Economic assessment of C-band reallocation

Tamaño: px
Comenzar la demostración a partir de la página:

Download "Economic assessment of C-band reallocation"

Transcripción

1 Economic assessment of C-band reallocation A REPORT PREPARED FOR THE GSMA October 2013 Frontier Economics Pty. Ltd., Australia

2

3 October 2013 Frontier Economics i Economic assessment of C-band reallocation 1 Introduction 3 2 Our results summary 5 3 Our approach High level description of methodology Estimating benefits of C-band re-allocation Estimating costs of C-band re-allocation Estimating indirect effects of C-band re-allocation Contents

4

5 October 2013 Frontier Economics 1 Economic assessment of C-band reallocation Figure 1. Benefits of C-band re-allocation in Australia and Indonesia 6 Figure 2. Cost of C-band re-allocation in Australia and Indonesia 6 Figure 3. Benefits of C-band re-allocation in the APAC Region 7 Figure 4. Costs of C-band re-allocation in the APAC Region 8 Figure 5. Consumer and producer surplus from C-band re-allocation 10 Figure 6. Social welfare analysis of C-band re-allocation 11 Figure 7. Benefits to Indonesia from re-allocating C-band spectrum 20 Figure 8. Benefits to Australia from re-allocating C-band spectrum 21 Figure 9. Benefits to the APAC region from re-allocating C-band spectrum 22 Figure 10. Rain fade issue in Indonesia 24 Figure 11. Sensitivity analysis - key cost drivers 31 Figure 12. Auction proceeds for the APAC region 34 Figure 13. Tax effects per cohort of countries 36 Figure 14. Employment effects for APAC countries (thousand jobs) 37 Table value per MHz Pop 17 Table 2. Adjusting for GDP and urbanisation levels 19 Table 3. Economic Benefits 19 Table 4. Costs associated with DTH TV and video distribution 28 Table 5. Estimated costs for VSAT users 29 Table 6. Costs associated with Earth stations 30 Table 7. Total costs of C-band migration for Indonesia and Australia 32 Table 8. Total costs of C-band migration for the APAC region 33 Table 9. % Tax revenues over GDP 35 Contents

6 2 Frontier Economics October 2013

7 October 2013 Frontier Economics 3 1 Introduction The increasing consumer demand for mobile data services implies that mobile operators will be required to provide significantly greater capacity to consumers in future. Forecasts predict that, notwithstanding significant increases in the efficiency of spectrum usage, this will require large increases in spectrum allocation to mobile operators. The C-band spectrum ( GHz) is a frequency range that is suitable for mobile operators, in particular for providing additional capacity to networks in urban areas. However, in the Asia Pacific (APAC) region C-band spectrum is currently mainly used by the satellite industry to deliver fixed satellite services, such as video distribution services for cable TV networks and for direct-to-home (DTH) television services and VSAT-based connectivity services for enterprises. The intensity of usage of the C-band varies by country in some countries, such as Indonesia, it is more intensively used whereas in other countries, such as Australia, it is less intensively used. Spectrum is a finite resource. From an economic perspective the objective of allocating spectrum is to maximise the efficiency with which it is used. This means maximising the difference between what users are willing to pay for it and the opportunity costs of supplying it, including the value of that spectrum to other users. Therefore, as the C-band spectrum is suitable for both satellite and mobile services a key question is, what is the most efficient way to allocate C- band spectrum? Determining precisely the most efficient allocation is difficult given the uncertainties surrounding some of the key estimates of the costs and benefits of re-allocation of spectrum. However, this report seeks to take the debate further by considering the costs and benefits associated with re-allocating 400MHz of C- band spectrum (the 3.4GHz-3.8GHz range) to mobile operators whilst the use of the upper part of the C-band spectrum remains with satellite operators. We analyse, in particular, the costs and benefits for two case study countries, Australia and Indonesia, and use these case studies to estimate the net benefits for the wider APAC region. There is limited information available about the current uses of the C-band spectrum, and the costs of shifting satellite usage to other frequency ranges. We have therefore had to make a number of assumptions to quantify the costs and the benefits, which are detailed below. Where possible we have relied on external sources for this information. The rest of this report is structured as follows: In section 2 we provide a summary of our main results for the case study countries and for the wider APAC region; and

8 4 Frontier Economics October 2013 In section 3 we explain in detail our approach and the key assumptions underlying our results. Introduction

9 2 Our results summary Below, we summarise the main results of our high level analysis of net benefits of C-band re-allocation and discuss our indicative estimates. Our modelling approach looks at the potential benefits to mobile operators of having access to 400MHz C-band spectrum by We then look at the cost such re-allocation would imply for satellite application that currently uses this frequency band. By estimating net benefits (benefits minus costs) of C-band reallocation, we approximate the gross-value added (GVA) created in the economy. Our approach relies on two key assumptions: C-band spectrum will be used in the future in a similar way that 2.6GHz is currently used by mobile operators, that is to build additional capacity networks in densely populated urban areas. We therefore assume that the access to 400MHz of C-band will help mobile operators to meet the future data demand in urban areas at lower costs than without this spectrum. The future demand for satellite services currently provided in C-band will be met after the re-allocation. This will be achieved through the remaining 400MHz of C-band and by shifting non-critical usage to other delivery means, such as other frequency bands assigned to satellite providers (Ku/Ka bands 2 ) or migration to alternative platforms (fixed and terrestrial networks). Our analysis focuses on estimating benefits and costs of C-band re-allocation in two case study countries Australia and Indonesia. Figure 1 and Figure 2 show estimated benefits and costs in these countries under three separate benefits scenarios and three separate cost scenarios. 1 In particular, we are considering lower part of C-band spectrum in the frequency range 3.4GHz to 3.8GHz. As our analysis is high level we do not explicitly consider the need to set aside spectrum for a protection band. 2 In this report, we use Ku-band when referring to frequencies between 11GHz and 18GHz and Kaband when referring to frequencies between 26.5GHz and 40GHz

10 6 Frontier Economics October 2013 Figure 1. Benefits of C-band re-allocation in Australia and Indonesia Source: Frontier Economics Figure 2. Cost of C-band re-allocation in Australia and Indonesia Source: Frontier Economics For Indonesia, the estimated benefits range from PPP $1.9 billion to PPP $17.5 billion, with estimated costs being between PPP $0.3 billion and PPP $1 billion. In our central base case scenario, the benefits of C-band re-allocation exceed costs by approximately 7 times. Our results summary

11 For Australia, the estimated benefits range from PPP $0.5 billion to PPP $4.7 billion, with estimated costs being between PPP $11 million and PPP $43 million. This implies that in our central base case scenario, the benefits of C-band reallocation exceed costs by approximately 47 times. Based on the results for these case study countries, and using some simplifying assumptions, we then derive indicative results for the whole APAC region, and show our indicative estimates of benefits and costs for the whole APAC region. The estimated regional benefits range from PPP $22 billion to approximately PPP $280 billion, with estimated costs being between PPP $2.6 billion and PPP $34.7 billion. This implies that in our central base case scenario, the benefits of C-band re-allocation exceed costs by approximately 8 times. Figure 3. Benefits of C-band re-allocation in the APAC Region Source: Frontier Economics

12 8 Frontier Economics October 2013 Figure 4. Costs of C-band re-allocation in the APAC Region Source: Frontier Economics It should be noted that benefits represent cumulative value of economic benefits derived from C-band spectrum by mobile operators, 3 the cost of re-allocation are mostly one-off in nature, as discussed in more detail below. Based on the estimates of net-benefits of C-band re-allocation, we also provide a high level estimate of the impact on government income and employment. We estimate that in the base case the total government income in the APAC region from C-band spectrum licencing and taxation of additional economic activity will be PPP $52 billion, with an addition of 108,000 new jobs. Our approach, as discussed below, relies on a number of assumptions due to limited availability of data, in particular on satellite usage and costs. Nevertheless, our approach could be considered conservative as we have not estimated additional benefits of higher quality of mobile services arising from the access to C-band spectrum. In addition, we have not explicitly modelled any indirect effects on the mobile supply chain and wider economy. 3 Operators benefit over time from having greater spectrum assigned to them (e.g. because they need less base stations) and they reflect this in the value they are willing to pay for that spectrum. Our results summary

13 3 Our approach This section describes in more detailed our approach and key assumptions underlying the results presented above. We first provide an overview of the methodology applied to estimate net benefits of re-allocating 400MHz of C-band spectrum. We then discuss in more detail how we estimated the direct benefits and costs of C-band re-allocation. We also present results for our case study countries and provide indicative estimates for the whole APAC region. Lastly, we discuss at the high level additional impacts of spectrum reallocation on the wider economy, in particular on tax revenues and employment in the APAC region. 3.1 High level description of methodology The mobile sector is an integral part of the economies in the APAC region and one of the significant drivers of economic growth. Demand for mobile services and mobile data in particular is rapidly increasing. To meet this demand, mobile operators are likely to need access to additional spectrum. 4 Our approach assumes that access to 400MHz C-band spectrum will help mobile operators to meet the future demand for mobile data services. We consider mobile operators will primarily use C-band to increase capacity of their mobile networks in urban areas where demand for mobile data is likely to be high. We therefore expect C-band spectrum to be used for similar purposes as 2.6GHz spectrum is currently used in the countries where this high frequency spectrum has been made available to mobile operators. By having access to additional C-band spectrum we assume mobile operators will be able to meet mobile data demand and serve future mobile users more efficiently, i.e. at lower costs than in the case without access to C-band spectrum. This is expected to lead to economic benefits (an increase in social welfare) through the following two channels: An increase in producer surplus implied by lower costs, i.e. mobile operators will need less resources to deliver the required services and meet 4 This is widely accepted and has been a consistent finding by many regulators. See, for example, Ofcom (2012): Securing long term benefits from scarce spectrum resources: A strategy for UHF bands IV and V, 29 March 2012.

14 10 Frontier Economics October 2013 the future demand for mobile data. It also implies that scarce capital and labour resources in the economy can be freed up and put to other more productive uses. If access to C-band spectrum lowers the marginal costs of mobile operators to some extent, in a competitive market, this would lead to lower prices and higher quantity or quality of services for end consumers, having further positive effects on social welfare. Figure 5 below illustrates a shift in supply curve, resulting from lower cost from C-band re-allocation, which leads to increases in both producer and consumer surplus and thus to an overall increase in social welfare. In addition, in this example the relationship between demand and supply would lead to lower prices for consumers and higher quantity of mobile data provided. 5 Figure 5. Consumer and producer surplus from C-band re-allocation Source: Frontier Economics At the same time, there are costs that current C-band users would incur as a result of spectrum re-allocation. In particular, certain fixed satellite services currently relying on C-band spectrum would need to be shifted to alternative means of delivery, e.g. other frequency bands allocated to satellite or fixed/terrestrial infrastructure. For the satellite applications that will continue to use C-band spectrum after re-allocation, there might be costs related to 5 Since the relative shift in costs and prices will be small, the primary welfare effect is an increase due to the lower costs for mobile operators, some of which will be passed on to consumers. Our approach

15 preventing possible interference from mobile networks operating in C-band. Also, there may be additional costs arising from a potential decrease in the quality of services and applications currently using C-band spectrum. Therefore, our aim is to estimate the incremental benefits of allocating 400MHz of C-band spectrum to mobile operators and compare that to the costs incurred by the satellite industry. If the benefits are greater than the costs, this would imply that there is an economic rationale for re-allocating a share of C-band spectrum as this would lead to an increase in social welfare. Figure 6 below, shows the situation where the marginal socio-economic benefit from allocating C-band spectrum to international mobile telecommunications (IMT) is greater than the marginal cost to the satellite industry. Figure 6. Social welfare analysis of C-band re-allocation Source: Frontier Economics Our analysis focuses on estimating benefits and costs of C-band re-allocation in two case study countries Australia and Indonesia. We chose Australia as an example of a country where it would be relatively easier to free up C-band spectrum for mobile usage and at relatively lower costs to the satellite industry compared to countries where the C-band spectrum is more heavily used by satellite providers. This is because the current satellite usage of C-band can be relatively easily shifted to alternative means of provision. 6 On the other hand, we 6 We understand that usage of C-band spectrum by the satellite industry in Australia is relatively limited, as local Australian satellite TV is delivered mostly via Ku-band (FTA and PayTV). We also understand that the satellite component of the National Broadband Network will be relying on

16 12 Frontier Economics October 2013 chose Indonesia as a country where there are a large number of satellite applications currently relying on C-band and where freeing up 400MHz of this spectrum is likely to be costly. Based on the results for Australia and Indonesia, and using some simplifying assumptions, we also provide indicative results for the whole APAC region. 3.2 Estimating benefits of C-band re-allocation Methodology and key assumptions In order to estimate the benefits of re-allocating C-band, we start by understanding what the supply and demand of mobile spectrum will look like in The two primary questions that arise out of this are: Would there be a need to allocate additional spectrum to the mobile sector? If yes, what would the benefits of the C-band spectrum re-allocation be to mobile operators? In order to answer the first question, we assess the pressure on the spectrum that is currently available to mobile operators to meet future demand: the more scarce this spectrum, the greater the need to re-allocate C-band. The pressure on spectrum that is already allocated to mobile will be determined by demand and supply. There is extensive evidence to suggest that there will be a shortfall in supply of spectrum to International Mobile Telecommunications (IMT) in the future. This shortage in supply arises primarily out of the increasing demand for video and data services on mobile devices (including smartphones, tables, laptops, etc.). For example, Cisco forecasts a 17-fold mobile data traffic growth in the APAC region between 2012 and The APAC region is expected to account for 47.1% of the global mobile traffic by The ITU Working Party 5D has estimated a total spectrum requirement of between MHz for IMT by This is significantly more than spectrum currently identified for IMT in Indonesia and Australia by 2020, which we understand will be approximately 900 MHz. Ku/Ka-band frequencies. There are some additional uses of C-band spectrum, for example fixed point-to-point and fixed point-to-multipoint services, however, we understand that these are also not intensive. 7 ITU (2013): Document /237-E, 17 July 2013 Our approach

17 In light of there being plans for only limited release of spectrum, it appears that the availability of new spectrum (such as C-band) will likely become necessary to meet the needs of mobile data users. 8 There are different ways of meeting the increase in demand, in addition to having access to additional spectrum. More efficient use of spectrum: advances in technology allow for a more efficient use of spectrum. For example, the Australian regulator ACMA suggests that the move to 3G and 4G technologies provides significant improvements in spectral efficiency and frequency re-use, especially compared with 2G technologies. 9 Similarly, Real Wireless estimated that technology developments associated to LTE deployment would provide between a three and ten times increase in mobile capacity for macro cells with three sector antennas by Increased network deployment: higher traffic demand could be accommodated by increasing the number of sites, i.e. by reducing the cell radius. However, as emphasized by UK regulator Ofcom, there are practical limits on how closely cells can be packed together without causing interference with one another and in securing planning permission for new sites. 11 In addition, increasing the number of sites in urban areas would likely imply significant deployment cost for operators. Off-loading indoor mobile traffic onto fixed networks: in the UK, Ofcom estimated that over half of mobile data traffic could potentially be offloaded onto fixed networks by 2030 using WiFi and Femtocells. 12 The feasibility of this alternative, though, depends on a number of factors, including the availability of WiFi access points, the availability and capacity 8 The UK government, for instance, decided in 2011 to free up additional 500MHz of spectrum to meet increasing mobile demand, including C-band frequencies 3.4GHz-3.6GHz, see https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/77460/spectru m-public-update-december-2011.pdf 9 ACMA (2011): Towards 2020 future spectrum requirements for mobile broadband, May Page OFCOM (2012): Securing long term benefits from scarce spectrum resources: a strategy for UHF bands IV and V, 29 march Paragraph Op. Cit., paragraph Op. Cit., paragraph 3.40.

18 14 Frontier Economics October 2013 of fixed networks 13 and the willingness of mobile users to off-load their mobile traffic. 14 However, most studies conducted by national regulatory authorities and international organisations have concluded that these alternative methods will not be sufficient to accommodate the very significant increases in future demand. 15 It is further a question to what extent these alternative measures can be applied in a cost-effective manner, in particular in the relatively less developed APAC countries with a limited deployment of fixed networks. Consequently, additional spectrum would be needed to cope with this. Given the likely value that mobile operators will place on C-band spectrum, and the assumption that this can be equated with the economic value that can be generated using that spectrum, we can address the question on the economic benefits of re-allocation. As it is envisioned that C-band spectrum will be used in a similar manner to the way the 2.6GHz is currently used by mobile operators, we use prices from the recent spectrum auctions of 2.6GHz FDD spectrum as a starting point for our estimates of economic value of C-band spectrum. 16 While we understand that C- band spectrum may primarily be used by technologies relying on unpaired spectrum (e.g. LTE-TDD), we assume that by 2025 these technologies will be widely applied and used by mobile operators in a similar way to how FDD services are currently used. Thus, the economic value of C-band spectrum was estimated as follows: The economic value of 2.6GHz spectrum was estimated based on benchmarking of auction prices. 13 See WIK Consult (2012): Study on Impact of traffic off-loading and related technological trends on the demand for wireless broadband spectrum, A study prepared for the European Commission DG Communications Networks, Content & Technology. 14 For instance, a recent survey of UK Everything Everywhere shows that 43% of their 4G subscribers uses fewer or no public Wi-Fi hotspots since having 4G, see https://explore.ee.co.uk/ourcompany/newsroom/4gee-transforming-britain-into-nation-of-nomadic-sharers-streamers-andshoppers 15 For instance, Study of Future Demand for Radio Spectrum in Canada explicitly takes into account alternative measures to meet spectrum demand, concluding that a substantial spectrum overhead would still be required to meet the expected mobile data demand 16 We understand that both 2.6GHz and C-band bands will and can be used to add capacity at particular traffic hotspots. Also both bands would be useful for small cells (indoor and outdoor) with high throughput. Our approach

19 These values were then adjusted to reflect the differences in the physical characteristics of C-band spectrum. Further adjustments were made to reflect country-specific factors, such as the degree of urbanisation and income levels. From this adjusted price, the economic value of C-band for Indonesia, Australia and the APAC region was derived. Further details on the methodology are provided below. Estimating value of high frequency spectrum based on auction benchmarking The future demand and supply of C-band spectrum is uncertain. However, as discussed above, it is expected that future demand for spectrum for IMT usage will exceed future supply. Rather than projecting forwards supply and demand, both of which are highly uncertain, our alternative proposition is that in 2025 (and beyond) mobile operators will use the C-band spectrum in the same way they are currently using, or planning to use, 2.6GHz spectrum. Effectively, we assume that the balance between future demand and supply of C-band can be approximated by the balance between the demand and supply for 2.6MHz. Thus, out starting point was auction benchmarking which gave a range of prices per MHz/population for the 2.6GHz spectrum. We have gathered publicly available data on auction trades to benchmark the value in different bands in the period 2009 to We have also concentrated on auction outcomes for paired spectrum as there currently is less demand for unpaired spectrum, in particular as there are a limited number of devices that are compatible with unpaired spectrum. Nevertheless, we assume that this will change in the future and that by 2025 the availability of network equipment and mobile devices will imply wide use of unpaired spectrum. In order to compare different spectrum auctions we have normalised each observation so that they are consistently expressed. In doing so we make a number of assumptions: we express all auctions in a price per MHz per population; we convert all prices in to US dollars on a purchasing power parity basis; 17 Earlier periods may be less relevant for the purpose of current benchmarking as they pre-date the widespread penetration of smartphones so mobile operators may have different valuations of the spectrum, and the period since 2009 coincides with the adoption of LTE as a mobile standard.

20 16 Frontier Economics October 2013 we express prices as 2013 prices by adjusting for CPI inflation since the auction; 18 we normalise all auctions to a common twenty year term using an assumed discount rate. 19 We recognise there are a number of external factors influencing the spectrum prices that we do not control for explicitly, which is why we work with a relatively wide range of estimates in our scenario analysis discussed below. In addition, as there is uncertainty surrounding what the exact equilibrium between demand and supply will be in the future, we used this range to develop 3 scenarios under which to analyse the potential benefits of re-allocating C-band. Base case: we use the simple mean of prices paid in 2.6GHz auctions as our starting point. We believe this approximates the economic value mobile operators believe to be able to extract from 2.6GHz spectrum today, taking into account expectations of future demand for mobile data and availability of spectrum. We note that the mean price is higher than the median price in the auction sample. However, we consider the mean as an appropriate measure in this case, as the value is skewed upwards by the auction results in countries in the APAC region (albeit by the countries which are currently the most developed), and thus may be more representative of the value of spectrum in the region. Low case: the median price was used as the input for the low case scenario. This case was used to model the scenario where demand for data is not as high as expected and the spectrum available is more likely to, at least partly, meet the mobile data demand in Thus, higher frequency spectrum would have a lower value for mobile operators. High case: the prices from recent auctions in Hong Kong and Korea were used to model the spectrum value in the high case. 20 Hong Kong placed the highest value on the 2.6GHz spectrum and prices in Korea were also well above the average. This case thus modelled the scenario where Hong Kong s and Korea s situation today are representative of the APAC region in Please note that throughout this report, all our results are presented in US dollar PPP prices for We estimate prices of a notional twenty year licence assuming that operators would be indifferent between choosing the twenty year licence term and the observed licence term given an assumed discount rate. 20 These were adjusted to take account of the high GDP per capita and urbanisation in these countries. Our approach

21 As shown in table below, these scenarios allow for a wide range of prices in our modelling, with a significant difference between low and high case scenario. This helps ensure that our results are robust Table value per MHz Pop Low Case Base Case High Case Source: Frontier Economics benchmarking analysis However, it is important to note that the prices from the auction do not give us the full estimate of economic value that mobile operators are able to generate as a result of using the spectrum. The auction prices are more likely to approximate the willingness to pay of the second-most efficient operator participating in the auction, which is directly related to the expected cost savings this operator can generate from having access to additional spectrum. Therefore, to estimate the full economic value of spectrum, we assumed the total economic value derived is approximately 50% above the price paid in the auction. 21 Adjusting for physical characteristics of C-band spectrum C-band is in a higher frequency band than the 2.6GHz spectrum and has different propagation characteristics, offering lower coverage (both outdoors and in building) at a given level of power. The higher the frequency of the spectrum, the more the number of cell sites needed to meet a given level of demand. It is estimated that C-band will need approximately twice as much cell sites than 2.6GHz, which means that the costs of network rollout will be higher for C-band than for the 2.6GHz spectrum. 22 We therefore need to take these differences in physical characteristics into account when estimating the value of C-band spectrum. For the purposes of our modelling, we rely on GSMA s estimate that the C-band will be valued at approximately 50% of 2.6GHz spectrum. 21 The winner s actual willingness to pay could be considerably more. For instance, bid data released by Ofcom on a recent auction revealed that the winner s willingness to pay was more than twice the amount actually paid, see 22 According to ITU (2011): Exploring the Value and Economic Valuation of Spectrum: This is because the service area covered by a base station is proportionate to the square of the frequency. For example, the minimum provision of service over a low population density region will require twice the number of base stations at 1 GHz than at 700 MHz, eight times more at 2 GHz and 14 times more at 2.6 GHz, and the cost of deploying a mobile network in such a region will rise in proportion. D/treg/Events/Seminars/GSR/GSR11/documents/BBreports_Economic-Valuation-of- Spectrum.pdf

22 18 Frontier Economics October 2013 Adjusting for urbanisation and GDP levels We looked at Australia and Indonesia as example countries. To derive the benefits for the APAC region, we divided it into two cohorts - High and Low. The High cohort consists of the more developed countries, such as Australia and Japan, with high GDP and urbanisation levels and the Low cohort consists of all the other countries in the APAC region. As C-band will primarily be used to provide capacity in densely populated urban areas C-band spectrum is likely to have higher value in countries with higher levels of urban population. However, our estimates of C-band value were derived from auctions for 2.6GHz spectrum that took place primarily in developed countries in Western Europe with high urbanisation and GDP. We adjusted for this by assuming a linear relationship between income/urbanisation and spectrum value, placing equal weight on both factors 23. We adjusted for the country/region by first calculating the country/region s urbanisation and GDP per capita as a percentage of the auction sample s average urbanisation and GDP per capita respectively. A simple average of these two proportions then gives us the adjustment factor. 23 Our methodology is broadly consistent to that applied by various spectrum benchmarking studies, which assumed a linear and cumulative relationship between income/population density (urbanisation) and the value of spectrum. See for instance DotEcon study for the Irish regulator available at Our approach

23 Table 2. Adjusting for GDP and urbanisation levels Country/Region GDP/capita as a percentage of auction sample average Urbanisation as a % of auction sample average Adjustment factor (simple average) Indonesia 23% 70% 47% Australia 158% 115% 137% APAC (high cohort) 140% 116% 128% APAC (low cohort) 31% 59% 45% Note: GDP per capita and Urbanisation levels in the APAC countries (projections for 2020) compared to the average of the countries the auction sample (as of 2012). Source: Frontier Economics Calculating benefits for Australia, Indonesia and the APAC region We multiplied the price per MHz/population of C-band spectrum after the above adjustments were made with 400MHz, which is the assumed amount of spectrum which will be made available, and the population of the country/region. This gave us the economic benefits of C-band spectrum for the country/region. Table 3. Economic Benefits Country/Region Low Case Base Case High Case Indonesia $1,854 $4,234 $17,520 Australia $471 $1,076 $4,453 APAC $21,789 $60,602 $279,726 Source: Frontier Economics Results Based on our calculations, we estimate that the benefits to Indonesia range from $1.9 billion to $17.5 billion, with the base case being $4.2 billion.

24 20 Frontier Economics October 2013 Figure 7. Benefits to Indonesia from re-allocating C-band spectrum Source: Frontier Economics Australia s lower population implies lower benefits from re-allocating C-band, but even the most conservative scenario, the low case, gives estimated benefits of PPP $500 million. The base case benefits are estimated at around PPP $1.1 billion. Our approach

25 Figure 8. Benefits to Australia from re-allocating C-band spectrum Source: Frontier Economics When estimating the benefits for the APAC region, the results range from $22 billion to around PPP $280 billion, with the base case benefits being estimated at PPP $67.6 billion.

26 22 Frontier Economics October 2013 Figure 9. Benefits to the APAC region from re-allocating C-band spectrum Source: Frontier Economics The above presented results may be conservative, as we have not explicitly considered additional benefits for consumers from higher quality of mobile services arising from the access to C-band spectrum. In addition, we have not explicitly modelled any indirect positive effects on the mobile supply chain and wider economy. 3.3 Estimating costs of C-band re-allocation Freeing up 400MHz of C-band spectrum will imply costs for the satellite industry. Below, we describe at a high level our methodology for estimating these costs and discuss in more detail our key assumptions. We then present our estimates of total costs incurred by satellite industry in Australia and Indonesia, also showing our results for the whole APAC region Methodology and key assumptions The starting point of our analysis is to identify the main applications or forms of usage of C-band spectrum by the satellite industry. 24 Based on the information 24 We note there are a number of non-satellite uses of C-band spectrum, such as fixed point-to-point and fixed point-to-multipoint services. We do not have information about the costs of moving these services to other delivery mechanisms. However, we understand that usage of these services is not intensive and therefore the costs are not expected to be significant. Our approach

27 from Northern Sky Research (NSR) 25, we identify four broad categories of applications relying on C-band: receive only consumer applications (e.g. satellite TV broadcasting); receive only commercial applications (e.g. cable head ends); two-way Very Small Aperture Terminals (VSAT) (e.g. banking terminals and ATM networks); and two-way large dishes (e.g. trunk telephony). There are a wide range of applications that fall into these four categories, including direct-to-home (DTH) broadcasting, satellite broadband internet services, military and government satellite use, corporate connectivity, as well as traditional telephony and carrier services. We then considered the cost of re-allocating 400MHz of C-band related to the individual applications. Our key assumption is that the ITU WRC decision will be passed in 2015 and that effective re-allocation of the lower part of C-band may not happen until We further assume that this will give the satellite industry enough time to gradually migrate non-critical applications, i.e. the applications that can be delivered cost-efficiently using alternative methods. We understand that it would not be possible to shift some of the applications currently relying on C-band to other frequencies (e.g. Ku/Ka band) or deliver them via alternative communication channels (e.g. over fixed networks). This is because some applications require high availability of service (e.g. VSAT terminals used for banking services etc.) and in certain countries they would still need to use the upper part of the C-band after 2025, since only C-band satellite services can easily provide the high availability required. This is particularly relevant for countries like Indonesia, where high rain attenuation is an important factor influencing of the availability of satellite services. 26 Broadly speaking, high rain attenuation limits the extent to which satellite services in higher frequencies (Ku/Ka band) can be used to deliver services that require high levels of availability. This is known as rain fade issue and its importance in Indonesia is illustrated in Figure 10 below. It shows that countries with high rain attenuation like Indonesia need a significantly higher 25 See Northern Sky Research (2013): Global Assessment of Satellite Supply & Demand, 10th Edition 26 Also the archipelago nature of Indonesia with 17,000 islands restricts the feasibility to deploy fibre infrastructure everywhere given the high cost of sub-marine fibre optic deployments to connect smaller population centres on islands.

28 24 Frontier Economics October 2013 fading margin 27 (expressed in db s) to achieve the same level of availability than countries with lower rain attenuation (e.g. UK). In addition, the figure shows that providing services with high availability in higher frequency bands (e.g. Ku/Ka bands) requires significantly higher fading margins, compared with providing the same services in C-band. In other words, in countries like Indonesia rain fade implies more outages for services provided in higher Ku/Ka bands in comparison to an equivalent service provided in C- band (unless other factors are varied to increase the resistance of the system to signal degradation, such as using bigger dishes). However the importance of this depends on the required availability of the service and the reaction of consumers to this. Figure 10. Rain fade issue in Indonesia Source: Newtec Webinar June 2013, Beyond Consumer and Ka-Band: The Future of Traditional VSAT (www.slideshare.net/newtec_satcom/the-future-of-traditional-vsat ) Therefore, for the purposes of our modelling exercise, we assume that for applications that require high availability of services, and can t be moved cost effectively to higher frequency bands, the 400MHz of C-band will be sufficient to fully meet the demand of these applications in the future. This is based on the assumptions that: 27 Fading margin is a measure of the amount by which a received signal level may be reduced without causing the signal availability to fall below the specified value.. Our approach

29 C-band spectrum will be used more efficiently in the future, through technological advancement (e.g. use MPEG-4 and potentially HEVC or DVB-S2); and applications that do not require C-band will be shifted to other bands or to fixed/terrestrial infrastructure, thus freeing up additional spectrum in the C-band. Given the limited availability of data on satellite usage, we focus on two main groups of applications that are likely to require significant amount of C-band spectrum in the future, in particular in the areas where rain fade may be a constraint to C-band re-allocation. direct-to-home satellite TV broadcasting (DTH) and video distribution to cable head ends; and different forms of two-way satellite connectivity with high service availability requirements. For example it may be that DTH will not require the same availability as some two way satellite services. As can be seen from Figure 10 a small reduction in required availability such as from 99.9% to 99.7% may have an impact on the link budget. For these applications, we explicitly modelled the costs of partial migration (where possible) to other frequency bands and additional costs related to dealing with potential interference from mobile services operating in the lower part of C- band. In addition, we took into account the expected costs of making some satellite equipment redundant as a result of freeing up the lower part of C-band in We also considered additional costs related to moving earth stations using C- band out of urban areas to areas where there is likely to be lower interference. This allows us to derive indicative estimates of the net benefits of re-allocating 400MHz to mobile use. We also applied a number of sensitivity checks on individual cost drivers to make sure that we have provided a sufficiently wide range of cost estimates. We recognise that this is not an exhaustive list of applications that would incur costs as a result of C-band re-allocation. Nevertheless, based on the information available to us, we believe that these applications are likely to represent major cost of C-band re-allocation. For instance, NSR estimates that video

30 26 Frontier Economics October 2013 distribution 28 and corporate connectivity services will account for almost 80% of C-band satellite capacity in South East Asia by We also note that according to NSR s data the overall demand for C-band satellite capacity in the region will decline gradually, which is in contrast to forecasted growth of usage in both Ku and Ka band. This is consistent with our assumption that a share of applications currently using C-band will gradually migrate to more cost-effective higher frequency bands. Below, we discuss in more detail our approach to estimating the costs of reallocation for the key applications described above. Cost of DTH TV and video distribution We assumed a proportion of DTH TV households would continue to use C- band, in particular in countries where rain fade is a significant issue. However, since our scenario considers mobile operators using the lower part of the C-band spectrum in urban areas, this would imply that filters would need to be installed to the satellite dishes of these customers. 29 We modelled this explicitly for Indonesia, assuming that there will be approximately 10 million DTH TV households close to urban areas relying on C- band satellite signal in We assumed that a proportion of these households (20% in our base case) would have chosen to switch to DTH TV services provided over Ku band by These would be users willing to accept somewhat lower availability of TV service due to rain fade, because they would have access to a potentially wider range of services since the capacity for SD and HD channels is much larger in Ku band. The remaining 8 million of DTH TV household that are assumed to continue to use C-band will require filters to prevent signal interference from mobile networks operating in lower parts of the C-band. We assumed the cost of LNB filters to be $10 in our base case, and that 20% of households would be able to self-install these filters. For the remaining 80%, a home visit by an engineer would be required, with an assumed cost of $25 per visit. This implies a one-off cost of C-band DTH filtering of approximately PPP $405 m According to NSR methodology, video distribution includes also DTH free-view broadcasting services 29 This is because the DTH satellite dishes have been designed to pick up signals across the C-band and would need to be adjusted so that they did not pick up the high power mobile signals in this range. 30 For the purposes of these cost calculations, we work with PPP exchange rate of 1.69 for Indonesia. This implies that one $ in nominal terms is converted into PPP $1.69. For Australia, we use PPP exchange rate of For simplicity we apply PPP adjustment to all cost items. Our approach

31 In addition, we assumed that cable head ends operating in urban areas would need to be equipped by filters at $200 per piece. Assuming a total number of 500 head ends in Indonesia and 30 pieces per head end, this implies a one-off cost of head end filtering of approximately PPP $5m. In total, we estimate the cost of DTH TV and video distribution for Indonesia to be approximately PPP $410m in our base case. We note that due to the decreased spectrum available for C-band video distribution by satellite in urban areas, the DTH proposition would be somewhat restricted. We do not have sufficient information to model the impact on consumer welfare of a somewhat lower range and quality of channels available for these users. However, we note that subscribers that place a high value on the wider range of channels would be those that would be more likely to switch to the new Ku-band satellite services, in any case. Therefore, this limits the welfare impact on the C-band subscribers. With regards to Australia, we understand that DTH TV in C-band is significantly less important than in Indonesia and is effectively limited to households capturing C-band satellite signal from foreign satellites (Indonesian, Indian, Chinese, Korean, Vietnamese, Thai, Arabic, etc and some International TV channels), that are aimed at cable head ends, for the purpose of further distribution or aimed at foreign DTH consumers. The main Australian Free-to- Air and Pay TV packages are offered on Ku-band satellites. 31 Moreover, as rain fade is unlikely to be a significant issue in Australia, we assume cable head ends currently relying on C-band would be able to switch to other frequencies in the 10 year period between 2015 and 2025 at no additional (significant) cost. Therefore, for the purposes of our modelling, we assume that total costs of accommodating DTH TV usage in C-band in Australia are likely to be insignificant The key Australian TV satellites such as Optus C1, Optus D1, Optus D2 and Optus D3 carry mainly Ku-band transponders, see https://www.optus.com.au/network/satellite/fleet 32 With regards to the cost of video distribution in Australia, we cover this in the section below where we discuss moving large earth stations outside urban areas.

32 28 Frontier Economics October 2013 Table 4. Costs associated with DTH TV and video distribution Scenario Potential costs for Indonesia (PPP $ mn) Low Case 198 Base Case 410 High Case 712 Source: Frontier Economics Cost of VSAT usage For the purposes of our modelling, we assumed that it would be feasible to shift 50% of corporate VSAT usage to alternative frequency uses (e.g. reconfiguring VSAT terminals on Ka/Ku frequencies or using fixed lines). We further assumed that that this could be done gradually, over a 10 year period, at no effective additional cost, as this migration would take place anyway. This is because we assume that for applications not requiring high service availability it would be cost efficient to use Ka/Ku band, in line with the forecasted increases in Ku/Ka band usage, according to NSR data. In particular broadband internet services require a lot of capacity and therefore a lower price per MByte. Ka-band offers more bandwidth per transponder resulting in a lower bandwidth cost, though at the expense of a lower availability due to rain fading. Some solutions even propose dual-band Ku/Ka to benefit from both more cost-effective bandwidth at Ka-band and better availability at Ku-band. We understand that broadband applications will, for bandwidth cost reasons, try to make more use of the higher satellite frequency band, as long as the reduced availability is acceptable for the customer (a typical example would be consumer satellite broadband). The remaining VSAT terminals would then need to be equipped with appropriate filters, at a cost of $200 per terminal. We assumed that there will be approximately 30,000 affected terminals in Indonesia and 5,000 in Australia. This would imply one-off costs of approximately PPP $10m and PPP $0.7 million respectively. Our approach

33 Table 5. Estimated costs for VSAT users Scenario Indonesia (PPP $ mn) Australia (PPP $ mn) Low Case Base Case High Case Source: Frontier Economics Cost of redundant satellite equipment By 2025, there will still be a number of old satellites (released before 2015) carrying C-band transponders configured for GHz range. As these transponders will not be well utilised after 2025, we consider their remaining asset lifetime to estimate the total cost of satellite equipment stranded as a result of C-band re-allocation. We assumed that there are currently around 50 satellites carrying C-band transmitters that are targeting the APAC region. We also assumed that C-band capacity of these satellites is 50% on average. We further assumed an average satellite lifetime of 15 years. Assuming that the satellites had been launched at the same rate over time, this implies that by 2025, there will be approximately 17 of these satellites still in operation, with a remaining asset lifetime of 2.5 years that would be stranded. 33 Assuming the cost of satellite capacity of approximately $250m per satellite, we estimate the total cost of C-band satellite equipment for the APAC region to be around $347m. We then approximately allocated this cost to Indonesia and Australia on the basis of population shares in the APAC region. 34 This implied a cost of satellite equipment in our base case of approximately PPP $48m and PPP $2m respectively. 33 We recognise that this is a very simplified approach relying on a number of assumptions, but we believe it should provide a reasonable estimate of the order of magnitude of the costs related to satellite assets. At the same time, this approach is conservative as these costs could be considered sunk and as such do not necessarily need to enter our calculation of benefits and costs of spectrum re-allocation. 34 We don t include China s population in this estimate, due to negligible share of DTH TV usage in C-band in China. This implies the adjusted population share of Indonesia is around 8% and about 1% for Australia.

34 30 Frontier Economics October 2013 Cost of earth stations We further assumed there would be a number of earth stations, many of which would be TV receive dishes, operating in urban areas that would be affected by potential mobile interference. For that reason, we assumed these earth stations would need to be moved to more suitable areas (e.g. valleys in remote locations). In our base case, we assumed the cost of moving a single teleport to be approximately $0.8m in Indonesia and $2m in Australia. We further assumed that there will be significant cost savings arising from using shared facilities. In our base case, we assumed it would be possible to achieve 75% saving in Indonesia as a result of facility sharing and 50% saving in Australia. Assuming total number of C-band earth stations in Indonesia to be 500 and in Australia 40, and assuming 75% of these stations would need to be relocated, this implied total cost of moving earth stations of approximately PPP $127m for Indonesia and PPP $22m for Australia. 35 Table 6. Costs associated with Earth stations Scenario Indonesia ($ PPP mn) Australia ($ PPP mn) Low Case 53 9 Base Case High Case Source: Frontier Economics Sensitivity analysis We recognise that our estimates rely on a number of assumptions, due to limited availability of data on the actual satellite usage and costs in C-band. We therefore identify the key cost drivers underlying the estimate for each of the key applications discussed above. 35 We understand that there are only about 12 C-band earth stations that would need to be relocated in Australia, see Appendix 5 here We also understand that some head ends will disappear as it becomes more cost-effective to distribute the TV signals over the fibre optic backbone (NBN) instead of building/maintaining cable head-ends all over Australia. Therefore, our assumption that 30 earth stations in Australia will be relocated appears conservative. Our approach

35 For each of these, we have assumed a range of values to model low, base and high cost scenarios. Our general approach is to apply approximately 50% increase (decrease) to the base value of the cost driver, to get the value for high and low scenario, respectively. The range of value for each of the key cost drivers in our calculation is illustrated in Figure 11 below. Figure 11. Sensitivity analysis - key cost drivers Low Base High Growth in C-band DTH user base (%) 20% 25% 30% Proportion of DTH HH in urban areas 60% 67% 75% Proportion C-band users that would switch (%) 30% 20% 15% Cost of filter (head end/vsat) ($) Costs of filter (mass production) ($) Share of self-instalations (%) 30% 20% 10% Home visit cost ($) Cost of satellite capacity ($m) Assume share of C-band capacity (%) 30% 50% 70% Share of affected earth stations (%) 50% 75% 100% Cost of moving ($m) (Indonesia) Cost of moving ($m) (Australia) Number of VSATs (Indonesia) 20,000 30,000 40,000 Number of VSATs (Australia) 3,000 5,000 7,000 Source: Frontier Economics Results Table 7 below summarises our cost estimates for Australia and Indonesia under different scenarios. The cost estimates for Indonesia range from PPP $0.3 billion to PPP $1 billion. The cost estimates for Australia range from PPP $11 million to PPP $43 million.

36 32 Frontier Economics October 2013 Table 7. Total costs of C-band migration for Indonesia and Australia Scenario Indonesia ($ PPP mn) Australia ($ PPP mn) Low Case Base Case High Case 1, Source: Frontier Economics We then extrapolated our cost estimates to the whole APAC region. We calculated the share of cost on benefits for Indonesia and Australia and apply this share to similar countries in the APAC region. 36 We use the relative share of costs on benefits for each country based on the Base Case scenario. The results are shown in Table 8 below. 36 The High Cohort consists of Australia, Japan, Hong Kong, Korea, New Zealand and Singapore. The Low Cohort consists of Brunei Darussalam, China, Indonesia, Vietnam, Philippines, Thailand, Malaysia, Sri Lanka, Fiji, Samoa, Vanuatu, Tonga, India, Bangladesh, Pakistan, Nepal, Cambodia, Lao, Myanmar, Papua New Guinea, Bhutan, and the Solomon Islands. Our approach

37 Table 8. Total costs of C-band migration for the APAC region Scenario Low cohort High cohort APAC total Cost to benefits ratio 14% 2% To be applied to the following benefits by cohort to get costs Low Case 2, ,574 Base Case 8, ,387 High Case 33, ,703 Source: Frontier Economics 3.4 Estimating indirect effects of C-band re-allocation Besides the direct net benefits from the re-allocation of C-band spectrum to the mobile sector estimated above, the additional value added generated will also have a number of indirect effects over the economy. These include: government income, new job creation, an impact on the wider economy and the promotion of new business creation. While our study estimates the impact on government income and employment, the impact on the wider economy is addressed only from a qualitative point of view. Hence, our estimates provide a lower bound on the potential benefits that the re-allocation of C-band spectrum to the mobile sector can bring to the economy Impact on government income and employment The re-allocation of additional C-band spectrum to the mobile industry will allow mobile operators to be more productive. This is because an additional amount of spectrum will allow operators to serve a higher level of demand with the given resources. Alternatively, an additional amount of spectrum will allow mobile operators to meet the increasing demand for mobile data services more efficiently. This higher efficiency implies freeing up of scarce resources from the mobile sector to other sectors, which will lead to increasing economic activity and additional gross value added (GVA) in the economy. As a high level approximation, we assume GVA is broadly comparable with our estimates of net benefits for mobile industry generated by re-allocation of C-band spectrum.

38 34 Frontier Economics October 2013 The extra GVA generated via the re-allocation of C-band spectrum will lead to additional government revenues and job creation, as discussed in more detail below. Government income There are two main sources of government income arising from the C-band reallocation. First, there is direct income arising from selling spectrum licences proceedings. Second there is additional tax income from increased economic activity. We estimate the overall government proceeding from spectrum licensing by simply multiplying the assumed price of C-band spectrum (per MHz/population) by 400MHz available and the total population in the APAC region. The results are summarised in Figure 12 below. Figure 12. Auction proceeds for the APAC region Source: Frontier Economics The additional taxation revenues will come via direct taxes (such as corporate taxes or income taxes) and indirect taxes (such as value added taxes VAT). In this study we have taken a simple approach to estimating the impact on tax revenues. Hence, our results are only indicative. Our approach

39 For the sample of countries considered in the study, we have estimated the proportion of tax revenues over GDP and estimated the weighted average tax rate, as shown in Table 9 below. Table 9. % Tax revenues over GDP Country % Tax revenue GDP (US$ billion) Australia 21% 1,541.8 Hong Kong SAR, China 14% Japan 10% 5,964.0 Korea, Rep. 16% 1,155.9 Singapore 14% New Zealand 28% Weighted average (High Cohort) 13% China 10% 8,227.0 Indonesia 12% Malaysia 15% India 10% 1,824.8 Pakistan 9% Thailand 18% Vietnam 30% Weighted average (Low Cohort) 11% Source: Frontier Economics based on information from the World Bank We note that the weighted average tax rates look rather low when compared with other regions. 37 Potential reasons identified to explain the low tax-to-gdp ratio 37 The OECD average 33.8% in 2010.

40 Tax revenues PPP $ millions 36 Frontier Economics October 2013 observed in India include 38 (i) a low per capita income; (ii) a large proportion of small and medium enterprises which benefit from a variety of exemptions and, in some cases, may not comply with its tax duties; and, (iii) the lack of policy initiatives to increase tax revenues. These reasons are likely to be applicable to the other countries in the region that present a low tax-to-gdp ratio. In order to estimate the additional tax revenues generated by the re-allocation of C-band we have applied the weighted average tax rate over the additional GVA generated in each cohort. The results are shown in Figure 13 below. Figure 13. Tax effects per cohort of countries $25,000 $22,274 $20,000 $15,000 $10,000 $5,000 $0 $5,383 $4,862 $1,635 $514 $1,175 Low Case Base Case High Case Low Cohort High Cohort Source: Frontier Economics Employment Similarly, we have simply estimated the impact on employment based on the average employment in each country compared to the total GVA. If we assume that this relationship stays constant, which may be a reasonable first order approximation, this implies an increase in employment which is proportional to the increase in the GVA generated by the re-allocation of C-band spectrum. The above assumption is reasonable in this context. This is because the additional GVA generated by the re-allocation of C-band spectrum will come via the additional activity generated in the economy by freeing up resources that would be used to provide mobile services. 38 See Our approach

41 Jobs added (thousands) Figure 14 below shows the effect on employment associated with the reallocation of C-band spectrum per country-cohort. Figure 14. Employment effects for APAC countries (thousand jobs) Low case Base case High case Source: Frontier Economics Impact on wider economy The re-allocation of C-band spectrum to mobile services is likely to exert two additional effects. First, the additional economic activity generated by the freeing up of resources from the mobile sector to other sectors in the economy (via the higher productivity attained from the re-allocation of C-band spectrum) will exert a multiplier effect over the whole economy. Second, the re-allocation of C-band spectrum to the mobile industry will likely result in a higher quality of service for mobile broadband services. By facilitating information flows, a higher quality for mobile broadband services may enhance the creation of new businesses. While we recognize the potential importance of these effects, we do not provide a quantitative estimation in our analysis. Hence, our estimation of the benefits from re-allocating additional C-band spectrum in the mobile sector should be considered a conservative estimate. Multiplier effect The multiplier effect created by an increase in the economic activity, in this case via freeing up resources by the mobile sector, refers to the subsequent spending rounds induced by the extra income generated. In other words, it captures the

42 38 Frontier Economics October 2013 fact that a proportion of the wages paid to the direct and indirect employees in the new economic activity, will be spent on domestically produced goods, thus stimulating further economic activity within other sectors. The multiplier effect is usually considered when assessing the impact assessment of a policy intervention or when estimating the economic contribution of a sector on the economy. In the context of mobile communications, existing studies estimating the contribution of the sector on the economy, have considered a multiplier effect in the range A recent study by AT Kearney 40 considers a multiplier of 1.5 for the Asia-Pacific region. By not accounting for a multiplier effect, our estimates represent a lower threshold of the potential benefits that the re-allocation of the C-spectrum band will generate in the APAC region. New business creation The potential increase in the quality of the mobile broadband service, due to the re-allocation of C-band spectrum for mobile services, is expected to enhance economic activity for different reasons. These include, among others, the following: An increased usage of mobile data services may also lead to lower input costs, for instance by promoting online shopping or by facilitating outsourcing to areas with lower labour costs. Higher quality mobile data services may facilitate a more efficient use of time and increase productivity of workers, by promoting telecommuting and while working while travelling. The literature on the impact of mobile services on new business creation is still new and scarce. With exceptions, most studies focus on developed countries and estimate the link between the availability and/or penetration of Internet services on new business creation. While the studies differ in the approach undertaken, they coincide on the existence of strong positive effects of internet on new businesses and 39 Deloitte (2008): Economic Impact of Mobile Communications in Serbia, Ukraine, Malaysia, Thailand, Bangladesh and Pakistan, A report prepared for Telenor ASA, January AT Kearney (2011): Asia Pacific Mobile Observatory 2011: Driving Economic and Social Development through Mobile Broadband, A report for the GSMA. Our approach

43 entrepreneurship. For example, Garcia-Murillo et al. (2013) 41 have found, for the case of México, that a 10% increase in fixed broadband penetration is associated to a 0.09% increase in new businesses density (new businesses per 100 people). Similarly, Kim and Orazem (2012), 42 estimate that a 10% increase in broadband availability raises firm entry by 1.6% but only in urban areas. In rural counties not adjacent to a metropolitan area, the implied increase in the probability of firm entry is only 0.2%. While this literature is not directly applicable to the current case, in areas where the availability of fixed broadband services is low, the impact of mobile broadband services may be of a similar magnitude. Further, the finding by Kim and Orazem (2012) that broadband availability may have a stronger impact on urban areas is particularly relevant for the current case. This is because the availability of C-band spectrum will mainly benefit urban areas. We therefore conclude that access to C-band spectrum might have positive effect on new business creation, in particular in countries where the availability of fixed broadband is low 41 Garcia-Murillo, M., J.A. Velez-Ospina and P. Vargas-Leon (2013): The techno-institutional leap and the formation of new firms, working paper version. 42 Kim, Y., Orazem, P. (2012): Broadband Internet and Firm Entry: Evidence from Rural Iowa, Iowa State University working paper.

44 40 Frontier Economics October 2013

45 Frontier Economics Pty Ltd in Australia is a member of the Frontier Economics network, which consists of separate companies based in Australia (Melbourne & Sydney) and Europe (Brussels, Cologne, Dublin, London & Madrid). The companies are independently owned, and legal commitments entered into by any one company do not impose any obligations on other companies in the network. All views expressed in this document are the views of Frontier Economics Pty Ltd. Disclaimer None of Frontier Economics Pty Ltd (including the directors and employees) make any representation or warranty as to the accuracy or completeness of this report. Nor shall they have any liability (whether arising from negligence or otherwise) for any representations (express or implied) or information contained in, or for any omissions from, the report or any written or oral communications transmitted in the course of the project.

46 FRONTIER ECONOMICS MELBOURNE SYDNEY Frontier Economics Pty Ltd 395 Collins Street Melbourne Victoria 3000 Tel: +61 (0) Fax: +61 (0) ACN: ABN:

47 Beneficios Económicos del Dividendo Digital para América Latina Por Dr. Raúl L. Katz Dr. Ernesto Flores-Roux El siguiente estudio fue comisionado por un consorcio conformado por la GSMA (GSM Association), AHCIET (Asociación Iberoamericana de Centros de Investigación y Empresas de Telecomunicaciones), Telefónica, América Móvil, Telecom Italia, Qualcomm, e Intel. TELECOM ADVISORY SERVICES, LLC

48 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA ÍNDICE PREFACIO RESUMEN EJECUTIVO EXECUTIVE SUMMARY (English) RESUMO EXECUTIVO (Português) 1. INTRODUCCIÓN 2. INVESTIGACIONES REALIZADAS HASTA LA FECHA SOBRE EL IMPACTO ECONÓMICO DEL DIVIDENDO DIGITAL 2.1. Estimación del valor económico privado de escenarios de cambio de atribución del espectro 2.2. Estimación del impacto socio-económico de escenarios alternativos de atribución de espectro 2.3. Valoración económica de atribución de espectro a las telecomunicaciones móviles 2.4. Simulación del valor social de atribución de espectro a las telecomunicaciones móviles 2.5. Conclusión 3. METODOLOGÍA PARA LA ESTIMACIÓN DEL IMPACTO ECONÓMICO Y SOCIAL DEL DIVIDENDO DIGITAL EN AMÉRICA LATINA 3.1. Análisis de la situación actual 3.2. Estimación de la demanda de servicios de banda ancha móvil 3.3. Estimación del valor del Dividendo Digital desde la óptica del operador 3.4. Comparación del impacto económico y social de escenarios de atribución de espectro 3.5. Extrapolación del estudio detallado de cinco países al resto de América Latina 4. SITUACIÓN ACTUAL DE LA ASIGNACIÓN DE LA BANDA DE 700 MHz EN AMÉRICA LATINA 4.1. La situación actual de espectro de las telecomunicaciones móviles en América Latina 4.2. La situación actual del espectro de 700 MHz en América Latina 5. EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 5.1. Crecimiento del tráfico de datos móviles a nivel mundial 5.2. Tendencias de la adopción de terminales generadores de tráfico de datos en América Latina 5.3. Estimación de tráfico de datos promedio por terminal 5.4. Impacto en el tráfico de voz y datos 6. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MOVIL 6.1. Cerramiento de la brecha de cobertura en zonas aisladas y rurales 6.2. Aumento de la penetración de banda ancha móvil como resultado de una reducción de precios

49 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA 6.3. Impacto en el ecosistema 6.4. Contribución directa a la economía 6.5. Contribución indirecta a la economía 6.6. Generación de excedente de consumidor 6.7. Beneficios sociales de la Banda Ancha móvil 7. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 7.1. Situación actual de la industria de la televisión en América Latina 7.2. Contribución al ecosistema 7.3. Contribución directa a la economía 7.4. Contribución indirecta a la economía 7.5. Contribución impositiva 7.6. Impacto económico acumulado 7.7. Impacto social y beneficios públicos 8.CONCLUSIONES APÉNDICES A. SOBRE LOS AUTORES B. BIBLIOGRAFÍA

50 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA PREFACIO El Dividendo Digital es sin lugar a dudas un elemento central en el futuro desarrollo y alcance de la Sociedad de la Información de todo el mundo. Nos encontramos hoy frente a un recurso crucial para democratizar el acceso a Internet de banda ancha. Si bien este es un tema que concierne a todos los gobiernos a través de sus políticas regulatorias de telecomunicaciones y radiodifusión, es la sociedad en su conjunto la que debe participar del debate acerca de cómo deben satisfacerse sus necesidades y de los mejores usos que pueden darse de los recursos de propiedad de una nación, tales como en este caso, del espectro radioeléctrico. El espectro que conforma el Dividendo Digital en América Latina es resultado del progreso tecnológico de las Tecnologías de la Información y la Comunicación (TICs) y cuenta con ventajas de propagación que permitirían expandir la frontera de alcance de los servicios de banda ancha existentes y conectar a segmentos de población aún no conectados reduciendo la llamada brecha digital. Sin embargo, la utilización de este espectro para la masificación de la banda ancha requiere un cambio de paradigma, básicamente una ruptura del statu quo en relación al uso que se le ha dado hasta el momento a esta porción del espectro. Por ser este un cambio crítico y determinante del futuro desarrollo económico y social de los países, creemos que es necesario contar con las mayores herramientas posibles a nuestro alcance que permitan tomar las decisiones de política pública más adecuadas y que maximicen los posibles beneficios para la sociedad. Es con esta motivación que hemos comisionado el presente estudio a Telecom Advisory Services LLC, a fin de que puedan profundizar en el análisis de los argumentos empíricos que nos ayuden a entender el impacto, tanto cualitativo como cuantitativo, que tendrían los usos alternativos del espectro del Dividendo Digital en algunos países clave de América Latina. Si bien ya se han realizado diversos estudios de este tipo, este es el primero que se lleva a cabo para los países de nuestra región, con una metodología que se adapta al contexto en el que estamos inmersos y hace previsiones en base a toda la información disponible al momento. Este estudio ha sido para nosotros un desafío muy importante que nos hemos animado a enfrentar a fin de hacer una contribución al debate que es necesario tener sobre el futuro que se le va a ofrecer a los ciudadanos de nuestra América Latina. Creemos que lo peor que puede suceder respecto de este debate y esta oportunidad es no tomar decisiones y retrasar o sub-utilizar la explotación de un recurso productivo que puede ser promotor de desarrollo individual y colectivo. 22 de Septiembre de 2011 Sebastian M. Cabello Director, GSMA Latin America Pablo Bello Secretario General, AHCIET RESUMEN EJECUTIVO 01

51 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA RESUMEN EJECUTIVO 1 Este estudio busca proporcionar a los tomadores de decisión de los países de América Latina una valoración cualitativa y cuantitativa del beneficio económico y social que puede ser generado si el "Dividendo Digital" es asignado a telecomunicaciones móviles, para su utilización primordial en el servicio de banda ancha móvil. El Dividendo Digital se define como el segmento superior de la banda de UHF 700 MHz en el caso de América Latina actualmente atribuido al servicio de radiodifusión en la mayoría de los países, y que, como consecuencia de la transición de la televisión analógica a digital, se libera, pudiendo así ser utilizado para prestar servicios de banda ancha. Esto permite dar una mayor capacidad a los servicios móviles para responder al crecimiento de tráfico de datos y aumentar la cobertura de estos servicios. El estudio está basado en un análisis en detalle de cinco países de la región (Argentina, Brasil, Colombia, México y Perú), a partir del cual se extrapolan 2 los resultados para el resto del continente. El Crecimiento del Tráfico de Datos en América Latina El tráfico de datos en América Latina está creciendo de manera importante, en buena medida por el advenimiento de la banda ancha móvil; esto está generando una necesidad cada vez más importante de espectro. Las telecomunicaciones móviles han alcanzado niveles masivos de penetración en el continente latinoamericano. El promedio continental, de 97,8% al primer trimetre de 2011, representa un nivel de adopción relativamente similar al observado en países industrializados. Basándonos en la situación actual, la tendencia histórica, y una estimación conservadora de los niveles de saturación esperados, estimamos que la penetración de la telefonía móvil alcanzará un promedio regional superior a 117% en el 2015, llegando a 3 130% en el De manera simultánea con la adopción acelerada de telecomunicaciones móviles, los operadores de América Latina están migrando sus redes de tecnologías de segunda (2G) a tercera generación (3G), mientras que algunos ya están haciendo 4 pruebas de cuarta generación (4G). Hacia el 2012 se observarán también comienzos de migración hacia plataformas 4G, con base principalmente en el estándar LTE (Long Term Evolution). La migración a tecnologías 3G es importante en la 5 medida en que los terminales que operan en estas normas, por ejemplo HSPA (High Speed Packet Access), son más adecuados económicamente para proveer un acceso de banda ancha eficiente a Internet que las tecnologías de banda ancha fija. Esta tecnología representa una respuesta económica y tecnológica a las necesidades de un mercado condicionado por los aún relativamente altos costos de adquisición de computadoras y los límites en el despliegue de banda ancha fija. La satisfacción de una necesidad de la población por tener acceso a Internet, combinada con una utilización más eficiente del espectro radioeléctrico, determina que la transición de la masa de abonados en la región a plataformas 3G se completará en el curso de la presente década. Así, nuestras proyecciones de la tasa de sustitución de terminales muestran que hacia el 2015, 46,2% de los abonados de la región estarán utilizando terminales 3G y 4G. Ciertos países de la región, debido al aumento dramático de abonados con tecnología HSPA y el ulterior despliegue de LTE, registrarán una mayoría de la base instalada en terminales de tercera y cuarta generación. Por ejemplo, de acuerdo a nuestras proyecciones, estimamos que hacia el 2020, 87% de la base 1. Este estudio fue comisionado por el consorcio conformado por la Asociación GSM (GSMA, por sus siglas en inglés), AHCIET (Asociación Iberoamericana de Centros de Investigación y Empresas de Telecomunicaciones), Telefónica, América Móvil, Telecom Italia, Qualcomm e Intel. 2. Los países considerados para la extrapolación de resultados incluyen Bolivia, Chile, Costa Rica, Ecuador, El Salvador, Guatemala, Honduras, Nicaragua, Panamá, Paraguay, República Dominicana, Uruguay y Venezuela. 3. Conservadoramente estimamos que en el 2015, la penetración en Argentina alcanzará el 150%, en Brasil 141%, en Colombia 100%, en México 95%, y en Perú 105%. 4. Comunicado de Prensa de la UIT, 6 de diciembre de Following a detailed evaluation against stringent technical and operational criteria, ITU has determined that LTE-Advanced and Wireless MAN-Advanced should be accorded the official designation of IMT-Advanced. As the most advanced technologies currently defined for global wireless mobile broadband communications, IMT-Advanced is considered as 4G, although it is recognized that this term, while undefined, may also be applied to the forerunners of these technologies, LTE and WiMax, and to other evolved 3G technologies providing a substantial level of improvement in performance and capabilities with respect to the initial third generation systems now deployed 5. En la segunda mitad del 2011, UNE (Empresa Publica de Medellín) en Colombia estará desplegando servicio en LTE en tres ciudades. En el 2012, ENTEL Chile lanzará servicio, y se espera que esto sea seguido de acciones por parte de Movistar (Telefónica) en Chile y Argentina, y Claro (América Móvil) en Chile. RESUMEN EJECUTIVO 02

52 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA instalada en Argentina, 73% en México y 76% en Brasil serán terminales de 3G y 4G. La migración hacia terminales 3G incluye una subtendencia importante: la adopción de smartphones. La funcionalidad de estos terminales es más avanzada que la de los teléfonos móviles básicos (llamados feature phones) en la medida que provee interfaces y formatos de pantalla más adecuados para acceder a internet. La adopción de smartphones representa una tendencia fundamental a ser estudiada en la medida que la conveniencia de estos terminales para el acceso a internet determina que los usuarios de smartphones tienden a utilizar la línea móvil de manera más intensa. Si bien en la actualidad la base instalada de este tipo de terminal en Argentina, Brasil, Colombia, México y Perú representa aproximadamente 20,9 millones (o 4,5% de la base instalada), ésta se estima que incrementará a una tasa anual de crecimiento compuesto del 50%, llegando a 157 millones en 2016 (o 28% de la base instalada de ese año). Es importante mencionar que una porción de la base instalada de smartphones será substituida por tablets, aunque dada la penetración embrionaria de estos terminales en la 6 región resulta difícil estimar con precisión este porcentaje. A la adopción de smartphones se suma el número de computadoras portátiles conectadas a internet mediante la banda ancha móvil. En el año en curso, se estima que estas conexiones alcanzan 1,3 millones en los cinco países arriba mencionados (o tan sólo 1,5% de la base instalada de unidades). Sin embargo, estos periféricos están creciendo a una tasa anual del 47%, lo que llevará la base instalada a un total de 9 millones en el año 2016 (ver figura A). La combinación de las tendencias arriba mencionadas, sumadas al despliegue de conexiones máquina a máquina (la llamada "internet de las cosas ) está resultando en un crecimiento dramático en el tráfico de datos que debe ser transportado por las redes móviles. Mientras que el tráfico de datos en los cinco países mencionados alcanzaba los 362 terabytes por mes en el año 2008, éste ha llegado a terabytes mensuales en el 2011, y proyectamos que alcanzará los terabytes por mes en el 2016, lo que implica una 7 tasa de crecimiento anual compuesta del 117%. Esta tasa de crecimiento produce una demanda creciente de capacidad en las redes móviles. Si no se da respuesta a este requerimiento mediante la asignación de espectro adicional, se produce una saturación de las redes y la consiguiente degradación del servicio, así también como el aumento de costos operaciones. FIGURA A. Crecimiento de base instalada de terminales generadoras de tráfico intensivo de datos (en millones de unidades) Smartphones PC y PDA corporativos móviles 157,4 9,0 Perú México 128,0 7,0 41,9 67,8 98,4 3,5 2,4 4,7 Colombia Brasil 2,7 8,3 1,0 21,0 1,3 0,7 0,1 0,3 Argentina Tasa anual de crecimiento compuesto 179% [ ] 50% [ ] 112% [ ] 47% [ ] Fuente: análisis TAS 6. Sin embargo, es importante considerar a futuro este aspecto dado que una tablet genera un promedio de 405 Mbps por mes en países avanzados, comparado con 79 Mbps por smartphone. Su uso crecerá al 122% (Fuente: Cisco. Visual Networking Index; Global Mobile Data Traffic Forecast Update, ). 7. Esta estimación es consistente con las proyecciones de Cisco, quien en su Visual Networking Index proyecta una tasa de crecimiento del tráfico de datos móviles en América Latina de 111% entre el 2010 y el De manera similar, IDATE, en su informe para el UMTS Forum, estima que el crecimiento mundial de tráfico de datos móviles entre 2010 y 2015 será de 94%. RESUMEN EJECUTIVO 03

53 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Necesidad de Espectro Radioeléctrico para Acomodar el Crecimiento de Tráfico Para responder a las necesidades crecientes de capacidad de red, la industria de las telecomunicaciones móviles necesita 8 acceder a más espectro radioeléctrico. Es en este contexto y en el caso de la Región 2 (Américas), que la Conferencia Mundial de Radiocomunicaciones 2007 (CMR-07) de la Unión Internacional de Telecomunicaciones (UIT), identificó la banda de MHz ( 700 MHz ) para IMT (Telecomunicaciones Móviles Internacionales). Los beneficios de esta posible asignación no sólo responden a la necesidad de acomodar de manera eficiente el tráfico de datos, sino que, dadas las características de mejor propagación de la señal en 700 MHz, permitirá promover el despliegue de redes de banda ancha en zonas rurales del continente, con el consiguiente impacto social positivo. Asimismo, la banda de 700 MHz permite mejorar la recepción de señal dentro de edificios en medios urbanos. Salvo algunos países de América Latina, la banda de 700 MHz está, en términos generales, poco utilizada. En Argentina la banda está prácticamente sin uso efectivo. Las licencias de radiodifusión habían sido mayormente otorgadas para servicio 9 de TV codificada por el ente regulador en carácter provisorio y ninguna es de alcance nacional, con muy bajo desarrollo (aproximadamente clientes). Sin embargo, en el mes de julio de 2011, de acuerdo a una Resolución de la AFSCA, se hizo pública la asignación de canales por arriba del 52 (en la banda de 700 MHz) a 15 universidades de Capital Federal y del Gran Buenos Aires. En el caso de Brasil, donde la banda se encuentra más ocupada que en el resto de la región, la banda por encima de 746 MHz está asignada a repetidoras de baja potencia y prevista para canales de televisión pública. En Colombia, la banda está relativamente ocupada, ya que existen 6 licencias nacionales, 7 regionales y 48 locales. México cuenta con 20 transmisoras en esta banda (19 abiertas y una por subscripción), la mayoría de ellas en ciudades 10 fronterizas. Finalmente, en Perú, la banda está marginalmente utilizada. Esto hace factible el escenario de reasignación de espectro en el corto plazo previo al apagón analógico. El Reglamento de Radiocomunicaciones de la UIT ya contempla la atribución del servicio móvil a título primario para la banda de 700 MHz en la Región 2 para promover el desarrollo de la banda ancha, y varios países en América Latina ya han adaptado sus respectivos cuadros nacionales de atribución de frecuencias. Adicionalmente, algunos gobiernos han comenzado a dar los primeros pasos prácticos tendientes a la reasignación de este espectro. Por ejemplo, en Perú, el Poder Ejecutivo ha establecido un plazo de 12 meses para reatribuir servicios de radiodifusión que operen en la banda de 700 MHz y se ha completado un proceso de consulta pública a tal 11 efecto. De manera similar, en Uruguay un decreto presidencial firmado en junio del 2011 determinó la liberación de la banda de 700 MHz para ofrecer telecomunicaciones móviles internacionales (IMT) y la sub-banda MHz para brindar servicios de televisión digital en el territorio nacional, con excepción del tramo comprendido entre los MHz. En Colombia, el Ministerio de las Tecnologías de Información y Comunicaciones (Ministerio TIC) anunció que el Dividendo 12 Digital en la banda de 700 MHz se asignará en el De manera similar, en México, siguiendo las recomendaciones de la UIT, COFETEL (Comisión Federal de Telecomunicaciones), la entidad regulatoria, tiene la intención de reasignar la banda de 700 MHz para servicios de telecomunicaciones, habiéndose concluido la primera consulta pública al respecto en diciembre En Argentina, por medio del Decreto Presidencial 1552/10 que crea el Plan Nacional de Telecomunicaciones Argentina Conectada, establece como prioridad planificar la utilización del espectro derivado del Dividendo Digital, originado por la adopción de la norma para la televisión digital". En el caso de Brasil, el regulador ANATEL ha expresado que la posible reasignación de la banda de 700 MHz habrá que esperar al final de la transición de la televisión analógica a la digital, prevista para El acceso a banda ancha móvil es una prioridad del gobierno brasilero, ya formulada por el Poder Ejecutivo en el Plan Nacional de Banda Ancha, lo que puede facilitar el estudio del uso de una parte de la banda de 700 MHz antes del cierre de la televisión analógica. Finalmente, es importante tomar en cuenta los trabajos que se están llevando a cabo dentro del marco de la Comisión Interamericana de Telecomunicaciones (CITEL), con el establecimiento de un grupo ad hoc para plantear el 8. La CITEL (Comisión Interamericana de Telecomunicaciones) estima que los operadores móviles de países latinoamericanos requerirán en el año 2020, 712 MHz adicionales en áreas de baja demanda y 1,161 MHz adicionales en áreas de alta demanda (CCPII/Rec.70 - XXII-02). 9. Autoridad Federal de Servicios de Comunicación Audiovisual (AFSCA), anteriormente llamada Comisión Federal de Radiodifusión (COMFER). 10. Adicionalmente, hay 8 emisoras planificadas pero no activadas todavía (Fuente: COFETEL, Dirección General de Radio y Televisión). 11. Ver Decreto Supremo MTC que modifica el artículo 28 del Reglamento General de la Ley de telecomunicaciones (http://www.osiptel.gob.pe/websiteajax/webformgeneral/sector/verlegislacionteleco.aspx) 12. Ministerio TIC abrirá proceso de asignación de espectro para servicios de 4G en el cuarto trimestre del 2011, 16 de junio de El 2 de septiembre de 2010 fue publicado el Decreto por el que "se establecen las acciones que deberán llevarse a cabo por la Administración Pública Federal para concretar la transición a la Televisión Digital Terrestre, que acelera las fechas originalmente planteadas en México. La transición que empezó en el 2004 deberá concluir las transmisiones de televisión analógica a partir del año 2011 y en su totalidad a más tardar el 31 de diciembre de El decreto está siendo combatido no sólo por empresas interesadas sino también por las cámaras legislativas que han dictaminado distintos puntos de acuerdo cuestionando sus términos. RESUMEN EJECUTIVO 04

54 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA espectro del Dividendo Digital resultante de la transición a la televisión digital y oportunidades para aplicaciones 14 convergentes. Apoyando el proceso decisorio encarado en la región, este estudio evalúa el escenario de asignación de la banda de 700 MHz que podría crear el mayor beneficio económico y social. Considerando la disponibilidad de la banda de 700 MHz, actualmente en uso para prestar servicios de radiodifusión, este trabajo estima el beneficio económico y social de dos escenarios alternativos: 1) asignar la banda de 700 MHz al servicio móvil para el despliegue de la banda ancha móvil o 2) preservar la banda de 700 MHz para ser utilizada por la radiodifusión. Para ello, este estudio compara la utilización del espectro en tres dimensiones: la contribución económica al ecosistema de las tecnologías de la información y comunicación (TIC), el beneficio económico, y el impacto social. En el primer aspecto, se estudia el beneficio comparado a ser generado para los proveedores de la industria de telecomunicaciones móviles (equipamiento de red, industria de la construcción, sistemas de información, etc.) y de televisión (producción de programas, industria de la construcción, etc.). Al mismo tiempo, se consideran los potenciales beneficios al tesoro público como resultado de la subasta de espectro a empresas privadas. También se realizó una estimación del beneficio a incurrir en el despliegue que sería posible al poder satisfacer la demanda de banda ancha en zonas no cubiertas mediante la utilización de la banda de 700 MHz, que se caracteriza por una mejor propagación de la señal. Asimismo, se estimó cuál sería la cobertura adicional que sería económicamente viable y que no sería 15 posible con la utilización de bandas superiores. Finalmente, se evaluó el excedente del productor a ser generado por la banda ancha móvil como resultado del cambio de asignación del espectro, suponiéndose que este excedente será trasladado en parte a los precios minoristas, beneficiando al usuario final. En el segundo módulo, se evalúa el impacto económico y social de ambos escenarios en términos de su contribución directa al PIB (producto interno bruto) nacional, como resultado de la oferta de bienes y servicios, y de la contribución indirecta, como resultado de los efectos de derrame y externalidades generadas en otros sectores de la economía. Al mismo tiempo, se calculó la creación de empleo directo e indirecto, la contribución a la recaudación impositiva y la creación de un excedente del consumidor. En el tercer módulo se evaluaron los beneficios sociales (por ejemplo, inclusión financiera, entrega de servicios de salud y educación, etc.) derivados de los dos escenarios de utilización de esta banda del espectro. A continuación, se detallan los resultados del estudio. Impacto Económico de la Asignación de la Banda de 700 MHz a la Banda Ancha Móvil En primer lugar, los resultados del análisis de contribución al ecosistema muestran una diferencia importante en la generación de valor en términos de adquisición de bienes y servicios según el escenario de utilización de la banda de 700 MHz. Así, si el Dividendo Digital fuese asignado a la banda ancha móvil, ésta contribuiría entre US $8.296 y millones en los cinco países estudiados en detalle ; para el resto de la región el valor es entre US $3.364 y millones. El rango está determinado por la proporción de espectro a ser subastado en caso de su asignación a la banda ancha móvil. Si la subasta típica en los cinco países estudiados en detalle fuese de 60 MHz, el monto estimado de valor a recaudar sumaría US $5.042 millones, mientras que si fueran subastados 90 MHz, el valor estimado sería de US $ Para el 17 resto de la región, el espectro podría recaudar entre US $1.259 y US $1.888 millones. Ésta es una estimación realizada tomando como base los precios alcanzados en las licitaciones que se han realizado hasta el momento en Europa y EEUU y las últimas subastas conducidas en América Latina. Consideramos importante resaltar que estos valores pueden verse modificados sustancialmente en función de las características de la metodología de asignación, de las condiciones concretas de los mercados en el momento en que se realicen las licitaciones, y, principalmente, las condiciones de las licencias, tales como cobertura, tiempo de despliegue, inversión mínima y otras obligaciones o restricciones. El resto de valor a ser generado incluye el correspondiente a inversiones en la adquisición de infraestructura, servicios operativos y servicios comerciales (ver figura B). Más allá del impacto en el ecosistema, el costo-beneficio de atribuir el espectro para la banda ancha móvil se manifiesta en otras dos áreas. En primer lugar, la utilización de ese espectro permite aumentar la cobertura de población y la velocidad de despliegue para satisfacer la demanda creciente de tráfico de datos. En caso de no contar con la banda de 700 MHz, los 14. Comité Consultivo Permanente II: Radiocomunicaciones incluyendo Radiodifusión, Resolución CCP.II/RES. 70 (XVI-10). Diciembre Obviamente, la extensión del servicio sería posible en bandas superiores, excepto que el aumento del número de sitios no es óptimo desde el punto de vista económico. 16. Esta suma es estimada si el pago por el espectro fuese de una sola vez. Se reconoce, sin embargo, que existen otros conceptos como el pago del espectro en sumas incrementales por año. 17. Esta suma no incluye pagos regulares por uso del espectro como tasas radioeléctricas (Argentina), pago de derechos (México), e impuestos municipales. RESUMEN EJECUTIVO 05

55 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA B. Impacto en la cadena productiva de la industria móvil (en millones de dólares) Adquisición de espectro Adquisición de bienes productivos Adquisición de servicios operativos Adquisición de servicios comerciales Inversión para la adquisición de espectro en licitación o concurso Sitios Equipamiento de acceso Red troncal Sistemas (OSS) Ingeniería civil Mantenimiento y reparación Distribución Logística Otros servicios Aplicaciones Publicidad Integración de sistemas comerciales (CRM, facturación, etc.) Argentina Brasil Colombia México Perú Resto AL Total Fuente: análisis TAS despliegues de las nuevas redes de tecnologías de 4G se haránen bandas superiores (1.900MHz, 1.700/2.100 MHz, 1.900/2.100 MHz, MHz), que tienen menor propagación, lo que requerirá un mayor número de sitios que si se utilizara la banda de 700 MHz. Un menor número de sitios en 700 MHz está además asociado con un gasto inferior de operación y mantenimiento. Finalmente, el menor número de sitios reduce el nivel de conflictividad con vecinos, quienes tienden a consistentemente oponerse al despliegue de infraestructura de torres y antenas. Aunado a esto, la utilización de la banda de 700 MHz por los servicios móviles permite conseguir mayor cobertura del territorio, ya que su alcance es significativamente mayor (10 kilómetros de radio, comparado con arangos inferiores a 5 18 kilómetros en otras frecuencias ). Así, el valor fundamental de la reasignación del espectro de 700 MHz conlleva la posibilidad de despliegue masivo de banda ancha móvil, ya que las inversiones adicionales requeridas para ofrecer este servicio a toda la población en todo el territorio en las otras bandas de frecuencia no se justifica económicamente. De esta manera, la cobertura de banda ancha móvil, que al día de hoy alcanza a 75% de la población en Argentina y Brasil, estimándose este número en 52% para Colombia, y 65% para Perú, podría extenderse significativamente ayudando a cerrar la brecha digital (ver figura C). Así, la cobertura de banda ancha móvil con el espectro de 700 MHz llegaría a un total estimado de la población latinoamericana de 92,7%, habiéndose incrementado la cobertura en 31.5 puntos porcentuales. Con una penetración promedio de banda ancha de 6,8% en América Latina, la cobertura adicional de banda ancha móvil permitirá aumentar significativamente la adopción de la banda ancha, el cual es un objetivo de política pública de la mayoría de los gobiernos de la región. Secundariamente, la utilización de la banda de 700 MHz por los servicios móviles permitiría cubrir aproximadamente 20,1 millones de personas en América Latina que hoy radican en áreas aisladas que incluso hoy no tienen cobertura de 19 telefonía móvil, lo que equivale a 4,8% de la población. Todo ello sería logrado con un ahorro de más de US $3.701 millones en el despliegue de las nuevas redes, de los cuales US $2.280 millones (equivalentes a US $3.690 millones nominales a lo largo de 8 años) corresponden a menor inversión y US $ Fuente: FCC. The broadband availability gap, OBI Technical Paper No.1. Abril La cobertura adicional sería aproximadamente de 1% en Argentina, 4% en Brasil, 6,4% en Colombia, casi 7% en México, y 4,5% en Perú. Estos valores se refieren al despliegue de todo tipo de red de telefonía móvil. Se debe considerar que para alcanzar cobertura universal de banda ancha móvil, la cobertura debería ser más elevada dado que estas cifras incluyen tecnología 2G. RESUMEN EJECUTIVO 06

56 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA C. Cobertura adicional de banda ancha móvil a ser alcanzada con base al espectro en 700 MHz 90% 95% 94% 95% 90% 92% 94% 89% 75% 75% 63% 52% 59% 65% 39% Argentina Brasil Colombia México Perú Fuente: análisis TAS Cobertura banda ancha fija Cobertura actual [3G] Cobertura estimada con la banda de 700 MHz millones a menores costos de operación. Para el resto de la 20 región, el ahorro en inversión y operación sería de US $ Éste es el valor del Dividendo Digital desde la perspectiva del despliegue de infraestructura. Alternativamente, si el espectro quedara atribuido y asignado a servicios de radiodifusión para la transmisión de señal de televisión, ésta contribuiría US $2.685 millones en los cinco países y US $823 millones en el resto de la región (ver figura D). Es importante mencionar que en esta estimación no se incluye costo alguno por la licitación de espectro, dado que es difícil establecer un valor dada la experiencia limitada en asignaciones 21 de este tipo. Como los números lo indican, una de las contribuciones más importantes en el ecosistema de la radiodifusión es la adquisición de servicios de programación para generar contenidos a ser transmitidos en las nuevas señales locales que podrían ser lanzadas en caso de que el espectro permaneciese usado por la televisión. Es importante destacar que este monto puede variar sustancialmente dependiendo del tipo de contenido que se está transmitiendo. El desarrollo de contenido original con producción de última generación se traduce en un costo elevado, mientras que programación básica (noticieros, compra de enlatados tales como documentales, novelas, películas y otro contenido disponible internacionalmente) puede implicar un costo significativamente menor al considerado en nuestra estimación. Se ha considerado un nivel básico de programación, estimado en US $1 millón por señal por mes. Evaluando el impacto económico y social, los resultados favorecen igualmente al escenario de asignación de espectro a la banda ancha móvil. En primer lugar, utilizar la banda de 700 MHz para prestar servicios de banda ancha móvil contribuye directa e indirectamente al PIB 7,0 veces más que la radiodifusión. En el caso de la contribución directa, consideramos la oferta de productos y servicios adicionales a los existentes generada por el acceso a la banda de 700 MHz. Ésta incluye abonados adicionales a banda ancha móvil derivados de una reducción de precios de 10% con base en una elasticidad de demanda de 0,6 (en zonas ya cubiertas), así como abonados nuevos residentes en zonas a ser cubiertas gracias a la banda de 700 MHz. En el caso de la televisión, los ingresos adicionales se refieren a la venta de publicidad por señales adicionales a las existentes por debajo del canal 51 y la suscripción pagada por 22 abonados a señales premium. Sin embargo, es importante mencionar que dado que la utilización del espectro por la radiodifusión sería hecha por canales de televisión abierta, la contribución en términos de venta de subscripción en ciertos países sería nula. 20. Cifra en valor presente neto descontada al 10% de 2012 a Un escenario posible es que, dadas las asignaciones previas, la licencia para la utilización de espectro sea otorgada sin cargo alguno. Aplicarían sólo los cargos recurrentes anuales por utilización del espectro. 22. En algunos países (como Argentina y Brasil), esta estimación puede ser considerada un tanto agresiva en la medida de que el Estado y organizaciones sin fines de lucro podrían ser el principal inversor en canales públicos adicionales, lo que determinaría que los servicios no serían por suscripción. RESUMEN EJECUTIVO 07

57 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA D. Impacto en la cadena productiva de la radiodifusión (en millones de dólares) Adquisición de espectro Adquisición de bienes productivos Adquisición de servicios operativos Adquisición de servicios comerciales Inversión para la adquisición de espectro en licitación o concurso Sitios Equipamiento de transmisión Repetidoras Ingeniería civil Mantenimiento y reparación Distribución Energía Programación (compra y producción) Facturación Argentina Brasil Colombia México Existe poca experiencia de licitación de espectro para radiodifusión El valor depende de cómo sea asignado el espectro Perú Resto AL Total Fuente: análisis TAS En el caso de la contribución indirecta de la banda ancha móvil, se estimaron las externalidades (o efectos de derrame en otros sectores de la economía). De manera conservadora, consideramos que solamente las conexiones de banda ancha móvil adicionales como consecuencia de la utilización de la banda de 700 MHz son las que generan un impacto económico. Estas suman US $1.062 millones en América Latina. La figura E presenta los resultados acumulados comparados entre ambos escenarios de estos efectos. FIGURA E. Ingresos adicionales del sector y contribución al crecimiento del producto interno bruto (en millones de unidades) 0 Argentina 300 Directa Indirecta 0 Brasil Colombia México Perú Resto AL América Latina x 7,0 Radiodifusión Banda ancha móvil Fuente: análisis TAS RESUMEN EJECUTIVO 08

58 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Así, al asignar el espectro de 700 MHz al servicio móvil, éste contribuye al PIB 7,0 veces más que la radiodifusión. En segundo lugar, la asignación de la banda de 700 MHz para el desarrollo de la banda ancha móvil contribuye significativamente más a la creación de empleo. En el caso de empleos directos (incluyendo aquí los empleados de las empresas de telecomunicaciones, así también como aquellos existentes en los proveedores a las mismas) a ser creados por cada industria, la radiodifusión tiende a crear un número mayor de puestos de trabajo: empleos comparados con para la industria móvil. Esto se debe a que en el caso de la televisión, la variable determinante es el número de señales de televisión digital nuevas a ser lanzadas al mercado, lo que multiplica de manera constante la cantidad de empleados necesarios por señal (entre 60 y 70, dependiendo del país).en la industria móvil, la variable determinante en la creación de empleos directos es el número de abonados adicionales incorporados como resultado de una mayor cobertura de los territorios nacionales y de un mayor número de líneas de banda ancha móvil. En este caso, dadas las importantes economías de escala de la industria móvil y el hecho de que en todos los países considerados la misma está operando a niveles óptimos de despliegue, el incremento marginal de empleados como resultado de las líneas adicionales es muy reducido. La diferencia importante en términos de generación de empleo entre ambos escenarios ocurre en virtud del efecto de creación de fuentes de trabajo en otras industrias. Dadas las externalidades ya debidamente verificadas de la banda ancha 23 móvil, la adopción de nuevas líneas de conexión de computadoras portátiles resulta en un efecto multiplicador en la 24 creación de empleo de otros sectores de la economía. Por otra parte, la televisión, más allá del efecto de publicidad adicional en las nuevas señales (factor que está limitado por la dimensión total del mercado publicitario), no genera empleo 25 indirecto significativo. El impacto en el empleo se presenta en la figura F. FIGURA F. Comparación de la contribución a la creación de empleo Argentina Directa Indirecta Brasil Colombia México Perú Resto AL América Latina x 2,1 Radiodifusión Banda ancha móvil Fuente: análisis TAS En tercer lugar, la utilización del espectro de 700 MHz para la banda ancha móvil contribuye a la recaudación impositiva en los cinco países estudiados en detalle US $2.141 millones más que la radiodifusión; para el resto de la región serían US $460 millones. En el caso de la radiodifusión, se consideran los impuestos directos (IVA o equivalente) acumulados en ocho años por ingresos adicionales de suscripción y publicidad generados por señales de televisión digital nuevas. En el caso de los servicios móviles, se consideran los impuestos directos (IVA o equivalente) por ingresos acumulados generados en 23. Ver, en particular, Katz, R. The impact of broadband on the economy: research to date and policy issues. International Telecommunication Union, Esto es medido de acuerdo a un modelo econométrico desarrollado con base en datos de panel de la economía chilena, el que indica que por cada 10% de penetración adicional de banda ancha, la tasa de empleo aumenta en 0,018% (ver Katz, R. "La contribución de la banda ancha al desarrollo económico", in Jordan, V., Galperin, H. y Peres, W. Acelerando la revolución digital: banda ancha para América Latina y el Caribe. CEPAL, 2010). 25. Por otra parte, como se mencionó arriba, todo aumento de empleo en producción y publicidad para la televisión está incluido en los efectos directos. RESUMEN EJECUTIVO 09

59 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA ocho años por el aumento de penetración de banda ancha móvil y expansión de cobertura en zonas rurales. Las cifras comparadas son presentadas en la figura G. FIGURA G. Comparación de la contribución impositiva acumulada ( ) (en millones de unidades) 0 Argentina Brasil Colombia México Perú Resto AL América Latina x 4, Radiodifusión Banda ancha móvil Nota: Esta contribución incluye sólo impuesto a las ventas, excluyéndose impuestos adicionales como tasa radioeléctrica por uso de espectro, derechos de uso, impuestos municipales, impuesto sobre la renta y otras tasas. Fuente: análisis TAS Finalmente, asignar la banda de 700 MHz para la banda ancha móvil genera beneficios importantes a través del excedente del consumidor. El excedente del consumidor mide la diferencia entre la voluntad de pago (como métrica de beneficio al consumidor) y el precio de un bien o servicio. En el caso de la radiodifusión, si bien la introducción de señales adicionales puede crear beneficios vinculados al servicio 26 público y la mejor información de ciudadanos, éstos son difíciles de cuantificar. Al mismo tiempo, corresponde mencionar que parte de estos efectos están también presentes en el caso de las telecomunicaciones móviles en la posibilidad de masificar la banda ancha ya que ésta permitiría a la ciudadanía poder acceder a las mismas señales de televisión y radio en línea, así también como poder tener acceso a redes sociales, blogs, etc. Por otra parte, es importante mencionar que, pese a la dificultad en cuantificar, la utilización del espectro por los servicios de radiodifusión genera un efecto de segundo orden que se traduce en mayor espacio publicitario creado por las señales adicionales con el consiguiente posible 27 excedente del productor y consumidor. Por el lado de la banda ancha móvil, el traslado de mayores economías de inversión de capital a precios de servicios de datos para computadoras conectadas y smartphones crea un excedente derivado de una reducción de precios en banda ancha móvil de aproximadamente 10%, beneficiando directamente al 28 consumidor. Midiendo esto en términos de la reducción acumulada de tarifas sobre la base total de usuarios a lo largo de ocho años, resulta en un excedente del consumidor de US $ millones para los cinco países estudiados en detalle (ver figura H) y US $ para el resto de los países latinoamericanos. Este excedente del consumidor contribuye a su vez al crecimiento del PIB en la medida que puede traducirse 26. Por ejemplo, calidad de vida, inclusión social, ciudadanía bien informada, y pertenencia a una comunidad (ver sección 7.7). 27. Es importante mencionar, sin embargo, que el posible aumento de espacio publicitario está limitado por el gasto total de publicidad que debe repartirse entre múltiples medios, los que incluyen, entre otros, el internet. 28. Una porción adicional de beneficios está determinada por la aplicación del modelo pre-pago a la banda ancha móvil, lo que tiene una importancia clave en facilitar su adopción. RESUMEN EJECUTIVO 10

60 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA H. Contribución por excedente del consumidor generado por la banda ancha móvil (en millones de dólares) Argentina Brasil Colombia México Perú Resto AL América Latina Fuente: análisis TAS 29 en mayor consumo. Por ejemplo, en el caso de México, la Encuesta Nacional de Ingresos y Gastos de los Hogares (ENIGH), realizada por el Instituto Nacional de Estadística, Geografía e Informática, indica que en el año 2008 la estructura de gastos promedio respondía a los siguientes rubros (ver figura I).Con base en este esquema de gastos, es posible concluir que una porción del excedente del consumidor estimado para México (US $530 millones) puede resultar en un aumento del consumo en rubros como artículos del hogar, 30 educación, esparcimiento y cuidado personal. FIGURA I. Grandes rubros del gasto corriente trimestral de los hogares (2008) Rubro Porcentaje del gasto Alimentos y bebidas 25,2% Vestido y calzado 3,9% Vivienda, energía y combustible 7,5% Artículos del hogar 4,5% Cuidados médicos y conservación de la salud 2,3% Transporte 13,8% Educación y esparcimiento 10,2% Cuidado Personal 7,5% Telecomunicaciones 1,1% Fuente: Gobierno de México. Instituto Nacional de Estadística, Geografía e Informática. 29. Ver en particular Greenstein, S. and McDevitt, R. The broadband bonus: accounting for broadband internet's impact on US GDP, NBER Working papers Investigaciones previas indican que gastos de alimentación y vivienda no aumentan como resultado de ahorro en otros rubros. Ver Martinez, A. Consumption Pattern Development across Mega-Cities: An Analysis of Sao Paulo and Shanghai. The Lauder Institute. University of Pennsylvania, April RESUMEN EJECUTIVO 11

61 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Impacto Social de la Asignación de la Banda de 700 MHz a la Banda Ancha Móvil Más allá de los beneficios económicos cuantificables, la asignación del espectro de 700 MHz a la banda ancha móvil en América Latina podrá tener una contribución social positiva en numerosas áreas. Por ejemplo, al expandir la provisión de banda ancha inalámbrica a zonas no cubiertas, la utilización del espectro permitirá a la población residente en áreas hoy sin acceso a banda ancha acceder a mayores recursos educativos, mejores servicios de salud y poder recibir servicios financieros. Al mismo tiempo, la banda ancha inalámbrica pasible de ser introducida en zonas rurales permitirá la provisión eficiente de servicios públicos a más alta velocidad de acceso, mejorando la interrelación entre la sociedad civil y la administración. De acuerdo a la comparación de ambos escenarios de uso del espectro, la banda ancha móvil representa una plataforma más eficiente para la provisión de servicios que aumentan el bienestar de los ciudadanos (ver figura J). Independiente de lo anterior, es importante mencionar que gran parte del impacto social de la radiodifusión ya están siendo cumplidos a partir del lanzamiento de señales de televisión digital. FIGURA J. Impacto social de escenarios alternativos de asignación de espectro Área de impacto Ejemplos Radiodifusión Banda ancha móvil Educación Conectividad a recursos edicacionales Educación a distancia Salud Telediagnóstico Comunicación interprofensional de salud Información sanitaria Inclusión financiera Acceso a plataformas de micropagos Educación para acceso a microfinanzas Acceso a servicios públicos Acceso a programas de gobierno electrónico Inclusión informativa Acceso a información estatal Programas de relación enrre ciudadanos y gobierno Fuente: análisis TAS Impacto Alto Impacto Nulo RESUMEN EJECUTIVO 12

62 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Conclusión En resumen, el estudio permite concluir que asignar el espectro de 700 MHz a los servicios de telecomunicaciones móviles en América Latina genera más valor económico y social que si éste permaneciese asignado a la radiodifusión (ver figura K). Estos resultados son consistentes con los generados por investigaciones realizadas en otras regiones del mundo (ver figura L). FIGURA K. Beneficio económico comparado según la utilización de la banda de 700 MHz (cifras en millones de dólares excepto empleo) Radiodifusión Banda ancha móvil Contribución al ecosistema de TIC (espectro, red y otros activos) $3.508 $ x 4,2 Ingresos adicionales del sector y contribución al crecimiento del PIB $513 $3.582 x 7,0 Generación de empleo directo e indirecto x 2,1 Impuestos (recaudación marginal adicional en ventas) $818 $3.420 x 4,2 Excedente del consumidor $0 (*) $5.157 (*) Efecto de segundo orden se traduce en mayor espacio publicitario con el consiguiente posible excedente del productor y consumidor Fuente: análisis TAS FIGURA L. Beneficio relativo si el espectro es asignado a la banda ancha móvil Américal Latina Asia (*) Unión Europea (**) Contribución al ecosistema de TIC (espectro, red y otros activos) x 4,2 N.A. x 2,9 (sin adquisición de espectro) Ingresos adicionales del sector y contribución al crecimiento del PIB x 7,0 x 9,3 x 4,8 Generación de empleo directo e indirecto x 2,1 x 22 x 1,3 Impuestos (recaudación marginal adicional en ventas) x 4,2 x 3,8 N.A. Excedente del consumidor $5,2 B N.A. 70 B (*) Boston Consulting Group Socio-economic impact of allocating 700 MHz band to mobile in Asia Pacific. Octubre. (**) SCF Associates The Mobile Provide Economic Impacts of Alternative Uses of the Digital Dividend. Septiembre Nota: Valor para banda ancha móvil dividido por el valor para la radiodifusión. Fuente: análisis TAS RESUMEN EJECUTIVO 13

63 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Así, los resultados del estudio indican los beneficios a ser generados por la asignación de la banda de 700 MHz a la telefonía móvil en los cinco países estudiados en detalle: Un aumento en cobertura de banda ancha a través de la disponibilidad de banda ancha móvil, elemento del cual hay consenso es fundamental para el crecimiento económico de América Latina; Un despliegue y operación de nuevas redes mas óptimo resultando en una reducción de inversión de US $3.701 millones comparada con el despliegue en bandas superiores, 31 alcanzando una mejor cobertura ; Contribución al ecosistema de TIC (adquisición de espectro y de equipamiento y servicios) que excede en más de US $8.129 millones a la contribución generada por la 32 radiodifusión ; Contribución directa (ingresos adicionales de la industria) e indirecta (externalidades positivas) al PIB que excede en más de US $ a la contribución generada por la 33 radiodifusión ; Creación de más de empleos directos e indirectos 34 adicionales a los generados por la radiodifusión ; Contribución impositiva adicional superior a US $2.142 millones; Excedente del consumidor adicional superior a US $3.842 millones. Extrapolando los resultados de los cinco países estudiados en detalle al resto de América Latina, los beneficios son, como es de esperar, mayores: Un incremento significativo en la cobertura de banda ancha móvil con una tecnología más eficiente resultando en un ahorro en despliegue y operación de red de US $5.440 millones comparado con redes en bandas superiores; Contribución al ecosistema de las Tecnologías de Información (adquisición de espectro y de equipamiento y servicios) que excede en más de US $ millones a la contribución generada por la radiodifusión; Contribución directa (ingresos adicionales de la industria) e indirecta (externalidades positivas) al PIB que excede en más de US $3.069 millones a la contribución generada por la radiodifusión; Creación de más de empleos directos e indirectos adicionales a los generados por la radiodifusión; Contribución impositiva adicional superior a US $2.602 millones; Excedente del consumidor adicional superior a US $5.157 millones. Alcanzar un aumento del 31,5% de la cobertura de banda ancha móvil de nueva generación, alcanzando un total estimado del 92,7% de la población latinoamericana. Esto permitirá aumentar significativamente la adopción de la banda ancha y alcanzar mayores velocidades, lo cual son objetivos de política pública de la mayoría de los gobiernos de la región. En conclusión, la reasignación de la banda de 700 MHz a los servicios móviles, en la medida que permite a estos ampliar la entrega de banda ancha móvil y acrecentar la cobertura del servicio conlleva efectos económicos y sociales substanciales, al mismo tiempo que responde a necesidades del mercado. Las autoridades públicas latinoamericanas no pueden soslayar este impacto y deben seguir el ejemplo de países como Colombia, Perú, Uruguay y México que al igual que otros países desarrollados están tomando decisiones que permitan la materialización de este impacto. 31. Valor nominal de US $6.155 millones (inversión: US $3.690 millones; costos operacionales: US $2.465 millones) calculado como Valor Presente Neto a ocho años descontado al 10%. 32. Gasto inicial no recurrente. 33. Contribución directa anual y contribución indirecta acumulada a ocho años. 34. Empleos directos anuales y empleos indirectos/año acumulados a ocho años. RESUMEN EJECUTIVO 14

64 GSMA / AHCIET ECONOMIC BENEFITS OF THE DIGITAL DIVIDEND FOR LATIN AMERICA PREFACE The Digital Dividend is, without any doubt, a central component of the future development and scope of the Information Society around the world. It embodies a critical resource for the democratization of Internet broadband access. While this topic concerns governments with regards to the policies regulating telecommunications and broadcasting, civil society needs to participate as well in the debate on how to meet its communication needs and what the most appropriate use of national resources, such as the frequency spectrum, is. The frequency spectrum that represents the Digital Dividend in Latin America is the result of technological progress of ICT (Information and Communication Technologies). It has significant propagation advantages that will allow expanding the present coverage of broadband services, reaching population that is not currently connected to the Internet, thereby reducing the "digital divide." However, to be able to use this portion of the spectrum to enlarge broadband coverage a change of paradigm is needed, that means basic change of the status quo of its current use. Given the criticality of such a change that will influence the future social and economic development of our countries, we believe in the need to gain access to the best possible tools in order to make public policy decisions that maximize the potential benefits to society. This is why we have commissioned a study to Telecom Advisory Services, LLC to deepen the analysis of empirical data that will help us understand the quantitative and qualitative impact of alternative usage scenarios of the Dividend Digital in some key Latin American countries. While several studies of this nature have already been conducted, this is the first one focused on countries of our region, relying on a methodology adapted to our context and generating forecast and estimations based on all the information available at this time. This study represents an important challenge we have decided to tackle in order to contribute to the much needed debate about the future of information and Internet access in our region. We believe that it is critical not to shy away from making decisions on the matter of what the best use of the Digital Dividend is. Delaying or underutilizing the use of a productive resource such us the radio spectrum that can promote individual and social development should definitely not be a policy option. September 22, 2011 Sebastian M. Cabello Director, GSMA Latin America Pablo Bello Secretary-General, AHCIET EXECUTIVE SUMMARY 15

65 GSMA / AHCIET ECONOMIC BENEFITS OF THE DIGITAL DIVIDEND FOR LATIN AMERICA EXECUTIVE SUMMARY 35 The purpose of this study is to provide decision makers in Latin American countries with a quantitative and qualitative estimation of the social and economic benefits of a prospective allocation of the "Digital Dividend" spectrum for the provision of mobile broadband services. The "Digital Dividend" is defined at the upper segment of the UHF band - "700 MHz" in the Latin American region - currently allocated to broadcasting services in most countries. As result of the transition of analog to digital television, this spectrum is being vacated, and could be reassigned to offer mobile broadband services in underserved areas. This would allow the provisioning of increased capacity to wireless services in order to meet the growth of data traffic, while increasing broadband coverage. This study is based on a detailed analysis of five countries in the Latin American region (Argentina, Brazil, Colombia, Mexico, and Peru), from which 36 results for the rest of the continent are extrapolated. The Growth of Data Traffic in Latin America Data traffic in Latin America is increasing significantly, partly driven by the introduction of mobile broadband services. This creates the need to gain access to more spectrum. Mobile telecommunications in the region have achieved massive adoption levels. Average penetration of wireless services in Latin America has reached 97.8% in 1Q01, which represents a level comparable to that of mature countries. Based on current penetration, the historical trend, and a conservative estimation of expected saturation levels, we estimate wireless telecommunications to reach an average penetration of 117% in 2015, 37 and 130% in In parallel with the diffusion of wireless telecommunications, operators in Latin America have been migrating their networks from second (2G) to third (3G) generation; some providers are 38 already testing fourth generation networks. In fact, in the course of 2012, the launch of 4G networks, based primarily on 39 the LTE (Long Term Evolution) standard, will begin. The migration to 3G technology represents a significant trend, since the devices that operate in standards such as HSPA (High Speed Packet Access) are better suited economically to offer a more efficient broadband access to the Internet than fixed broadband. As such, this technology represents an economic and technological response to market needs still constrained by the relatively high computer acquisition costs and the limits in fixed broadband deployment. The response to the need to gain access to the Internet, combined with a more efficient use of the frequency spectrum, will result in the migration of most subscribers in Latin America to 3G platforms in the course of the next ten years. According to our forecast of device substitution, by % of Latin American subscribers will be using either 3G or 4G devices. In some countries, driven by the significant increase of subscribers relying first on HSPA, and later on LTE, a large part of the installed base will be either 3G or 4G. For example, we estimate that by 2020, 87% of the installed base of wireless devices in Argentina, 73% in Mexico and 76% in Brazil will be 3G or 4G devices. The migration to 3G devices subsumes a second important trend: the adoption of smartphones. These devices are more developed functionally than feature phones in terms of both their user interface and screen formats. Therefore, since they are more suited to gain broadband access to the Internet than 35. This study was commissioned by a consortium formed by the GSM Association (GSMA), AHCIET (Asociación Iberoamericana de Centros de Investigación y Empresas de Telecomunicaciones), Telefónica, América Móvil, Telecom Italia, Qualcomm, and Intel. 36. The countries considered for extrapolating results include Bolivia, Chile, Costa Rica, Ecuador, El Salvador, Guatemala, Honduras, Nicaragua, Panama, Paraguay, Dominican Republic, Uruguay and Venezuela. 37. We estimate, conservatively that in 2015 wireless penetration will reach 150% in Argentina, 141% in Brazil, 100% in Colombia, 95% in Mexico, and 105% in Peru. 38. ITU Press Release, December 6th, Following a detailed evaluation against stringent technical and operational criteria, ITU has determined that LTE-Advanced and Wireless MAN-Advanced should be accorded the official designation of IMT-Advanced. As the most advanced technologies currently defined for global wireless mobile broadband communications, IMT-Advanced is considered as 4G, although it is recognized that this term, while undefined, may also be applied to the forerunners of these technologies, LTE and WiMax, and to other evolved 3G technologies providing a substantial level of improvement in performance and capabilities with respect to the initial third generation systems now deployed 39. In the second half of 2011, UNE (Empresa Publica de Medellín) in Colombia will be deploying LTE services in three cities. In 2012, ENTEL Chile will launch LTE services, which will be followed by Movistar (Telefónica) in Chile and Argentina, as well as Claro (AMX) in Chile. EXECUTIVE SUMMARY 16

66 GSMA / AHCIET ECONOMIC BENEFITS OF THE DIGITAL DIVIDEND FOR LATIN AMERICA conventional phones, smartphone subscribers tend to be heavy broadband users. While the smartphone installed base in Argentina, Brazil, Colombia, Mexico, and Peru currently amounts to approximately 20.9 million (or 4.5% of the total base), we estimate that it will grow at a compound annual growth rate of 50%, reaching 157 million by 2016 (or 28% of the installed base of that year). It is estimated, however, that a significant share of the smartphone installed base will be substituted by tablets, although given the embryonic penetration of these devices in the region, it is difficult to estimate 40 accurately this percentage. In addition to the growing diffusion of smartphones, the increasing number of personal computers connected to the Internet through mobile broadband peripherals (USB, air-cards, and dongles) should be considered. During 2011, the total number of personal computers connected to mobile broadband will reach 1.3 million in the five countries studied in detail (which amounts to 1.5% of the installed base). However, these peripherals are growing at a rate of 47%, which will result in an installed base of 9 million units in 2016 (see figure A). FIGURE A. Installed base of devices generating heavy data traffic (in millions) Smartphones 157,4 Mobile broadband connected PCs and other devices 9,0 Peru 128,0 7,0 Mexico 41,9 67,8 98,4 2,4 3,5 4,7 Colombia Brazil 1,0 2,7 8,3 21,0 0,1 0,3 0,7 1,3 Argentina Compound annual growth rate 179% [ ] 50% [ ] 112% [ ] 47% [ ] Source: TAS analysis The trends mentioned above, combined with deployment of "machine to machine" connections (what is labeled the "Internet of things"), result in a sizable growth of data traffic that has to be transported through mobile networks. Thus, while monthly mobile data traffic in the five countries under study was 362 terabytes in 2008, it reached 11,906 terabytes in 2011, and is projected to reach 180,214 terabytes in 2016, 41 which implies a compound annual growth rate of 117%. This growth rate results in a growing demand for wireless network capacity. If nothing is done to meet this requirement by assigning additional frequency spectrum, the networks risk of becoming saturated, with the consequent service degradation and an increase in operating costs. 40. It is important to consider this future trend since in mature countries a tablet user generates an average monthly traffic of 405 Mbps, compared to 79 Mbps for smarthpone subscribers. It is also estimated that in the future, tablet-generated traffic will grow at 122% (Source: Cisco. Visual Networking Index; Global Mobile Data Traffic Forecast Update, ). 41. This calculation is consistent with the estimates of Cisco, which projects a rate of mobile data growth in Latin America of 111% between 2010 and 2015 (See Cisco Visual Networking Index). Similarly, IDATE estimates in its report to the UMTS Forum that worldwide growth of mobile data traffic between 2010 and 2015 will be 94%. EXECUTIVE SUMMARY 17

67 GSMA / AHCIET ECONOMIC BENEFITS OF THE DIGITAL DIVIDEND FOR LATIN AMERICA Frequency Spectrum Needs to Support the Growth in Mobile Data Traffic In order to respond to the increasing needs for network capacity, the mobile telecommunications industry needs to gain 42 access to additional frequency spectrum. In this context, the World Radio Communications Conference 2007 (WRC-07) of the International Telecommunication Union (ITU) recommended the reallocation of the MHz ( 700 MHz ) band for IMT (International Mobile Telecommunications) in Region 2 (Americas). The benefits of such a move are not only to provide a more efficient way to handle data traffic, but also, given the propagation characteristics of signals in the 700 MHz band, allocating this spectrum to mobile broadband will allow to deploy broadband service in rural areas of the continent, with the resulting positive social impact. In addition, the 700 MHz band will improve indoor signal reception in urban areas. With the exception of a few countries in Latin America, the 700 MHz band is, typically, under-utilized. For example, in Argentina the band has very limited use. Until recently, broadcasting licenses had been provisionally assigned by the regulatory agency for use in encoded TV service. None of them were national in scope and all of them had limited development (approximately 30,000 clients). However, in July 2011, following 43 a resolution by AFSCA, channels above 52 (in the 700 MHz band) were licensed to 15 universities in the Buenos Aires region. In Brazil, the band is relatively more utilized than in the rest of the region. The band above 746 MHz is used by low power repeaters and is reserved to be licensed to public television channels. In Colombia, the band is relatively occupied by six national, seven regional, and 48 local licenses. Mexico has 20 signals in this band (19 free broadcasting television channels 44 and one paid), most of them in border towns. Finally, in Peru, the band is marginally utilized. This situation allows for the spectrum to be reassigned in the short term without waiting for the complete migration to digital television. The Radio Communications Rules of the International Telecommunication Union envision the primary allocation of the 700 MHz band in Region 2 to promote the development of broadband services. Accordingly, several countries in Latin America have already adapted their national frequency plans. Furthermore, some governments have already begun taking the first practical steps aimed at the reallocation of this band. For example, the Peruvian Executive has established a 12 month lapse to relocate broadcasting services operating in the 700 MHz band; a public consultation to that effect has already been 45 completed. Similarly, in Uruguay a presidential decree signed in June 2011 ordered the 700 MHz band to be vacated by any broadcasting signals in order to be utilized by international mobile telecommunications services, while the MHz sub-band would be utilized to offer national digital television services. In Colombia, the Ministry of Information and Communication Technologies announced that the Digital Dividend in 46 the 700 MHz band will be assigned in Similarly, in Mexico, following the recommendations of the ITU, COFETEL (Comisión Federal de Telecomunicaciones), the regulatory agency, is intended to reassign the 700 MHz for telecommunications services, having concluded the first public consultation 47 to that effect in December In Argentina, the Presidential Decree 1552/10 which creates the National Telecommunications Plan Argentina Conectada, established, as a priority, "to plan the use of the Digital Dividend spectrum freed by the migration to digital television." In Brazil, ANATEL, the regulatory agency, is considering that the potential reallocation of the 700 MHz band will occur at the completion of the transition to digital television, to be achieved by Access to mobile broadband is a priority for the Brazilian government, as stated by the Executive Branch in the National Broadband Plan. This could ease the consideration of using a portion of the 700 MHz band before the completion of the digital transition. Finally, it is important to consider the work conducted by the Inter-American Commission of Telecommunications (CITEL), which has "established an ad-hoc group to study the Digital Dividend spectrum resulting from the transition to digital television and 48 the opportunities to launch converging applications. In support of the decision-making process conducted in the region, this study evaluates two potential allocation options of the 700 MHz band that would create the larger economic and social benefit. Considering the availability of the 700 MHz band currently used for broadcasting services, the study estimates the social and economic benefits of two alternative scenarios: 1) assigning the 700 MHz band to mobile services for deployment of mobile broadband, or 2) using the band by broadcast television, maintaining the status quo. 42. The CITEL (Comisión Interamericana de Telecomunicaciones) estimates that Latin American mobile operators will require additional 712 MHz of spectrum in low demand areas and 1,161 MHz in high demand areas by 2020 CCPII/Rec.70 - XXII-02). 43. Autoridad Federal de Servicios de Comunicación Audiovisual (AFSCA), previously called Comisión Federal de Radiodifusión (COMFER). 44. Additionally, eight signals are planned but not in service as of now (Source: COFETEL, Dirección General de Radio y Televisión). 45. See Supreme Decree MTC, which modifies article 28 of the General Rule of the Telecommunications Law (http://www.osiptel.gob.pe/websiteajax/webformgeneral/sector/verlegislacionteleco.aspx) 46. Ministerio TIC abrirá proceso de asignación de espectro para servicios de 4G en el cuarto trimestre del 2011 June 16, On September 2, 2010 the Decreto por el que "se establecen las acciones que deberán llevarse a cabo por la Administración Pública Federal para concretar la transición a la Televisión Digital Terrestre, was published, resulting in accelerating the original transition plan. The transition, which began in 2004, will have to cease the broadcasting on analog signals by December, 31, This decree is being questioned not only by interested companies but also by Congress. 48. Consultative Permanent Committee II: Radio communications including Broadcasting, Resolution CCP.II/RES. 70 (XVI-10), diciembre EXECUTIVE SUMMARY 18

68 GSMA / AHCIET ECONOMIC BENEFITS OF THE DIGITAL DIVIDEND FOR LATIN AMERICA The assessment of alternative scenarios proceeds along three dimensions: economic contribution to the Information and Communication Technologies (ICT) ecosystem, the public economic benefit, and the social impact. In the first dimension, we compared the additional economic benefit to be generated to the providers of the telecommunications industry (network equipment, construction industry, information systems, etc.) and the television industry (program production, construction industry, etc.). In addition, we considered the potential benefits to the public treasury as a result of the licensing of spectrum to private sector firms. Finally, the study estimated the savings to be generated if additional broadband coverage in unserved areas were to be achieved through the use of the 700 MHz band, which has a better signal propagation. Similarly, we estimated the additional broadband coverage to be reached with the 700 MHz spectrum that would otherwise 49 not be achieved if using of higher bands. Based on these analyses, the study calculates the producer surplus to be generated as a result of a change in the allocation of spectrum, assuming that a portion of it will be transferred to retail prices, thereby benefiting end users. In the second module, the study evaluates the social and economic impact of each scenario in terms of its direct contribution to the Gross Domestic Product (GDP), as a result of the offer of additional goods and services enabled by the 700 MHz spectrum, as well as the spillover effects and positive externalities generated in other sectors of the economy. In addition, we calculated the creation of direct and indirect employment, the contribution to additional taxes and the creation of consumer surplus. In the third module the study evaluates the social benefits (for example, financial inclusion, e-health service delivery, and education) derived from each spectrum allocation scenario. The study results are reviewed in the following sections. Economic Impact of Allocating the 700 MHz Band to Mobile Broadband In the first place, the results of the analysis focusing on the contribution to the ICT ecosystem show an important difference between scenarios in terms of value to be generated from the acquisition of goods and services. If the Digital Dividend were to be used to provide to mobile broadband, it would contribute between US $8.3 and US $10.82 billion in 50 the five countries studied in detail ; for the rest of Latin America, the value would be between US $3.36 and US $3.99 billion. The range is driven by the portion of the spectrum to be licensed in case it were to be assigned to mobile broadband. If the typical licensed in the five countries studied in detail were of 60 MHz, the estimated proceeds would reach US $5.04 billion, while if 90 MHz were licensed, the amount would 51 reach US $7.56 billion. For the rest of Latin America, the spectrum to be licensed could generate between US $1.26 and US $1.89 billion. This estimation was made using, as a starting point, the prices observed in licensing processes conducted until now in Europe and the United States, as well as the latest ones held in Latin America. It is important to mention, however, that these values can change substantially as a function of the licensing approach to be followed, the specific market conditions at the time of occurrence, and, especially, the licensing conditions, such as coverage, deployment timeline, minimum investment, and other obligations and restrictions. The rest of the value to be generated comprises investment in infrastructure acquisition, operational services, and commercial services (see figure B). Beyond the contribution to the ICT ecosystem, the cost-benefit of allocating spectrum to mobile broadband is apparent in other areas. First, the utilization of the 700 MHz band allows the increase of speed of deployment of broadband services. If this band were not available for mobile broadband services, the deployment of 4G technology would have to be achieved using higher spectrum bands (1,900MHz, 1,700/2,100 MHz, 1,900/2,100 MHz, 2,500 MHz), which will require a larger number of radio base stations. Conversely, a smaller number of sites would also result in lower operations and maintenance expenses. Finally, a smaller number of sites would reduce the level of potential conflict due to the location of towers and antennas. As a corollary, the use of the 700 MHz for mobile broadband allows a larger coverage of the territory. In fact, signal propagation in 700 MHz allows for a 10 kilometer radius (or higher), 52 compared to 5 kms in other bands. Therefore, the fundamental value of the reallocation of the 700 MHz spectrum is embodied in the possibility of significantly increasing profitably in the deployment of mobile broadband, promoting a more suitable technology to foster adoption. The coverage of mobile broadband, which today reaches 75% in Argentina and Brazil, 52% in Colombia and 65% in Peru, could increase significantly, helping to close the digital divide (see figure C). 49. Obviously, the extension of service coverage would also be possible with higher bands; however, due to lower signal propagation, this would require an increase in the number of radio base stations, which would render the investment uneconomical. 50. This amount is estimated as a one-time payment. Other payments would have to be made annually for spectrum usage. 51. This sum does not include regular payments for the use of spectrum such as Frequency Levies (Argentina), payment of duties (Mexico), and municipal taxes. 52. Source: FCC, The broadband availability gap, OBI Technical Paper No.1, April EXECUTIVE SUMMARY 19

69 GSMA / AHCIET ECONOMIC BENEFITS OF THE DIGITAL DIVIDEND FOR LATIN AMERICA FIGURE B. Impact on the production chain of the mobile industry (in million US $) Spectrum acquisition Acquisition of production goods Acquisition of operational services Acquisition of commercial services In vestment for the acquisition of spectrum in either auction or contest Sites Access equipment Backbone Systems (OSS) Civil engineering Mantenance and repairs Distribution Logistics Other services Applications Advertising Commercial systems integration (CRM, billing, etc.) Argentina Brazil Colombia Mexico Peru Rest of Latam Total Source: TAS analysis FIGURE C. Additional mobile broadband achievable with the 700 MHz band 90% 95% 94% 95% 90% 92% 94% 89% 75% 75% 63% 52% 59% 65% 39% Argentina Brazil Colombia Mexico Peru Source: TAS analysis Fixed broadband coverage Current mobile broadband coverage [3G] Estimated mobile broadband coverage with 700 MHz band EXECUTIVE SUMMARY 20

70 GSMA / AHCIET ECONOMIC BENEFITS OF THE DIGITAL DIVIDEND FOR LATIN AMERICA Thus, mobile broadband coverage with the use of the 700 MHz band would reach an estimated total of 92.7% of the population of Latin America, increasing the reach of networks by 31.5 percentage points. With an average broadband penetration of 6.8% in the continent, the additional mobile broadband coverage would result in a significant increase in broadband adoption, which is a fundamental public policy objective of the majority of governments in the region. Additionally, the use of the 700 MHz band for mobile services will allow covering approximately 20.1 million people (or 4.8% of the population) in Latin America living today in isolated areas, 53 which today do not have access to mobile telephony. This 54 would be achieved with savings of more than US $3.7 billion in the deployment of new networks, of which US $2.28 billion (or US $3.69 billion in nominal value over 8 years) represent less investment and US $1.42 billion comprise lower operations costs. For the rest of the region beyond the five countries studied in detail, savings in capital investment and operations would reach US $1.74 billion. This is the Digital Dividend value form the perspective of infrastructure deployment. Alternatively, if the spectrum remained licensed to broadcasting services for the transmission of television signals, it would contribute US $2.69 billion in the five countries under study and US $823 million in the rest of the region (see figure D). It is important to mention that this estimate does not include contributions for spectrum licensing, given that it remains difficult to establish a firm value due to the limited experience 55 in licensing of this type. As the estimates indicate, one of the largest contributions to the broadcasting ecosystem is the acquisition of programming services to generate content to be distributed through the new local signals to be launched in case the spectrum remained licensed to the television industry. It is important to emphasize that this amount can vary substantially depending on the type of content that is being transmitted. For example, the development of original content with state-of-the-art production resources results in high costs, while basic programming (such as newscast, acquisition of syndicated programming like documentaries, soap operas, films, or other content available internationally) would result in significantly lower costs. In the study, we have considered acquisition of basic programming costs, amounting to monthly US $1 million per signal. FIGURE D. Impact on the production chain of the broadcasting industry (in million US $) Spectrum acquisition Acquisition of production goods Acquisition of operational services Acquisition of commercial services Investment for the acquisition of spectrum at auction or contest Sites Transmission equipment Repeters Civil engineering Maintenance and repair Distribution Energy Programming (acquisition and production) Billing Argentina Brazil Colombia Mexico There is little experience in broadcasting spectrum auctioning Value is dependent on mode of assignment Peru Rest of Latam Total Source: TAS analysis 53. The additional coverage would be approximately 1% in Argentina, 4% in Brazil, 6.4% in Colombia, 7% in Mexico, and 4.5% in Peru. These values refer to the deployment of all standard types. As mentioned above, in order to reach massive mobile broadband coverage, the coverage targets are higher given that these figures include 2G technology. 54. Amount calculated as the Net Present Value discounted at 10%, between 2012 and A potential scenario could be that, given the prior assignments, the licenses would be assigned without charge. Only the recurring charges for spectrum use would apply in this case (although for comparative purposes with the prior scenario, these were not included). EXECUTIVE SUMMARY 20

71 GSMA / AHCIET ECONOMIC BENEFITS OF THE DIGITAL DIVIDEND FOR LATIN AMERICA When it comes to evaluating the economic and social impact, the results again favor the allocation of spectrum to mobile broadband. First, the utilization of the 700 MHz band to offer mobile broadband services contributes directly and indirectly to the GDP seven times more than broadcasting. In the case of the direct contribution, the study estimates the amount generated by offering additional products and services to be introduced as a result of gaining access to the 700 MHz band. This amount comprises additional mobile broadband adoption resulting from a reduction of prices of 10% (a transfer of a portion of the savings mentioned above), assuming a demand elasticity of 0.6 (in areas already covered). In addition, it considers new subscribers to be gained in new areas to be covered as a result of gaining access to the 700 MHz band. In the case of television, the additional revenues are based on the sale of advertising by the new signals to be broadcast under channel 51 as well as the subscription paid for premium 56 signals. It is important to mention, however, that given that in most cases the spectrum in the broadcasting case would be used by public television, the contribution in terms of subscription would in some countries be nil. In the case of the indirect contribution of mobile broadband to the GDP, the study estimates the externalities (or spillover effects on other sectors of the economy); conservatively, we have considered only the economic impact derived from additional mobile broadband connections resulting from the use of the 700 MHz band. This is estimated to reach US $1.06 billion in Latin America. Figure E compares the cumulative results for both allocation scenarios. As shown in figure E, using the 700 MHz band to provide mobile broadband services would contribute seven times more to the GDP than broadcasting. Secondly, by assigning the 700 MHz band to the development of mobile broadband, the industry could contribute significantly more than broadcasting to job creation. In the case of direct employment (which includes the employees of the telecommunications service providers, as well as those selling services to the carriers) to be created by each industry, broadcasting tends to generate a higher number of jobs: 3,870 compared to 3,000 for the mobile telecommunications industry. This is because in the case of broadcasting, the key variable driving new employment is the number of digital television signals to be deployed, which multiplies by a constant the number of employees required by signal (between 60 and 70 depending on the country). In the mobile telecommunications industry, the key variable in direct job creation is the number of additional subscribers resulting from a larger coverage of the territory and the decline in prices. In this case, given the important economies of scale of the mobile telecommunications industry and the fact that the service providers in most countries in the region are operating at optimal levels, the marginal increase of employees as a result of additional mobile subscribers is small. FIGURE E. Additional revenues and contribution to GDP growth (in million US $) 0 Argentina 300 Direct Indirect 0 Brazil Colombia Mexico Peru Rest of Latam Latin America x 7,0 Broadcasting Mobile broadband Source: TAS analysis 56. In some countries (such as Argentina and Brazil), this estimate can be considered aggressive because the State and non-profit organizations could be the main investors in public broadcasting signals; thus, these services would not be subscription-based and may not be able to fetch advertising revenues. EXECUTIVE SUMMARY 21

72 GSMA / AHCIET ECONOMIC BENEFITS OF THE DIGITAL DIVIDEND FOR LATIN AMERICA The important difference in terms of job creation between both scenarios occurs in terms of their impact on other industries. Given the externalities already documented for the broadband 57 industry, the adoption of new mobile broadband connections will have a multiplier effect in job creation in other sectors of the 58 economy. On the other hand, beyond the impact of additional advertising of the new broadcasting signals (a factor which is limited by the total size of the advertising market), the television 59 industry does not generate significant indirect jobs. The impact of alternative scenarios is presented in the figure F. Thirdly, the utilization of the 700 MHz spectrum for mobile broadband contributes to the collection of taxes in the five countries under study US $2.14 billion more than broadcasting; for the rest of the region the difference would amount to US $460 million. In the case of broadcasting, we considered direct taxes (VAT or equivalent) accumulated over eight years resulting from additional sales in advertising and subscription generated by new digital television signals. In the case of mobile services, we have considered direct taxes (VAT or equivalent) on cumulative sales generated over eight years due to the growth in mobile broadband penetration and the increase in coverage. The comparative amounts are presented in figure G. FIGURE F. Comparative contribution to job creation Argentina Direct Indirect Brazil Colombia Mexico Peru Rest of Latam Latin America x 2,1 Broadcasting Mobile broadband Source: TAS analysis Finally, by using the 700 MHz band to provide to mobile broadband services, important benefits can be generated in terms of consumer surplus. This is measured in terms of the difference between willingness to pay (as a measure of user benefit) and the price paid for a good or service. In the case of broadcasting, while the introduction of additional signals can create benefits linked to the delivery of public services and better information to 60 citizens, they are difficult to quantify. Furthermore, these effects would also be present in the case of mobile broadband since the service would allow the population to gain access to the same radio and television signals through the internet, in addition to access to social networks, blogs, etc. Furthermore, it is important to mention that, while difficult to quantify, the use of spectrum for broadcasting generates a second order effect that can result in more advertising vehicles (e.g. additional television signals) 61 with the consequent potential producer and consumer surplus. From the mobile broadband perspective, the transfer of larger savings in capital investment to retail prices of mobile broadband services results in a consumer surplus derived from a 62 reduction of prices of 10%, directly benefiting consumers. When measuring this benefit in terms of the cumulative decrease in tariffs over the total user base during 57. See, in particular, Katz, R. The impact of broadband on the economy: research to date and policy issues. International Telecommunication Union, This is measured according to an econometric model developed based on a data panel for the Chilean economy, which indicates that for each 10% increase in additional broadband penetration, the employment rate increases by 0.018% (see Katz, R. "La contribución de la banda ancha al desarrollo económico", in Jordan, V., Galperin, H. y Peres, W. Acelerando la revolución digital: banda ancha para América Latina y el Caribe. CEPAL, 2010). 59. On the other hand, as mentioned above, all increases in employment in production and advertising are included in the direct effects. 60. For example, quality of life, social inclusion, well-informed citizenry, and social belonging (see section 7.7. in the study). 61. It is important to mention however, that the potential increase in advertising vehicles is limited by the total spending of advertising which would have to be split among multiple media, which also includes the Internet. 62. An additional portion of benefits results from applying the pre-paid model to mobile broadband, which plays a key role in stimulating adoption. EXECUTIVE SUMMARY 22

73 GSMA / AHCIET ECONOMIC BENEFITS OF THE DIGITAL DIVIDEND FOR LATIN AMERICA FIGURE G. Comparative tax collection ( ) (in million US $) 0 Argentina Brazil Colombia Mexico Peru Rest of Latam Latin America x 4, Broadcasting Mobile broadband Note: This contribution includes only taxes on revenues, excluding additional levies such as any radioeletric frequency tax for spectrum usage, usage rights, municipal taxes, and other taxes. Source: TAS analysis eight years, we can estim ate a total surplus of US $3.84 billion for the five countries studied in detail and US $1.32 billion for the rest of the countries in Latin America (see figure H). FIGURE H. Contribution resulting from consumer surplus generated by mobile broadband (in million US $) Argentina Brazil Colombia Mexico Peru Rest of Latam Latin America Source: TAS analysis EXECUTIVE SUMMARY 23

74 GSMA / AHCIET ECONOMIC BENEFITS OF THE DIGITAL DIVIDEND FOR LATIN AMERICA This consumer surplus contributes in turn to GDP growth 63 insofar that it can lead to more consumption. For example, in the case of Mexico, the National Census of Household Income and Spending (ENIGH), conducted by the National Institute of Statistics, Geography and Informatics, indicates that in 2008 average household spending was broken down according to the following items (see figure I). FIGURE I. Larger items of current household spending by quarter (2008) Item Percent of spending Food and beverage 25,2% Clothing 3,9% Housing, energy, and fuel 7,5% Household articles 4,5% Health care 2,3% Transportation 13,8% Education and entertainment 10,2% Personal care 7,5% Telecommunications 1,1% Source: Gobierno de México. Instituto Nacional de Estadística, Geografía e Informática Considering the above breakdown, it is reasonable to conclude that a portion of the consumer surplus estimated for Mexico (US $530 million) could yield an increase in consumption of ítems such as household articles, education, entertainment and personal care 64. Social Impact of Allocating the 700 MHz Spectrum to Mobile Broadband Beyond the quantifiable economic benefits, the allocation of the 700 MHz band to mobile broadband in Latin America will have a positive social contribution in several areas. For example, the expansion of wireless broadband to unserved zones will allow the population without current coverage to gain access to more educational resources, improved health services, and financial services. At the same time, wireless broadband to be introduced in rural areas will enable the efficient provision of public services at a greater speed of access, improving the interrelationship between civil society and governments. Based on the comparison of both scenarios, mobile broadband represents a more efficient platform for the provision of services that improve the welfare of citizens (see figure J). Furthermore, it should be emphasized that a large part of the social impact of broadcasting has already been achieved through the launch of digital television signals. 63. See, in particular, Greenstein, S. and McDevitt, R. The broadband bonus: accounting for broadband internet's impact on US GDP, NBER Working papers Prior research indicates that spending in food and housing tend not to increase as a result of savings in other ítems. See Martinez, A. Consumption Pattern Development across Mega-Cities: An Analysis of Sao Paulo and Shanghai. The Lauder Institute. University of Pennsylvania, April EXECUTIVE SUMMARY 24

75 GSMA / AHCIET ECONOMIC BENEFITS OF THE DIGITAL DIVIDEND FOR LATIN AMERICA FIGURE J. Comparative social impact of spectrum alternative allocation scenarios Impact area Examples Broadcasting Mobile broadband Education Connectivity to educational resources Distance learning Health Tele-diagnostics Health professional communications Preventive health care information Financial inclusion Access to micro-payments platforms Education for micro-finance access Acces to public services Acces to e-government programs Information inclusion Access to government information Programs linking citizens and government Source: TAS analysis High impact No impact Conclusion To sum up, the study concludes that the allocation of the 700 MHz spectrum to mobile broadband in Latin America will generate more economic and social value than if it were to continue being assigned to broadcasting (see figure K). FIGURE K. Comparative economic benefits according to the utilization of the 700 MHz band (in million US $, except for employment) Broadcasting Mobile broadband Contribution to the ICT ecosystem (spectrum, network, other assets) $3.508 $ x 4,2 Additional revenues and contribution to GDP growth $513 $3.582 x 7,0 Direct and indirect employment creation x 2,1 Taxes (collected on additional sales) $818 $3.420 x 4,2 Consumer surplus $0 (*) $5.157 (*) Second-order effect which translates into more advertising space with the consequent potencial impact on producer and consumer surplus. Source: TAS analysis EXECUTIVE SUMMARY 25

76 GSMA / AHCIET ECONOMIC BENEFITS OF THE DIGITAL DIVIDEND FOR LATIN AMERICA These results are consistent with those generated by research in other regions of the world (see figure L). FIGURE L. Relative benefit if spectrum is used for mobile broadband services Latin America Asia (*) European Union (**) Contribution to the ICT ecosystem (spectrum, network, other assets) x 4,2 N.A. x 2,9 (without spectrum licensing ) Additional revenues and contribution to GDP growth x 7,0 x 9,3 x 4,8 Direct and indirect employment creation x 2,1 x 22 x 1,3 Taxes (collected on additional sales) x 4,2 x 3,8 N.A. Consumer surplus $5,2 B N.A. 70 B (*) Boston Consulting Group Socio-economic impact of allocating 700 MHz band to mobile in Asia Pacific. October. (**) SCF Associates The Mobile Provide Economic Impacts of Alternative Uses of the Digital Dividend. September Note: Value generated by mobile broadband divided by value generated by broadcasting. Source: TAS analysis Thus, the results of the study indicate the benefits to be generated as a result of using the 700 MHz band to provide mobile broadband services in the five countries studied in detail: An increase in broadband coverage resulting from increased availability of mobile broadband, a fundamental variable to ensure economic growth in Latin America; More optimal deployment and operation of new networks, resulting in a reduction of capital investment of US $3.7 billion compared to deployment of infrastructure in higher frequency 65 bands, while achieving better coverage ; Contribution to the ICT ecosystem (acquisition of spectrum, equipment, and services) in excess of US $8.13 billion 66 compared to the contribution generated by broadcasting ; Direct (additional revenues to the industry) and indirect contribution (positive externalities) to GDP that exceeds US 67 $2.47 billion to the contribution generated by broadcasting ; Creation of more than 4,600 direct and indirect additional 68 jobs than those created by broadcasting ; Additional taxes in excess of US $2.14 billion; A consumer surplus greater than US $3.84 billion. By extrapolating the results of the five countries studied in detail to the rest of Latin America, the benefits are, as expected, or a greater magnitude: 65. Nominal value of US $6.16 billion (investment: US $3.69 billion, and operating expenses: US $2.47 billion) calculated as the Net Present Value discounted at 10% over eight years. 66. Initial non-recurring expenses. 67. Annual direct contribution and cumulative indirect contribution over eight years. 68. Annual direct employment and indirect job/years accumulated over eight years. EXECUTIVE SUMMARY 26

77 GSMA / AHCIET ECONOMIC BENEFITS OF THE DIGITAL DIVIDEND FOR LATIN AMERICA A significant increase in mobile broadband coverage with a more efficient technology, resulting in savings in capital investment and network operations of US $5.44 billion when compared to deploying similar networks in higher bands; Contribution to the ICT ecosystem (acquisition of spectrum licenses, equipment, and services) in excess of US $11.3 billion to the contribution generated by broadcasting; Direct (additional revenues of the industry) and indirect (positive externalities) contribution to GDP in excess of US $3.07 billion to that generated by the broadcasting industry; Creation of 5,540 direct and indirect more jobs than those generated in broadcasting; Tax contribution of US $2.60 billion more than in broadcasting; Consumer surplus in excess of US $5.16 billion; An increase of 31.5% of mobile broadband coverage, leading to a total estimated coverage of 92.7% of the population of Latin America. This will lead to significantly increase broadband adoption and deliver services at higher speeds, which represent a public policy objective of most governments in the region. In summary, the reassignment of the 700 MHz band to mobile services, by allowing the enhanced deployment of mobile broadband and increasing coverage, entails significant economic and social effects, while responding to market needs. Policy makers in Latin America should consider these effects and follow the example of countries like Colombia, Peru, Uruguay, and Mexico, as well as that of other mature countries that are making spectrum reallocation decisions leading to their realization. EXECUTIVE SUMMARY 27

78 GSMA / AHCIET BENEFÍCIOS ECONÔMICOS DO DIVIDENDO DIGITAL PARA A AMÉRICA LATINA PREFÁCIO O Dividendo Digital é, sem nenhuma dúvida, um elemento central no futuro desenvolvimento e alcance da Sociedade da Informação do mundo inteiro. Encontramo-nos hoje face a um recurso crucial para democratizar o acesso à internet de banda larga. Embora este seja um tema que atinge todos os governos a través de suas políticas regulatórias de telecomunicações e radiodifusão, é a sociedade em seu conjunto quem deve participar do debate sobre o modo em que devem ser satisfeitas as suas necessidades, bem como dos melhores usos que se pode dar aos recursos de propriedade de um país, tais como, neste caso, o espectro radioelétrico. O espectro que conforma o Dividendo Digital na América Latina é resultado do progresso tecnológico das Tecnologias da Informação e da Comunicação (TICs) e conta com vantagens de propagação que permitiriam expandir a fronteira de alcance dos serviços de banda larga já existentes e conectar a segmentos de população ainda não conectados reduzindo a chamada brecha digital. Porém, a utilização deste espectro para a massificação da banda larga requer uma mudança de paradigma, basicamente uma ruptura do status quo em relação ao uso que se deu até agora a esta porção do espectro. Por ser esta uma mudança crítica e determinante do futuro desenvolvimento econômico e social dos países, julgamos que é preciso contar com as maiores ferramentas possíveis ao nosso alcance, que permitam tomar as decisões de políticas públicas mais adequadas e que maximizem os possíveis lucros para a sociedade. É com esta motivação que comissionamos o presente estudo à Telecom Advisory Services, LLC, a fim de que possam aprofundar na análise dos argumentos empíricos que nos ajudem a entender o impacto, tanto qualitativo como quantitativo, que teriam os usos alternativos do espectro do Dividendo Digital em alguns países da América Latina. Embora já se tenham realizado diversos estudos deste tipo, este é o primeiro que se realiza para os países da nossa região, com uma metodologia que se adapta ao contexto no qual estamos imersos e faz previsões com base em toda a informação disponível no momento. Este estudo foi para nós um desafio muito importante que nos animamos a enfrentar a fim de fazer uma contribuição ao debate que é necessário ter sobre o futuro que se vai oferecer aos cidadãos da nossa América Latina. Achamos que o pior que pode acontecer a respeito deste debate e desta oportunidade é não tomar decisões e atrasar ou subutilizar a exploração de um recurso produtivo que pode ser um promotor de desenvolvimento quer individual, quer coletivamente. 22 de Septiembre de 2011 Sebastian M. Cabello Director, GSMA Latin America Pablo Bello Secretario General, AHCIET RESUMO EXECUTIVO 28

79 GSMA / AHCIET BENEFÍCIOS ECONÔMICOS DO DIVIDENDO DIGITAL PARA A AMÉRICA LATINA RESUMO EXECUTIVO 69 Este estudo visa proporcionar aos tomadores de decisão dos países da América Latina uma avaliação qualitativa e quantitativa do benefício econômico e social que pode ser obtido se o "dividendo digital" for atribuído à indústria de telecomunicações móveis para a sua utilização primordial em serviços de banda larga móvel. O dividendo digital é definido como o segmento superior da faixa UHF 700 MHz no caso da América Latina atualmente atribuído aos serviços de radiodifusão na maioria dos países, e que, como conseqüência da transição da televisão analógica à digital, fica liberado, podendo assim ser atribuído aos serviços de banda larga móvel. Isto permite dar uma maior capacidade à telefonia móvel para responder ao crescimento de tráfego de dados e aumentar a cobertura destes serviços. O estudo está baseado numa análise detalhada de cinco países da região (Argentina, Brasil, Colômbia, México e Peru), a partir do qual se extrapolam os resultados 70 para o resto da região. O Crescimento do Tráfego de Dados na América Latina O tráfego de dados na América Latina vem aumentando de maneira importante, em grande medida pela chegada da banda larga móvel; isto está gerando uma necessidade cada vez mais importante de espectro. As telecomunicações móveis atingiram níveis massivos de penetração no continente latino-americano. A média continental, de 97,8 % no primeiro trimestre de 2011, representa um nível de adoção relativamente semelhante ao observado em países industrializados. Baseando-nos na situação atual, na tendência histórica, e numa estimativa conservadora de níveis de saturação esperados, estimamos que a penetração da telefonia móvel atingirá uma média regional 71 superior aos 117% em 2015, chegando aos 130% em De maneira simultânea com a adoção acelerada das telecomunicações móveis, os operadores da América Latina estão migrando as suas redes de tecnologias de segunda (2G) para terceira geração (3G), enquanto alguns já começaram fazer 72 testes com a quarta geração (4G). Por volta de 2012, serão observados o início de migração para plataformas 4G, com base 73 principalmente no padrão LTE (Long Term Evolution). A migração para tecnologias 3G é importante na medida em que os terminais que operam nestas tecnologias, por exemplo, HSPA (High Speed Packet Access), são mais adequados economicamente para prover um acesso de banda larga eficiente à internet do que aqueles com as tecnologias de banda larga fixa. O 3G representa uma resposta relativamente altos custos de aquisição de computadores e os limites na implantação da banda larga fixa. A satisfação da necessidade da população de ter acesso à internet, combinada com uma utilização mais eficiente do espectro radioelétrico, determina que a transição dos usuários na região a plataformas 3G será completada até o final da presente década. Assim, as nossas projeções da taxa de substituição de terminais mostram que em 2015, 46,2% dos assinantes da região estarão utilizando terminais 3G e 4G. Certos países da região, devido ao aumento dramático de usuários com tecnologia HSPA e a posterior implementação do LTE, irão registrar uma maioria da base instalada em terminais de terceira e quarta geração. Por exemplo, de acordo com as nossas projeções, estimamos que em 2020, 87% da base instalada na Argentina, 73% no México e 76 % no Brasil serão terminais de 3G e 4G. 69. Este estudo foi patrocinado pelo consórcio formado pela AHCIET (Associação Ibero-americana de Centros de Pesquisa e Empresas de Telecomunicações), a Associação GSM (GSMA, pelas suas siglas em inglês), Telefônica, América Móvil, Telecom Italia, Qualcomm e Intel 70. Os países considerados para a extrapolação de resultados incluem a Bolívia, Chile, Costa Rica, Equador, El Salvador, Guatemala, Honduras, Nicarágua, Panamá, Paraguai, República Dominicana, Uruguai e Venezuela. 71. Conservadoramente estimamos que em 2015, a penetração na Argentina poderá atingir 150%, no Brasil 141%, na Colômbia 100%, no México 95% e no Peru 105%. 72. Comunicado de imprensa a UIT, 6 de dezembro de 2010,.Following a detailed evaluation against stringent technical and operational criteria, ITU has determined that LTE-Advanced and Wireless MAN-Advanced should be accorded the official designation of IMT-Advanced. As the most advanced technologies currently defined for global wireless mobile broadband communications, IMT-Advanced is considered as 4G, although it is recognized that this term, while undefined, may also be applied to the forerunners of these technologies, LTE and WiMax, and to other evolved 3G technologies providing a substantial level of improvement in performance and capabilities with respect to the initial third generation systems now deployed 73. No segundo semestre de 2011, UNE (Empresa Pública de Medellín) na Colômbia estará lançando serviços em LTE em três cidades. Em 2012, ENTEL Chile fará o lançamento e espera-se que seja seguida por ações por parte da Movistar (Telefônica) no Chile e na Argentina, e Claro (América Móvil) no Chile. RESUMO EXECUTIVO 29

80 GSMA / AHCIET BENEFÍCIOS ECONÔMICOS DO DIVIDENDO DIGITAL PARA A AMÉRICA LATINA A migração para terminais 3G inclui uma tendência importante: a adoção de smartphones. A funcionalidade destes terminais é mais avançada que a dos telefones móveis básicos (feature phones) pelo fato de prover interfaces e formatos de tela mais adequados para o uso da internet. A adoção de smartphones representa uma tendência fundamental a ser estudada na medida em que a conveniência destes terminais para o acesso à internet determina que os usuários de smartphones tendam a utilizar a linha móvel de modo mais intenso. Embora na atualidade a base instalada deste tipo de terminais na Argentina, no Brasil, na Colômbia, no México e no Peru represente aproximadamente 20,9 milhões (4,5% da base instalada), estima-se que esta aumente a uma taxa anual de crescimento composto de 50%, chegando a 157 milhões em 2016 (28% da base instalada desse ano). É importante mencionar que uma parte da base de terminais existente de smartphones será substituída por tablets, embora, dada a penetração embrionária destes terminais na região, torne-se 74 difícil estimar com precisão esta percentagem. À adoção de smartphones deve-se acrescentar o número de computadores portáteis (notebooks) conectados à internet via banda larga móvel. Durante o ano de 2011, estima-se que estas conexões possam atingir 1,3 milhões nos cinco países acima citados (ou apenas 1,5% da base de terminais existente). Porém, estes periféricos estão crescendo a uma taxa anual de 47%, o que vai elevar a base de terminais existente a um total de 9 milhões em 2016 (ver figura A). A combinação das tendências supracitadas, somadas à implantação de conexões máquina a máquina (a chamada "internet das coisas ) está resultando num crescimento dramático do tráfego de dados que deve ser transportado pelas redes móveis. Enquanto o tráfego de dados nos cinco países mencionados atingia os 362 terabytes por mês no ano de 2008, este chegou a terabytes mensais em 2011, e projetamos que irá atingir terabytes por mês em 2016, o que supõe uma taxa de crescimento anual composta 75 de 117%. Esta taxa de crescimento gera uma demanda crescente de capacidade nas redes móveis. Se este requerimento não for atendido, através da alocação de espectro adicional, as redes ficarão saturadas, com conseqüente degradação do serviço, bem como o aumento dos custos operacionais. FIGURA A. Crescimento da base instalada de terminais geradores de tráfego intensivo de dados (em milhões de unidades) Smartphones PC e PDA corporativos móveis 157,4 9,0 Peru 128,0 7,0 México 41,9 67,8 98,4 2,4 3,5 4,7 Colômbia Brasil 1,0 2,7 8,3 21,0 0,1 0,3 0,7 1,3 Argentina Taxa anual de crescimento composto 179% [ ] 50% [ ] 112% [ ] 47% [ ] Fonte: análise TAS 74. Porém, é importante no futuro considerar este aspecto, pois uma tablet gera na média 405 Mbps ao mês em países desenvolvidos, comparado com 79 Mbps gerados por um smatrphone. O seu uso crescerá a uma taxa de 122% ao ano nos próximos anos (Fonte: Cisco. Visual Networking Index; Global Mobile Date Traffic Forecast Update, ). 75. Esta estimativa é consistente com as projeções da Cisco, que no seu Visual Networking Index projeta uma taxa de crescimento do tráfego de dados móveis na América Latina de 111% entre os anos 2010 e De maneira similar, a IDATE, em seu relatório para o UMTS Forum, estima que o crescimento mundial de tráfego de dados móveis entre 2010 e 2015 será de 94%,. RESUMO EXECUTIVO 30

81 GSMA / AHCIET BENEFÍCIOS ECONÔMICOS DO DIVIDENDO DIGITAL PARA A AMÉRICA LATINA Necessidade de Espectro Radioelétrico para Suportar o Crescimento de Tráfego Para responder às necessidades crescentes de capacidade de rede, a indústria de telecomunicações móveis necessita ter 76 acesso a mais espectro radioelétrico. É neste contexto e no caso da Região 2 (Américas), que a Conferência Mundial de Radiocomunicações de 2007 (CMR-07) da União Internacional de Telecomunicações (UIT), identificou a faixa de MHz ( 700 MHz ) para IMT (Telecomunicações Móveis Internacionais). Os benefícios desta possível atribuição não se referem apenas à possibilidade de acomodar de maneira eficiente o tráfego de dados, mas também, dadas as características de melhor propagação do sinal em 700 MHz, promover a implementação de redes de banda larga em zonas rurais do continente, com o conseqüente impacto social positivo. De igual modo, a faixa de 700 MHz permite melhorar a recepção do sinal dentro dos prédios em meios urbanos. Com exceção de alguns países da América Latina, a faixa de 700 MHz está, em termos gerais, pouco utilizada. Na Argentina a faixa está praticamente sem uso efetivo. As licenças de radiodifusão foram concedidas para serviço de TV codificada 77 pela órgão regulador com caráter provisório e nenhuma delas é de alcance nacional, com uma penetração muito baixa (aproximadamente 30,000 clientes). Porém, em julho de 2011, segundo uma resolução da AFSCA, foi publicada a outorga de canais acima do canal 52 (na faixa de 700 MHz) para 15 universidades da Capital Federal e Grande Buenos Aires. No caso do Brasil, onde a faixa está mais ocupada que no resto da região, a faixa acima de 746 MHz está destinada à retransmissoras de baixa potência e prevista para canais de televisão pública. Na Colômbia, a faixa está relativamente ocupada; existem 6 licenças nacionais, 7 regionais e 48 locais. México tem 20 transmissoras nesta faixa (19 de televisão aberta e uma por assinatura), a maior parte delas em cidades 78 da fronteira. Finalmente, no Peru, a faixa está pouco utilizada. Isto torna factível o cenário de reatribuição do espectro para telecomunicações móveis no curto prazo antes do apagão analógico. O Regulamento de Radiocomunicações da UIT já considera a atribuição da faixa de 700 MHz na Região 2 para o serviço móvel em caráter primário para promover o desenvolvimento da banda larga, evários países na América Latina já alteraram os seus respectivos quadros nacionais de atribuição de freqüências. Adicionalmente, alguns governos já começaram a dar os primeiros passos direcionados à nova utilização do espectro. Por exemplo, no Peru, o Poder Executivo estabeleceu um prazo de 12 meses para reacomodar os serviços de radiodifusão que operem na banda de 700 MHz e foi completado um processo de consulta pública nesse sentido. De 79 modo similar, no Uruguai um decreto presidencial assinado em junho de 2011 determinou a liberação da faixa de 700 MHz para oferecer telecomunicações móveis (IMT) e a sub-faixa MHz para serviços de televisão terrestre, com exceção da faixa de MHz. Na Colômbia, o Ministério de Tecnologias da Informação e Comunicação (Ministério TIC) declarou que o dividendo digital na faixa de MHz será outorgado em De maneira similar, no México, seguindo as recomendações da UIT, COFETEL (Comisión Federal de Telecomunicaciones), a entidade reguladora, tem intenções de realocar a banda de 700 MHz à serviços de telecomunicações terrestres, tendo-se concluído a 81 primeira consulta pública a esse respeito em dezembro Na Argentina, o Decreto Presidencial 1552/10, que cria o Plano Nacional de Telecomunicações Argentina Conectada, estabelece como prioridade planejar a utilização do espectro derivado do dividendo digital, originado pela adoção da norma para a televisão digital". No Brasil, a ANATEL tem expressado que a possível realocação da faixa de 700 MHz terá que esperar até o final da transição da televisão analógica à digital, prevista para Entretanto,o acesso a banda larga móvel é uma prioridade do governo brasileiro, já formulada pelo Poder Executivo no Plano Nacional de Banda Larga, o que poderia antecipar o uso de uma parte da faixa de 700 MHz antes do fim da televisão analógica. Finalmente, é importante considerar os trabalhos que estão sendo conduzidos dentro da Comissão Interamericana de Telecomunicações (CITEL), com o estabelecimento de um grupo ad-hoc para planejar o uso do espectro do dividendo digital, resultado da transição à televisão digital e oportunidades para aplicações 82 convergentes. 76. A CITEL (Comissão Interamericana de Telecomunicações) estima que os operadores móveis de países latino-americanos requeiram no ano de 2020, 712 MHz adicionais em áreas de baixa demanda e MHz adicionais em áreas de alta demanda (CCPII/Rec.70 - XXII-02). 77. Autoridad Federal de Servicios de Comunicación Audiovisual (AFSCA), antigamente denominada Comisión Federal de Radiodifusión (COMFER). 78. Adicionalmente, há 8 emissoras planejadas mas não em operação (Fonte: COFETEL, Dirección General de Radio y Televisión). 79. Ver Decreto Supremo MTC que modifica o artigo 28 do Regulamento Genral da Lei de Telecomunicações (http://www.osiptel.gob.pe/websiteajax/webformgeneral/sector/verlegislacionteleco.aspx) 80. Ministerio TIC abrirá proceso de asignación de espectro para servicios de 4G en el cuarto trimestre del junho Em 2 setembro de 2010 foi publicado o Decreto em que "se establecen las acciones que deberán llevarse a cabo por la Administración Pública Federal para concretar la transición a la Televisión Digital Terrestre, que antecipa as datas originalmente planejadas no México. A transição, começada em 2004, deverá concluir as transmissões de televisão analógica, começando em 2011 y finalizando em 31 dezembro de O decreto está sendo combatido não só pelas empresas interessadasmas também pelo Poder Legislativo que está questionando o conteúdo. 82. Comitê Consultivo Permanente II: Radiocomunicações incluindo Radiodifusão, Resolução CCP.II/RES. 70 (XVI-10), dezembro de RESUMO EXECUTIVO 31

82 GSMA / AHCIET BENEFÍCIOS ECONÔMICOS DO DIVIDENDO DIGITAL PARA A AMÉRICA LATINA Apoiando o processo decisório enfrentado na região, este estudo avalia o cenário de atribuição da faixa de 700 MHz que poderia criar um maior benefício econômico e social. Considerando a disponibilidade da faixa de 700 MHz, atualmente em uso para prestar serviços de radiodifusão, este trabalho estima o benefício econômico e social de dois cenários alternativos: 1) atribuir a faixa de 700 MHz à banda larga móvel, ou 2) preservar a faixa de 700 MHz para ser utilizada pela radiodifusão. Para isto, este estudo compara a utilização do espectro em três dimensões: a contribuição econômica ao ecossistema das tecnologias da informação e da comunicação (TIC), os benefícios econômicos e o impacto social. No primeiro módulo, se estudou o valor comparado a ser gerado para os fornecedores da indústria de telecomunicações móveis (equipamento de rede, indústria da construção, sistemas de informação, etc.) e de televisão (produção de programação, indústria da construção, etc.). Ao mesmo tempo, se consideram os potenciais benefícios à receita pública como resultado da licitação de espectro a empresas privadas. Também se realizou uma estimativa das economias na construção e operação de rede que seriam possíveis em caso de poder satisfazer a demanda por banda larga em regiões não atendidas a partir da utilização da faixa de 700 MHz, caracterizada por uma maior cobertura devido a maior propagação do sinal. De igual modo, estimou-se qual seria a cobertura adicional que resultaria economicamente viável e que não seria possível com a utilização de faixas superiores. Finalmente, avaliou-se o excedente do produtor a ser 83 gerado pela banda larga móvel como resultado da mudança de atribuição do espectro, supondo-se que este excedente será transferido em parte aos preços no varejo, beneficiando o usuário final. No segundo módulo, avalia-se o impacto econômico e social de ambos cenários em termos da sua contribuição direta ao PIB (produto interno bruto) nacional, como resultado da oferta de bens e serviços, e da contribuição indireta, como resultado dos efeitos de espalhamento e impactos externos gerados em outros setores da economia. Ao mesmo tempo, calculou-se a criação de emprego direto e indireto, a contribuição à arrecadação tributária e a criação de um excedente de consumidor. No terceiro módulo, foram avaliados os benefícios sociais (por exemplo, inclusão financeira, serviços de saúde, educação, etc.) decorrentes dos cenários de utilização desta faixa de espectro. A seguir, estão detalhados os resultados do estudo. Impacto Econômico da Alocação da Faixa de 700 MHz à Banda Larga Móvel Em primeiro lugar, os resultados da análise de contribuição ao ecossistema mostram uma diferença importante na geração de valor em termos de aquisição de bens e serviços segundo o cenário de utilização da faixa de 700 MHz. Assim, se o dividendo digital fosse alocado à banda larga móvel, esta iria contribuir entre US $8,296 e 10,815 bilhões nos cinco países 84 estudados em detalhe ; para o resto da região o valor é de entre US $3,364 e 3,993 bilhões. O intervalo é determinado pela proporção de espectro a ser leiloado no caso da sua mudança de alocação à banda larga móvel. Se o leilão típico nos cinco países estudados em detalhe fosse de 60 MHz, o valor estimado a arrecadar seria de US $5,042 bilhões, enquanto se fossem leiloados 90 MHz, o valor estimado seria 85 de US $7,561 bilhões. Para o resto da região, esse espectro poderia arrecadar entre US $1,259 e 1,888 bilhões. Esta estimativa foi realizada considerando como base os preços do espectro pagos nas licitações que aconteceram até hoje na Europa e nos Estados Unidos, junto com as licitações recentes na América Latina. Consideramos importante destacar que estes valores podem variar substancialmente em função das características da metodologia de licitação, das condições existentes nos mercados no momento da realização da licitação, e, principalmente, das condições das licenças, tais como obrigações de cobertura, tempo para implementação, investimento mínimo e outras obrigações ou restrições. O resto do valor a ser gerado inclui o correspondente a investimentos na aquisição de infra-estrutura, serviços operacionais e serviços comerciais (ver figura B). Mais além do impacto no ecossistema, o custo-benefício de alocar o espectro à banda larga móvel se manifesta em outras duas áreas. Em primeiro lugar, a utilização desse espectro permite aumentar a população atendida a velocidade de implementação para satisfazer a demanda crescente de tráfego de dados. No caso de não contar com a faixa de 700 MHz, os desenvolvimentos das novas redes com tecnologias 4G se farão em faixas superiores (1.900 MHz, MHz/2.100 MHz, MHz), que têm menor propagação, o que requererá um maior número de sites do que se fosse utilizada a banda de 700 MHz. Um menor número de sites está também associado a um custo inferior de operação e manutenção. Finalmente, o menor número de sites reduz o nível de conflitos com a sociedade, as quais tendem, consistentemente, a se opor à instalação de infra-estrutura de torres e antenas. Somado a isto, a utilização da faixa de 700 MHz pelos serviços móveis permite conseguir uma maior cobertura, uma vez que o seu alcance é significativamente maior (10 quilômetros de raio, comparado com alcances inferiores a 5 quilômetros em outras 86 freqüências ). Deste modo, o valor fundamental da realocação 83. Obviamente, a extensão do serviço seria possível em bandas superiores, mas o aumento do número de ERBs não seria otimizado do ponte de vista econômico. 84. Este valor é estimado se o pagamento pela licença de uso do espectro fosse feito de uma única vez. Porém, se reconhece que existem outros conceitos como o pagamento do espectro em somas incrementais por ano. 85. Este valor não inclui pagamentos regulares por uso do espectro, tais como taxas radioelétricas (Argentina), pagamento de direitos (México), e impostos municipais. 86. Fonte: FCC, The broadband availability gap, OBI Technical Paper No.1, abril de 2010 RESUMO EXECUTIVO 32

83 GSMA / AHCIET BENEFÍCIOS ECONÔMICOS DO DIVIDENDO DIGITAL PARA A AMÉRICA LATINA FIGURA B. Impacto na cadeia produtiva da indústria móvel (em milhões de dólares) Aquisição do espectro Aquisição de bens produtivos Aquisição de serviços operativos Aquisição de serviços comerciais Investimento para a aquisição de espectro em licitação ou concurso Bens imóveis Equipamento de aceso/ ERBs Rede de transporte Sistemas (OSS) Engenharia civil Manutenção e reparo Distribuição Logística Outros serviços Aplicações Propaganda Integração de sistemas comerciais (CRM, faturamento, etc.) Argentina Brasil Colômbia México Peru Resto AL Total Fonte: análise TAS FIGURA C. Cobertura adicional da banda larga móvel a ser alcançada com a utilização da faixa de 700 MHz 90% 95% 94% 95% 90% 92% 94% 89% 75% 75% 63% 52% 59% 65% 39% Argentina Brasil Colômbia México Peru Fonte: análise TAS Cobertura banda larga fixa Cobertura anual [3G] Cobertura estimada com a faixa de 700 MHz RESUMO EXECUTIVO 33

84 GSMA / AHCIET BENEFÍCIOS ECONÔMICOS DO DIVIDENDO DIGITAL PARA A AMÉRICA LATINA do espectro de 700 MHz traz a possibilidade do desenvolvim ento massivo da banda larga móvel, já que os investimentos adicionais necessários para prestar este serviço a toda a população em todo o território utilizando outras faixas de freqüência não se justifica economicamente. Assim, a cobertura da banda larga móvel, que hoje alcança 75% da população na Argentina e no Brasil, 52% na Colômbia e 65% no Peru, poderá aumentar significativamente, ajudando a diminuir a exclusão digital (ver figura C). Deste modo, a cobertura da banda larga móvel utilizando o espectro de 700 MHz poderá chegar a aproximadamente 92,7% da população latinoamericana, o que implica um aumento de cobertura de 31,5 pontos percentuais. Com uma penetração média de banda larga de 6,8% na América Latina, a cobertura adicional da banda larga móvel permitirá aumentar de maneira importante o atendimento com banda larga, o que é um objetivo de política pública da maioria dos governos na região. Como efeito secundário, a utilização da faixa de 700 MHz pelos serviços móveis permitirá cobrir aproximadamente 20,1 milhões de pessoas na América Latina que hoje se encontram em áreas isoladas que não dispõem de cobertura 87 de telefonia móvel, o que equivale a 4,8% da população. Tudo isto seria conseguido com uma economia de mais de US $3,7 88 bilhões, dos quais US $2,28 bilhões (equivalentes a US $ 3,69 bilhões nominais ao longo de 8 anos) correspondem a um menor investimento e US $1,42 bilhões a menores custos de operação. Para o resto da região, a economia em investimento e operação seria de US $1,74. Este é o valor do dividendo digital desde a perspectiva da construção de infra-estrutura. Alternativamente, se o espectro ficasse atribuído à radiodifusão para a transmissão de sinais de televisão, esta contribuiria com US $2,685 bilhões nos cinco países e US $ 823 milhões no resto da região (ver figura D). É importante mencionar que nesta estimativa não se inclui valor algum pela licitação do espectro, uma vez que é difícil estabelecer um 89 valor dada a experiência limitada em atribuições deste tipo. Como os números indicam, uma das contribuições mais importantes no ecossistema da radiodifusão é a aquisição de serviços de programação para gerar conteúdos a serem FIGURA D. Impacto na cadeia produtiva da radiodifusão (em milhões de dólares) Aquisição de espectro Aquisição de bens produtivos Aquisição de serviços operativos Aquisição de serviços comerciais Investimento para a aquisição de espectro em licitação ou concurso Bens imóveis Equipamento de transmissão Repetidoras Engenharia civil Manutenção e reparo Distribuição Energia Programação (compra e produção) Faturamento Argentina Brasil Colômbia México Peru Existe pouca experiência de leiloes de espectro para radiodifusão O valor depende da forma de alocação do espectro Resto AL Total Fonte: análise TAS 87. A cobertura adicional seria aproximadamente de 1% na Argentina, de 4% no Brasil, de 6,4% na Colômbia, de quase 7% no México, e 4,5% no Peru. Estas quantias referem-se a implementação de todo tipo de rede de telefonia móvel. Deve-se considerar que para alcançar uma cobertura universal de banda larga móvel, a cobertura deveria ser mais elevada dado que estas quantias incluem tecnologia 2G. 88. Cifra em valor presente líquido descontado os 10% de 2012 a Um cenário possível é que, dadas as atribuições prévias, a licença para a utilização de espectro seja outorgada sem cargo algum. Aplicariam apenas os encargos recorrentes anuais por utilização do espectro. RESUMO EXECUTIVO 34

85 GSMA / AHCIET BENEFÍCIOS ECONÔMICOS DO DIVIDENDO DIGITAL PARA A AMÉRICA LATINA transmitidos nos novos sinais locais que poderiam ser lançados no caso do espectro permanecer utilizado pela televisão. É importante destacar que esta quantia pode variar substancialmente dependendo do tipo de conteúdo que estiver sendo transmitido. O desenvolvimento de conteúdo original com produção de última geração se traduz num custo elevado, enquanto a programação básica (noticiários, compra de enlatados tais como documentários, filmes e outros conteúdos disponíveis internacionalmente) implica num custo significativamente menor ao considerado em nossa estimativa. Considerou-se um nível básico de programação, estimado em US $1 milhão por sinal por mês. Avaliando o impacto econômico e social, os resultados favorecem novamente o cenário de alocação de espectro à banda larga móvel. Em primeiro lugar, utilizar a faixa de 700 MHz para prestar serviços de banda larga móvel contribui direta e indiretamente ao PIB 7,0 vezes mais do que a radiodifusão. No caso da contribuição direta, consideramos a oferta de produtos e serviços adicionais aos existentes, gerada pelo acesso à faixa de 700 MHz. Esta inclui assinantes adicionais à banda larga móvel devido à uma redução de preços de 10% com base em uma elasticidade da demanda de 0,6 (em zonas já cobertas), assim como novos assinantes em zonas rurais que estariam cobertos graças a faixa de 700 MHz. No caso da televisão, as receitas adicionais referem-se à venda de publicidade por canais adicionais aos existentes abaixo do canal e à assinatura paga para acesso a canais premium. É importante destacar que, como a utilização do espectro por serviços de radiodifusão seria feita por canais de televisão aberta, a contribuição em termos de venda de assinaturas em alguns países seria nula. No caso da contribuição indireta da banda larga móvel, se estimaram impactos externos (ou efeitos de espalhamento em outros setores da economia); de maneira conservadora, consideramos que apenas as conexões adicionais de banda larga móvel como conseqüência da utilização da faixa de 700 MHz são as que geram um impacto econômico. Estas somam US $1,062 bilhões na América Latina. A figura E apresenta os resultados acumulados destes efeitos, comparados entre ambos os cenários. Assim, ao atribuir o espectro de 700 MHz ao serviço móvel, este contribui ao PIB 7,0 vezes mais do que a radiodifusão. Em segundo lugar, a alocação da faixa de 700 MHz para o desenvolvimento da banda larga móvel contribui significativamente mais à criação de emprego. No caso de empregos diretos (incluindo aqui os empregados das empresas de telecomunicações, bem como aqueles dos fornecedores das mesmas) a serem criados por cada indústria, a radiodifusão tende a criar um número maior de postos de trabalho: empregos comparados com para a indústria móvel. Isto FIGURA E. Receitas adicionais do setor e contribuição ao crescimento do Produto Interno Bruto (em milhões de dólares) 0 Argentina 300 Direta Indireta 0 Brasil Colômbia México Peru Resto AL América Latina x 7,0 Radiodifusão Banda larga móvel Fonte: análise TAS 90. Em alguns países (como Argentina e Brasil), esta estimativa pode ser considerada um tanto agressiva na medida em que o Estado e ONG s poderiam ser o principal investidor em canais públicos adicionais, o que determinaria que os serviços não seriam por assinatura. RESUMO EXECUTIVO 35

86 GSMA / AHCIET BENEFÍCIOS ECONÔMICOS DO DIVIDENDO DIGITAL PARA A AMÉRICA LATINA é devido a que no caso da televisão, a variável determinante é o número de canais novos de televisão digital a serem lançados ao mercado, o que multiplica de maneira constante a quantidade de empregados necessários por canal (entre 60 e 70, dependendo do país). Na indústria móvel, a variável determinante na criação de empregos diretos é o número de assinantes adicionais incorporados como resultado de uma maior cobertura dos territórios nacionais e de um maior número de linhas de banda larga móvel. Neste caso, dadas as importantes economias de escala da indústria móvel e o fato de que em todos os países considerados a mesma está operando a níveis ótimos de desenvolvimento, o incremento marginal de empregados como resultado dos acessos adicionais é muito reduzido. A diferença importante em termos de geração de emprego entre ambos cenários acontece em virtude do efeito de criação de fontes de trabalho em outras indústrias. Tendo em conta os impactos externos já devidamente verificados da banda larga 91 móvel, a implementação de novas linhas de conexão de computadores portáteis resulta num efeito multiplicador na 92 criação de emprego em outros setores da economia. De outra forma, a televisão, além do efeito de publicidade adicional nos novos canais (fator que vem limitado pela dimensão total do 93 mercado publicitário), não gera emprego indireto significativo. O impacto no emprego se apresenta na figura F. FIGURA F. Comparação da contribuição à criação de emprego Argentina Direta Indireta Brasil Colômbia México Peru Resto AL América Latina x 2,1 Radiodifusão Banda larga móvel Fonte: análise TAS Em terceiro lugar, a utilização do espectro de 700 MHz para a banda larga móvel contribui para a arrecadação de impostos nos cinco países estudados em profundidade com US $2,141 bilhões a mais do que a radiodifusão; para o resto da região seriam US $460 milhões. No caso da radiodifusão, consideramse os impostos diretos (IVA ou equivalente) acumulados em oito anos por receitas adicionais de assinatura e publicidade gerados por canais novos de televisão digital. No caso das telecomunicações móveis, se consideram os impostos diretos (IVA ou equivalente) por receitas acumuladas geradas em oito anos pelo aumento de penetração da banda larga móvel e a expansão de cobertura em zonas rurais. As quantias comparadas são apresentadas na figura G. 91. Ver, em particular, Katz, R. The impact of broadband on the economy: research to date and policy issues. International Telecommunication Union, Isto é medido de acordo com um modelo econométrico desenvolvido com base em dados de painel da economia chilena, o qual indica que para cada 10% de penetração adicional de banda larga, a taxa de emprego aumenta em 0,018 % (ver Katz, R. "La contribución de la banda ancha al desarrollo económico", em Jordan, V., Galperin, H. y Peres, W. Acelerando la revolución digital: banda ancha para América Latina y el Caribe. CEPAL, De outra forma, como já se mencionou acima, todo aumento de emprego em produção e publicidade para a televisão está incluído nos efeitos diretos. RESUMO EXECUTIVO 36

87 GSMA / AHCIET BENEFÍCIOS ECONÔMICOS DO DIVIDENDO DIGITAL PARA A AMÉRICA LATINA FIGURA G. Comparação da contribuição de impostos acumulada ( ) (em milhões de dólares) 0 Argentina Brazil Colômbia México Peru Resto AL América Latina x 4, Radiodifusão Banda larga móvel Nota: Esta contribuição inclui somente o imposto sobre vendas, excluindo-se impostos adicionais como a taxa radioelétrica por uso de espectro, direitos de uso, impostos municipais, imposto de renda e outras taxas. Fonte: análise TAS Finalmente, alocar a faixa de 700 MHz ao serviço de banda larga móvel gera benefícios importantes através do excedente do consumidor. O excedente do consumidor mede a diferença entre a vontade de pagamento (como métrica de benefício ao consumidor) e o preço de um bem ou serviço. No caso da radiodifusão, embora a introdução de canais adicionais possa criar benefícios relacionados ao serviço público e a melhor 94 informação dos cidadãos, estes são difíceis de quantificar. Ao mesmo tempo, vale mencionar que parte destes efeitos também estão presentes no caso das telecomunicações móveis na possibilidade de massificar a banda larga, já que esta permitiria aos cidadãos poder ter acesso aos mesmos sinais de televisão e rádio online, bem como poder ter acesso a redes sociais, blogs, etc. De outra forma, é importante lembrar que, pese a dificuldade em quantificar, a atribuição de espectro à radiodifusão gera um efeito de segunda ordem que se traduz num maior espaço publicitário criado pelos canais adicionais com o possível excedente do produtor e consumidor. Do lado da telefonia móvel, a transferência de 95 maiores economias de investimento de capital a preços de serviços de dados para computadores conectados e smartphones cria um excedente derivado de uma redução dos preços da banda larga móvel de aproximadamente 10%, beneficiando diretamente o consumidor. Medindo isto em termos da redução 96 acumulada de tarifas sobre a base total de usuários ao longo de oito anos, resulta num excedente do consumidor de US $3,842 bilhões para os cinco países estudados em detalhe (ver figura H) e US $1,315 para o restante dos países latinoamericanos. Este excedente do consumidor contribui por sua vez ao crescimento do PIB na medida em que pode se traduzir num maior 29 consumo. Por exemplo, no caso do México, a Encuesta Nacional de Ingresos y Gastos de los Hogares (ENIGH), realizada pelo Instituto Nacional de Estadística, Geografía e Informática, indica que no ano 2008 a estrutura de gastos médios respondia aos seguintes itens (ver figura I). 94. Por exemplo, qualidade de vida, inclusão social, cidadãos bem informados, e participação de uma comunidade (ver seção 7.7). 95. É importante mencionar que o possível aumento de espaço publicitário está limitado pela despesa total de publicidade que deve ser repartirda entre múltiplos meios, os quais incluem, entre outros, a internet. 96. Uma porção adicional de benefícios é determinada pela aplicação do modelo pré-pago à banda larga móvel, o que tem uma importância chave em facilitar a sua adoção. 97. Ver em particular Greenstein, S. and McDevitt, R. The broadband bonus: accounting for broadband internet's impact on US GDP, NBER Working papers RESUMO EXECUTIVO 37

88 GSMA / AHCIET BENEFÍCIOS ECONÔMICOS DO DIVIDENDO DIGITAL PARA A AMÉRICA LATINA FIGURA H. Contribuição por excedente do consumidor gerado pela banda larga móvel (em milhões de dólares) Argentina Brasil Colômbia México Peru Resto AL América Latina Fonte: análise TAS FIGURA I. Grandes itens do gasto corrente trimestral dos domicílios (2008) Item Porcentagem do gasto Alimentos e bebidas 25,2% Vestido e calçado 3,9% Habitação, energia e combustível 7,5% Artigos para casa 4,5% Cuidados médicos e conservação da saúde 2,3% Transporte 13,8% Educação e lazer 10,2% Cuidado pessoal 7,5% Telecomunicações 1,1% Fonte: Governo do México. Instituto Nacional de Estadística, Geografía e Informática (INEGI) Com base neste esquema de gastos, é possível concluir que uma porção do excedente do consumidor estimado para o México (US $530 milhões) pode resultar em um aumento do consumo em itens como artigos para o lar, educação, lazer e 97 cuidado pessoal. 97. Pesquisas prévias indicam que despesas de alimentação e habitação não aumentam como resultado de economia em outros itens. Ver Martínez, A. Consumption Pattern Development Across Mega-Cities: An Analysis of Sao Paulo and Shanghai. The Lauder Institute. University of Pennsylvania, April RESUMO EXECUTIVO 38

89 GSMA / AHCIET BENEFÍCIOS ECONÔMICOS DO DIVIDENDO DIGITAL PARA A AMÉRICA LATINA Impacto Social da Alocação da Faixa de 700 MHz à Banda Larga Móvel Mais do que os benefícios econômicos quantificáveis, a atribuição do espectro de 700 MHz à banda larga móvel na América Latina poderá ter uma contribuição social positiva em muitas áreas. Por exemplo, ao expandir o acesso a banda larga móvel em zonas sem cobertura, a utilização do espectro permitirá à população residente em áreas hoje não cobertas por banda larga ter acesso a maiores recursos educativos, melhores serviços de saúde e poder receber serviços financeiros. Ao mesmo tempo, a banda larga sem fio passível de ser introduzida em zonas rurais permitirá a provisão eficiente de serviços públicos com maior velocidade de acesso, melhorando a inter-relação entre a sociedade civil e a administração. De acordo com a comparação de ambos cenários de uso do espectro, a banda larga móvel representa uma plataforma mais eficiente para a prestação de serviços que aumentam o bem-estar dos cidadãos (ver figura J). Independentemente dos fatos acima mencionados, é importante destacar que grande parte do impacto social da radiodifusão já está sendo atingido com o lançamento de sinais de televisão digital. FIGURE J. Impacto social de cenários alternativos de uso do espectro Área de impacto Exemplos Radiodifusão Banda larga móvel Educação Conectividade a recursos educacionais Educação à distância Saúde Tele-diagnóstico Comunicação inter-profissional de saúde Informação sanitária Inclusão financeira Aceso a plataformas de micro-pagamentos Educação para aceso a micro-finanças Aceso a serviços públicos Aceso a programas de governo eletrônico Inclusão informativa Aceso a informação do Estado Programas de relacionamento entre cidadãos e governo Fonte: análise TAS Impacto alto Impacto nulo RESUMO EXECUTIVO 39

90 GSMA / AHCIET BENEFÍCIOS ECONÔMICOS DO DIVIDENDO DIGITAL PARA A AMÉRICA LATINA Conclusão Em resumo, o estudo permite concluir que atribuir o espectro de 700 MHz aos serviços de telecomunicações móveis na América Latina gera mais valor econômico e social que se este permanecesse atribuiído à radiodifusão (ver figura K). Estes resultados são consistentes com os gerados por pesquisas realizadas em outras regiões do mundo (ver figura L). FIGURA K. Benefício econômico comparado segundo a utilização da banda de 700 MHz (em milhões de dólares correntes exceto emprego) Radiodifusão Banda larga móvel Contribuição ao ecossistema de TIC (espectro, rede e outros ativos) $3.508 $ x 4,2 Receitas adicionais do setor e contribuição ao crescimento do PIB $513 $3.582 x 7,0 Geração de emprego direto e indireto x 2,1 Impostos (arrecadação marginal adicional nas vendas) $818 $3.420 x 4,2 Excedente do consumidor $0 (*) $5.157 (*) Efeito de segunda ordem que se traduz em mais espaço para propaganda, tendo possíveis efeitos no excedente do produtor e do consumidor Fonte: análise TAS FIGURA L. Benefício relativo se o espectro é atribuído à banda larga móvel América Latina Asia (*) União Europeia (**) Contribuição ao ecossistema de TIC (espectro, rede e outros ativos) x 4,2 N.A. x 2,9 (sem aquisição de espectro) Receitas adicionais do setor e contribuição ao crescimento do PIB x 7,0 x 9,3 x 4,8 Geração de emprego direto e indireto x 2,1 x 22 x 1,3 Impostos (arrecadação marginal adicional nas vendas) x 4,2 x 3,8 N.A. Excedente do consumidor $5,2 B N.A. 70 B (*) Boston Consulting Group. Socio-economic impact of allocating 700 MHz band to mobile in Asia Pacific. Outubro 2010 (**) SCF Associates. The Mobile Provide Economic Impacts of Alternative Uses of the Digital Dividend. Setembro 2007 Nota: Valor para banda larga móvel dividido pelo valor para a radiodifusão. Fonte: análise TAS RESUMO EXECUTIVO 40

91 GSMA / AHCIET BENEFÍCIOS ECONÔMICOS DO DIVIDENDO DIGITAL PARA A AMÉRICA LATINA Deste modo, os resultados do estudo indicam os benefícios a serem gerados pela alocação da faixa de 700 MHz à banda larga móvel nos cinco países estudados em detalhe: Um aumento de cobertura de banda larga devido à disponibilidade de acessos, variável em que existe consenso ser fundamental para apoiar o crescimento econômico na América Latina; Uma implementação e operação de novas redes mais otimizadas resultando em uma economia nos investimentos de US $3,701 bilhões em comparação com a instalação de redes em 98 faixas superiores, alcançando uma melhor cobertura ; Contribuição ao ecossistema de TIC (aquisição de espectro, equipamento e serviços) que excede em mais de US $8, bilhões à contribuição gerada pela radiodifusão ; Contribuição direta (receitas adicionais da indústria) e indireta (externalidades positivas) ao PIB que excede em mais de US $2,474 bilhões à contribuição gerada pela 100 radiodifusão ; Criação de mais de empregos diretos e indiretos 101 adicionais aos gerados pela radiodifusão ; Contribuição tributária adicional superior aos US $2,142 bilhões; Excedente do consumidor adicional superior aos US $3,842 bilhões. Extrapolando os resultados dos cinco países estudados em detalhe ao resto da América Latina, os valores são, como era de se esperar, maiores: Um aumento significativo na cobertura da banda larga móvel usando uma tecnologia mais eficiente, resultando numa economia no desenvolvimento e operação da rede de US $5,440 bilhões, comparado com redes em faixas superiores; Contribuição ao ecossistema de TIC (aquisição de espectro, de equipamento e serviços) que excede em mais de US $11,300 bilhões à contribuição gerada pela radiodifusão; Contribuição direta (receitas adicionais da indústria) e indireta (impactos externos positivos) ao PIB que excede em mais de US $3,069 bilhões à contribuição gerada pela radiodifusão; Criação de mais de empregos diretos e indiretos adicionais aos gerados pela radiodifusão; Contribuição tributária adicional superior aos US $2,602 bilhões; Excedente do consumidor adicional superior aos US $5,157 bilhões. Alcançar um aumento de 31,5% na cobertura da banda larga móvel de nova geração, permitindo assim alcançar quase 92,7% da população da América Latina. Isto permitirá aumentar significativamente a adoção da banda larga e levar para os consumidores finais serviços em velocidades superiores às atuais, o que é um objetivo comum à maioria dos governos da região. Concluindo, a atribuição da faixa de 700 MHz aos serviços móveis, na medida em que permite a estes aumentar a oferta de banda larga móvel e acrescentar a cobertura do serviço, traz efeitos econômicos e sociais substanciais, ao mesmo tempo em que está respondendo às necessidades do mercado. As autoridades públicas latino-americanas não podem ignorar este impacto e devem seguir o exemplo de países como a Colômbia, Peru, Uruguai e México, assim como vários países desenvolvidos, que estão no processo decisório para permitir a concretização deste impacto. 98. Valor nominal de US $6,155 bilhões (investimento: US $3, 690; custos operacionais: US $2,465) em Valor Presente Líquido em oito anos descontado aos 10%. 99. Despesa inicial não recorrente Contribuição direta anual e contribuição indireta acumulada em oito anos Empregos diretos anuais e empregos indiretos ao ano acumulados em oito anos. RESUMO EXECUTIVO 41

92 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA 1. INTRODUCCIÓN El apagón analógico conlleva la liberación de una importante franja de espectro radioeléctrico en la banda de UHF cuya asignación a los servicios de telecomunicaciones móviles de nueva generación podría resultar en el despliegue óptimo de nuevas tecnologías de transmisión y la introducción de servicios avanzados. En particular, el cambio de atribución y uso de la banda de 700 MHz en las Américas conocida como el Dividendo Digital representa una opción clave para la masificación de la banda ancha móvil de cuarta generación en América Latina. El Dividendo Digital se define como el segmento superior de la banda de UHF 700 MHz en el caso de América Latina - actualmente atribuido al servicio de radiodifusión en la mayoría de países, y que, como consecuencia de la transición de la televisión analógica a digital, se libera, pudiendo así ser atribuido al servicio de telecomunicaciones móvil. Esto permite dar una mayor capacidad a la telefonía móvil para responder al crecimiento de tráfico de datos y aumentar la cobertura de la banda ancha móvil. El cambio de atribución y asignación de la banda de 700 MHz al servicio móvil depende de una decisión de política pública. Ésta ya ha sido tomada en un número importante de países desarrollados. Japón, uno de los pioneros en el lanzamiento de servicios avanzados de telecomunicaciones, tomó la decisión de cambiar la atribución del espectro de Dividendo Digital; la 103 reasignación está planeada para 2011, una vez que se haya concluido el apagón analógico. Australia definió recientemente que subastará 700MHz en el segundo semestre de En 2002, Estados Unidos subastó 18 MHz en espectro de entre 710 MHz y 746 MHz, por lo que se recaudó US $88,651 millones. Seis años más tarde, en 2008, Estados Unidos, procedió a subastar 62 MHz del espectro en la banda de 700 MHz, con una recaudación de US $ millones. Las primeras ofertas comerciales en LTE fueron lanzadas en diciembre En Europa se ha avanzado de manera importante en la materia. El Dividendo Digital en Europa consiste de la banda de frecuencia de MHz, también conocida como la banda de 800 MHz. Alemania concluyó el apagón analógico en 2008 y la licitación del Dividendo Digital en mayo 2010, habiendo recaudado millones. Suecia subastó el Dividendo Digital en marzo 2011 y recaudó US $323, millones. Se espera que Noruega, Finlandia, Austria, Dinamarca, Polonia, Portugal, Francia, Suiza, España, Irlanda, Eslovaquia y la República Checa también completen el proceso en Holanda, el Reino Unido, Italia, Bélgica y Croacia han anunciado la reasignación para el En América Latina, Perú y México, que han anunciado que la banda de 700 MHz sería licitada en 2011 y 2012 respectivamente, con anticipación al apagón analógico. Uruguay ha emitido un decreto que libera la banda de 700 MHz para habilitar el despliegue de servicios móviles de voz y transmisión de datos con tecnologías de cuarta generación (4G) en 105 el 2012, y algunas indicaciones de responsables políticos en otros países tales como Colombia, la situación todavía no ha sido claramente resuelta. Por un lado, siguiendo recomendaciones de la Unión Internacional de Telecomunicaciones (UIT), el calendario para la desconexión de televisión analógica está 106 avanzando (ver figura 1-1). Sin embargo, pese a que la mayoría de los países ha completado el proceso de definición de estándar para la televisión digital, las fechas de transición pueden extenderse como 108 mínimo cinco años, pudiendo llegar a diez Fuente: GSMA Se subastaron 3 licencias de de 2x10 MHz, por lo que se recaudó 2,054 millones de Coronas, lo que implica al tipo de cambio del momento US $0,576 MHz/POP Comunicado de Prensa, Ministerio de Industria, Energía y Minería, En 2012 Uruguay liberará espectro en la banda UHF para servicios de transmisión de datos por banda ancha Ver la resolución de COMTELCA (Comisión Técnica Regional de Telecomunicaciones para América Central) publicada en el documento ITU-D RPM-AMS09/28-E que establece como fecha límite el El 2 de septiembre de 2010 fue publicado el DECRETO por el que se establecen las acciones que deberán llevarse a cabo por la Administración Pública Federal para concretar la transición a la Televisión Digital Terrestre, que acelera las fechas originalmente planteadas en México. La transición que empezó en el 2004 deberá concluir las transmisiones de televisión analógica a partir del año 2011 y en su totalidad a más tardar el 31 de diciembre de El decreto está siendo combatido no sólo por empresas interesadas sino también por las cámaras legislativas que han dictaminado distintos puntos de acuerdo cuestionando sus términos En México, esto ocurrirá si el decreto recientemente publicado, es derogado, como todo parece indicar basado en la reciente decisión de la Suprema Corte de Justicia, donde este es declarado ilegal. INTRODUCCIÓN 42

93 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA 1-1. Fechas de decisión de estándar de la TV digital y apagón analógico País Argentina Bolivia Brasil Chile Colombia Costa Rica Ecuador El Salvador México Paraguay Perú Uruguay Venezuela Fecha de desición de estándar 8/2009 5/2010 6/2006 9/2009 8/2008 4/2010 3/2010 4/2009 7/2004 6/2010 2/2009 8/2007 9/2009 Fecha de apagón analógico 2019 Sin definir Entre 2016 y Sin definir 2023 Sin definir /2009 Fuente: AEGIS (27 de mayo de 2010) En este contexto, tres escenarios para el futuro de la atribución de la banda de 700 MHz son factibles. El primero es que el espectro liberado fuese atribuido al servicio de radiodifusión, lo que determinaría que el desarrollo de la banda ancha móvil de cuarta generación en la banda de 700 MHz se vería obstaculizado de manera permanente. El segundo escenario es que la decisión de cambio de atribución sea hecha con posterioridad al apagón analógico. Por ejemplo, en Brasil, la asignación del espectro al servicio móvil está condicionada al apagón de la televisión analógica. Si esto ocurre, se estaría prolongando, como lo probaremos en este estudio, una situación de uso ineficiente de espectro, con el consiguiente costo de oportunidad para la economía y la sociedad (en términos de obstáculos a la introducción de servicios de última generación hasta por diez años) ocasionado por la extensión del plazo en la toma de decisión. El tercer escenario es que se tomen decisiones de cambio de atribución de la banda de 700 MHz, con su consiguiente asignación, sin esperar al apagón analógico. Este escenario es factible, y el más deseable, en la medida de que la televisión en América Latina se emite principalmente en la banda de VHF y en el segmento inferior de la banda de UHF. La banda de 700 MHz muestra una utilización muy baja en varios países, resultando su cambio de atribución y la asignación antes del apagón posible. Por ejemplo, en Argentina, el rango de frecuencia de 512 a 806 MHz está atribuido a la radiodifusión 109 a título primario, y ha sido básicamente subutilizada con licencias precarias, hasta que recientemente se empezara a promover el despliegue de la TV Digital terrestre. En Chile, la utilización de la banda de 700 MHz también es baja (por ejemplo, la Región Metropolitana utiliza sólo 7 de las licencias disponibles ). En México, existen 20 transmisoras (varias en manos de instituciones públicas) en esta banda; la mayoría están en ciudades fronterizas con Estados Unidos. Por otra parte, la industria de servicios de comunicaciones móviles necesita incrementar su acceso a espectro radioeléctrico. Por un lado, las redes de servicios móviles se han transformado en la infraestructura que ha permitido a las telecomunicaciones alcanzar niveles de universalización. Las estadísticas del primer trimestre del 2011 indican niveles de penetración de 136% en Argentina, 109% en Brasil, y 79% en México, alcanzando un promedio continental de 97.8% y niveles de penetración extremadamente altos en la base de la pirámide socio-demográfica (por ejemplo, 50% en Argentina, % en Colombia, y 57% en México). Por otro lado, la banda ancha móvil inalámbrica continúa aumentando su penetración 109. Espectro atribuido a título primario a radiodifusión y con categoría secundaria al Servicio Fijo (Resoluciones SC 2467/98 y 433/98) 110. Fuente: Telecom Argentina. Banda de 700Mhz: Oportunidad para el desarrollo de los servicios de comunicaciones y banda ancha móvil de 4G / LTE en Latinoamérica. Febrero, Fuentes: Wireless Intelligence; UIT; Euromonitor. INTRODUCCIÓN 43

94 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA en la región, demostrando su valor como plataforma sustituta al ADSL y al cable modem, pero en otros casos es la red principal para el acceso a Internet. En Chile, la banda ancha móvil en la segunda mitad el 2009 creció 45%, alcanzando a 25% de todas líneas de banda ancha; en Brasil, 23% de las conexiones 112 de banda ancha (3.5 millones de líneas) son móviles. Considerando la difusión acelerada de smartphones, tablets, netbooks y laptops y, por consiguiente, el crecimiento de usuarios de banda ancha móvil, la necesidad de espectro radioeléctrico en la banda de 700 MHz en América Latina para el despliegue de nuevas redes se torna un problema con implicancias económicas fundamentales. Esto es aun más importante si se reconoce, 113 como lo han planteado numerosos estudios, que la banda de 700 MHz es la más adecuada, por sus características de propagación de señal, para satisfacer las necesidades de zonas suburbanas y rurales. Así, el problema económico incorpora también una dimensión social. Es en este contexto que este estudio visa proporcionar a los tomadores de decisión en los países de América Latina una valoración cualitativa y cuantitativa del beneficio económico y social a generar si el "Dividendo Digital" es asignado al servicio móvil, para su utilización primordialmente para el despliegue de redes de banda ancha móvil. Conceptualmente, se reconoce que la atribución de espectro de 700 MHz a los servicios de radiodifusión también genera beneficios sociales y económicos. Sin embargo, como se demostrará en este estudio así como ha sido hecho en estudios anteriores en otras geografías, la asignación del Dividendo Digital a la banda ancha móvil puede generar un impacto económico mayor. De esta manera, el objetivo primordial de este trabajo es generar una valoración cualitativa y cuantitativa de la diferencia en el análisis relativo de costo-beneficio. Esta valoración se enfoca fundamentalmente en un análisis de impacto socio-económico en tres áreas: Impacto macroeconómico agregado, lo que incluye la contribución al crecimiento del producto interno bruto como resultado del incremento de la productividad, la promoción de inversiones, la creación de valor y la generación de empleo; Impacto social, medido éste en términos de excedente del consumidor (que no es capturado en las variables macroeconómicas) y bienestar económico; Impacto en el sector de telecomunicaciones: ingresos del sector de telecomunicaciones, consiguiente recaudación impositiva, efectos sobre la cadena de valor de la banda ancha móvil y la creación de empleo. Para generar conclusiones suficientemente robustas, el estudio se basa en el análisis detallado de cinco países latinoamericanos cuyas economías representan 82,5% del PIB continental: Argentina, Brasil, Colombia, México y Perú. Una vez generadas las conclusiones del análisis para cada uno de estos países, los 114 resultados fueron extrapolados al resto de América Latina. El estudio comienza en el capítulo 2 pasando revista a los diferentes estudios realizados hasta la fecha con el objetivo de estimar el valor del Dividendo Digital. El propósito es sentar las bases de un marco teórico riguroso, aprovechando las metodologías utilizadas anteriormente, al mismo tiempo que se genera un contexto que permita evaluar los resultados de este estudio en el marco de las conclusiones reportadas en investigaciones anteriores. Habiendo sentado las bases teóricas, presentamos en el capítulo 3 las metodologías utilizadas en este estudio para evaluar el impacto socio-económico del Dividendo Digital en América Latina. Éstas incluyen la definición de los escenarios, el análisis de costo-beneficio de la utilización del espectro para satisfacer las necesidades de tráfico de datos, y el análisis de beneficio económico y social comparado entre un escenario de atribución del espectro a la radiodifusión (para uso de la señal televisiva) y un cambio de atribución a la telefonía móvil. Una vez que el marco metodológico ha sido descrito, comenzamos a adentrarnos en el estudio de la situación latinoamericana. El capítulo 4 presenta la situación de asignación de espectro en la región detallando los planes de transición digital de la señal televisiva, la situación de la banda de 700 MHz en términos de utilización actual y las iniciativas tendientes al cambio de atribución de uso de esta porción de espectro. Después de analizar la situación actual de asignación de la banda de 700 MHz, examinamos en el capítulo 5 las tendencias al crecimiento de tráfico móvil en la región. En este capítulo se presenta una proyección de tráfico de voz y datos al 2020 basada en las tendencias de demanda (crecimiento de la penetración de las telecomunicaciones móviles), sustitución tecnológica (reemplazo de las plataformas 2G por 3G y 4G), 112. Fuente: IDC/Cisco Ver, en particular, Federal Communications Commision. The Broadband Availability Gap. OBI Technical Paper No.1, April 2010, y la Estrategia Nacional de Banda Ancha en Alemania BMWi (2009a). Breitbandstrategie der Bundesregierung, Stand Feburar 2009, Los países considerados para la extrapolación de resultados incluyen Bolivia, Chile, Costa Rica, Ecuador, El Salvador, Guatemala, Honduras, Nicaragua, Panamá, Paraguay, República Dominicana, Uruguay y Venezuela. INTRODUCCIÓN 44

95 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA adopción de nuevos terminales (smartphones, tablets, netbooks y periféricos para la conexión de computadoras portátiles), y nuevas aplicaciones (e.g. video, conexiones máquina a máquina, etc.). En el contexto del análisis cuantitativo de crecimiento de tráfico de voz y datos, el capítulo 6 presenta el análisis de impacto socio-económico en caso de que la telefonía móvil se viese atribuida y asignada la banda de 700 MHz. Éste incluye un análisis de costo-beneficio para las telecomunicaciones móviles en caso de que éstas tuvieran o no acceso a la banda de 700 MHz. El objetivo es examinar cuantitativamente cuál es el valor a ser generado en términos de menor inversión de capital y costo de operaciones si la industria pudiera desplegar una red con base en una señal que tiene mejores condiciones de propagación, requiriendo de esta manera, menos infraestructura que las bandas tradicionales de MHz, 1.700/2.100 MHz, MHz y MHz. Para complementar el análisis de costo-beneficio para las telecomunicaciones móviles, se analizan una multiplicidad de efectos económicos como la creación de empleo, la contribución al crecimiento del producto interno bruto, la recaudación impositiva y el crecimiento del excedente del consumidor. Como alternativa, el capítulo 7 presenta los mismos efectos en caso de que el espectro continuase siendo atribuido a la radiodifusión para su utilización en la expansión de señales de televisión digital. Los capítulos 6 y 7 sientan las bases para la presentación en el capítulo 8 de los efectos comparados para ambos escenarios de asignación: la TV digital o la banda ancha móvil. Para ello, se presentan primero los resultados comparados de ambos escenarios para cada uno de los países de América Latina estudiados en detalle y finalmente se extrapolan los resultados al resto de los países de la región. Finalmente, se presentan las conclusiones del estudio. INTRODUCCIÓN 45

96 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA 2. INVESTIGACIONES REALIZADAS SOBRE EL IMPACTO ECONÓMICO DEL DIVIDENDO DIGITAL En el momento en que la digitalización de la señal televisiva permitió una mejor utilización del espectro radioeléctrico, el análisis de impacto económico, social y tecnológico cobró una importancia fundamental. Numerosos estudios han sido realizados en entornos geográficos diferentes, principalmente en los continentes europeo y asiático, con reportes particulares en países como Francia y Australia. Es importante mencionar, sin embargo, que pese a que las conclusiones de estos estudios tienden a ser esencialmente consistentes, puntualizando el valor económico del cambio de atribución y reasignación del Dividendo Digital a la telefonía móvil, los abordajes y metodologías difieren en aspectos esenciales. El propósito de este capítulo es presentar, de manera resumida, las diferentes metodologías y conclusiones de los principales estudios para crear un referente teórico que permita definir una metodología para el análisis del impacto económico de según la utilización del espectro en la banda de 700 MHz en América Latina. En términos generales, los diferentes estudios realizados a la fecha pueden ser clasificados de acuerdo a cuatro categorías: Estimación del valor económico privado de escenarios de cambio de atribución del espectro: Esta metodología estima el valor agregado a ser generado por escenarios alternativos de asignación de uso del espectro banda ancha móvil o radiodifusión. El abordaje no calcula el valor económico externo o público derivado de la contribución indirecta al crecimiento del PIB o a la creación de empleo. Estimación del impacto socio-económico de escenarios de asignación del espectro: A diferencia del enfoque anterior, este abordaje define dos escenarios alternativos de utilización de la banda de 700 MHz asignación a la radiodifusión o a las telecomunicaciones móviles midiendo el impacto económico comparado de ambos escenarios (por ejemplo, contribución directa e indirecta al producto interno bruto, creación directa e indirecta de empleo, contribución tributaria). Valoración económica de asignación de espectro a las telecomunicaciones móviles: Este enfoque analiza dos escenarios alternativos para satisfacer la demanda creciente de espectro por la telefonía móvil utilización de las bandas previamente asignadas en 800 MHz y MHz, o la asignación de la banda de 700 MHz y mide la diferencia en inversión requerida por ambas alternativas para establecer en una medida única (valor a la sociedad en términos de riqueza creada) para cada uno de los escenarios y proceder a efectuar algunas comparaciones relevantes. Suponiendo mayor complejidad, ciertos estudios definen múltiples escenarios de demanda de servicios audiovisuales y de telecomunicaciones y evalúan el valor económico de diferentes alternativas de asignación de espectro. Simulación del valor social de asignación de espectro a las telecomunicaciones móviles: Este análisis pone el énfasis en la estimación del beneficio a consumidores en términos de valor social. Las siguientes secciones presentan algunos de los estudios y conclusiones realizados en cada una de estas cuatro categorías Estimación del valor económico privado de escenarios de cambio de atribución del espectro El objetivo fundamental de los estudios que siguen esta metodología ha sido determinar políticas públicas óptimas de asignación de una porción del espectro en la banda de 700 MHz a telecomunicaciones móviles. Llevados adelante en momentos donde existía cierta incertidumbre con respecto a la velocidad con la que el tráfico de datos móvil crecería, estos estudios enfatizan la formulación de escenarios de crecimiento del sector. Por otra parte, dado el estado incipiente de la investigación sobre el impacto económico y externalidades de la banda ancha en el momento en que estos estudios fueron realizados, los mismos tienden a estipular el impacto potencial sin cuantificarlo Estudio comisionado por ARCEP (regulador de telecomunicaciones de Francia) En el año 2008, las firmas Analysis Mason y Hogan & Hartson realizaron un estudio evaluando para el caso de Francia dos escenarios de asignación del espectro liberado como consecuencia del "Dividendo Digital". El primer escenario atribuye todo el espectro a la transmisión de señal televisiva, mientras que el segundo escenario establece una compartición de espectro, donde la industria móvil recibe la sub-banda de 790 MHz a 862 MHz y el resto es asignado a la televisión. El análisis INVESTIGACIONES REALIZADAS SOBRE EL IMPACTO ECONÓMICO DEL DIVIDENDO DIGITAL 46

97 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA comparado de ambos escenarios es realizado con base en tres dimensiones: Análisis microeconómico: El escenario de compartición de espectro genera valor agregado superior al escenario de asignación exclusiva de espectro a la televisión, medido en millones entre 2012 y 2024; Análisis macroeconómico: El escenario de compartición de espectro contribuye al crecimiento de PIB francés en millones, comparado con millones del escenario de asignación exclusiva a la televisión. Este análisis no incluye el impacto en la productividad ocasionado como consecuencia de la introducción de servicios de banda ancha móvil en el escenario de compartición; Análisis de externalidades: Este módulo no es evaluado cuantitativamente, pese a que concluye que las externalidades acumuladas son superiores en el caso de la compartición de espectro Estudio comisionado por la Comisión Europea En el año 2009, las firmas Analysis Mason, DotEcon y Hogan & Hartson llevaron adelante un estudio requerido por la Comisión Europea para evaluar los beneficios relativos en la reasignación de espectro liberado por el Dividendo Digital en los países de la Unión Europea. El abordaje de este estudio consistió en la estimación del "valor privado" (beneficio directo recibido por individuos por el consumo de un servicio específico menos el costo de entrega, o en otras palabras, la diferencia entre los excedentes del consumidor y productor) generado de acuerdo a varios escenarios de demanda y oferta de espectro. Los escenarios de oferta de espectro considerados son esencialmente dos: un escenario de compartición temporaria de sub-bandas y otro de asignación única a la banda ancha móvil. Los escenarios de demanda incluyen seis alternativas combinando demanda elevada o baja de servicios audiovisuales y de banda ancha móvil. El análisis es realizado comparando los valores generados por la asignación de las alternativas de asignación de espectro a los diferentes escenarios de crecimiento de la demanda, e identificando cuál es el escenario de oferta óptimo en términos de creación de valor para satisfacer la demanda. Así, considerando un escenario de alta demanda de banda ancha móvil y bajo crecimiento de la demanda de servicios de televisión digital terrestre, la alternativa óptima de creación de valor es el de compartición temporaria de sub-bandas de espectro, generando esta millones en beneficio económico. El estudio concluye que existe un argumento sólido para atribuir y asignar al menos una sub-banda entre 790 MHz y 862 MHz a la banda ancha móvil para satisfacer la demanda de este tipo de servicios. Dado que el estudio se enfoca solamente en el valor "privado", excluyendo así todo valor público, como efectos de derrame en otros sectores de la economía, los autores deben utilizar un multiplicador calculado en un estudio de Ofcom, el regulador inglés, quien estima que el valor público externo es entre 5% y 15% del valor privado. Con base en este factor, los autores concluyen que el valor total del Dividendo Digital para la Unión Europea es entre y millones 115 (VPN descontado a 15 años) Estimación del impacto socio-económico de escenarios alternativos de atribución del espectro Estos estudios se enfocan principalmente en el área dejada parcialmente de lado por los estudios descritos arriba: el impacto de externalidades. Las métricas calculadas con base en benchmarks o modelos cuantitativos se enfocan en el impacto económico externo a ser obtenido como resultado de escenarios alternativos de atribución y asignación de espectro: contribución al crecimiento del PIB, aumento de la productividad y creación de empleo son algunos de los indicadores calculados en estos estudios Estudio comisionado por Deutsche Telekom para la Unión Europea En uno de los primeros estudios realizados para evaluar el Dividendo Digital, Deutsche Telekom comisionó a la firma SCF Associates un estudio enfocado en la Unión Europea. El estudio define dos escenarios de asignación de la banda de 800 MHz, donde la mayor parte de la banda es atribuida a la televisión o a la telefonía móvil. Ambos escenarios presuponen que 15% de la banda es atribuida a otros usos, como aplicaciones militares. El estudio se basa en métricas generadas por investigación previa en áreas tales como impacto de la banda ancha en el PIB, generación de empleo directo e indirecto y excedente del consumidor. Asimismo, modela el impacto de los escenarios en áreas como ingresos y gastos en compra de infraestructura. Los autores concluyen que los efectos económicos directos en la Unión Europea (servicios, ingresos, etc.) para los operadores de telefonía móvil son estimados en millones, comparados con millones para la televisión. Los gastos en el ecosistema de proveedores suman millones en la telefonía móvil y millones para la televisión. El excedente de consumidor y productor es estimado en 165,000millones en la industria móvil y millones en la televisión. Finalmente, el sector móvil generaría 2.3 millones de puestos de trabajo, mientras que la televisión generaría 1.8 millones empleos Valor presente neto, utilizando un flujo de 5 años; no se especifica la tasa de descuento INVESTIGACIONES REALIZADAS SOBRE EL IMPACTO ECONÓMICO DEL DIVIDENDO DIGITAL 47

98 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Estudio comisionado por Telenor, LM Ericsson y la GSMA para Asia En el año 2010, la firma Boston Consulting Group encaró para la región asiática el análisis del impacto socio-económico de la reasignación del espectro de 700 MHz a las telecomunicaciones móviles en comparación con su actual asignación a la radiodifusión. Focalizando el análisis en cuatro países Corea del Sur, Malasia, India e Indonesia y extendiendo las conclusiones al resto del continente, el estudio se enfoca en estimar el aumento en la adopción de banda ancha como resultado de la reatribución de la banda de 700 MHz y la consiguiente contribución al crecimiento del PIB, aumento de la productividad, creación de fuentes de trabajo y otros indicadores macroeconómicos. La premisa de partida de este estudio es la siguiente: la asignación de la banda de 698 a 806 MHz a la telefonía móvil para la provisión de banda ancha reduce los costos de inversión y operación de los proveedores de servicio, quienes trasladan el excedente del productor a los precios. Esta disminución de precios produce un aumento en la penetración de banda ancha, impactando las variables macroeconómicas (productividad, PIB, empleo, recaudación tributaria) y contribuyendo a la creación de beneficios sociales, como mejor acceso a educación y servicios de salud. Suponiendo una disminución de precios de banda ancha móvil de entre 6% y 10% como consecuencia del traspaso de parte 116 del excedente del productor, el estudio estima que la penetración de banda ancha móvil crecerá más allá de las tasas proyectadas en 14% en Corea, 23% en Malasia, 22% en Indonesia y 21% en India, lo que se traduce en una contribución económica significativa. Por ejemplo, utilizando una metodología que desagrega a nivel macroeconómico el impacto de la banda ancha en la productividad de seis sectores económicos, el estudio estima el impacto de la tecnología en la productividad total de los factores y la consiguiente contribución al PIB. En otra área de impacto, utilizando coeficientes de correlación entre la penetración de banda ancha y la creación de nuevas empresas, el estudio calcula la contribución de la penetración incremental de la banda ancha móvil a la creación de nuevas empresas, principalmente en el ecosistema de TIC (ver figura ). La contribución de la radiodifusión al utilizar la banda de 700 MHz incluye las señales adicionales a ser lanzadas con el correspondiente impacto en empleo, ingresos publicitarios, y contribución impositiva. Comparando la contribución al PIB de un escenario de reasignación del Dividendo Digital a la telefonía móvil con su permanencia en la radiodifusión, el estudio concluye que la contribución al PIB agregado de Asia del primero suma US $ millones mientras que el segundo genera US $ millones. Finalmente, pese a no ser estimados cuantitativamente, el estudio menciona los impactos sociales en salud, educación y cohesión social. FIGURA 2-1. Estructura del modelo de impacto económico y social del Dividendo Digital en Asia Establecimientos industriales existentes Productividad Asignación de banda de 700 MHz a telefonía móvil Crecimiento en la penetración de banda ancha Creación de nuevas empresas Rentabilidad promedio Contribución al PIB PIB Operadores de telecomunicaciones Compra de equipamiento Costos adquisición de licencia Salarios de empleados Márgenes Fuente: adaptado del estudio de BCG 116. Los autores del estudio suponen que la construcción de una red en la banda de 700 MHz representa una reducción en costos de infraestructura de 50%, lo que resulta en una reducción de costos totales de entre 6% y 10% que es transferida a los consumidores 117. Tecnologías de la información y comunicación INVESTIGACIONES REALIZADAS SOBRE EL IMPACTO ECONÓMICO DEL DIVIDENDO DIGITAL 48

99 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA 2.3. Valoración económica de atribución de espectro a las telecomunicaciones móviles Estos estudios, aunque abordan temas como beneficios a los consumidores (medidos en excedente del consumidor), el énfasis es puesto en el análisis de costo-beneficio para la industria móvil en comparación con la televisión como resultado de la atribución del Dividendo Digital. El costobeneficio requiere la construcción de modelos detallados de demanda de banda ancha móvil y crecimiento de tráfico sobre los que estructuran redes configuradas de manera alternativa dependiendo del acceso o no a la banda de 700 MHz Estudio comisionado por Ericsson, Nokia, Orange, Telefónica y Vodafone para la Unión Europea Este estudio, realizado en la primera mitad del 2008 por la firma Spectrum Value Partners, analiza el impacto económico en la Unión Europea en caso de que una porción de la banda de UHF fuese asignada a la telefonía móvil. Como el estudio se enfoca en el análisis de costo-beneficio de la asignación de espectro, su punto de partida es la elaboración de escenarios de demanda de tráfico, sobre todo en la transmisión de datos. Basándose en presupuestos de penetración de terminales y utilización, el modelo genera una estimación de capacidad a ser acomodada por la red móvil. Dada la incertidumbre existente en ese momento respecto a la tasa de crecimiento de la demanda de tráfico de datos, el estudio elabora tres escenarios. Sobre la base de estos tres escenarios de crecimiento del tráfico, el estudio estima el impacto económico de dos alternativas tendientes a satisfacer la demanda requerida: con o sin acceso al Dividendo Digital. El análisis de costo-beneficio de ambos escenarios está estructurado alrededor de cuatro dimensiones: Excedente del productor, definido como los beneficios recibidos por la reducción de precios en la producción y pasados al consumidor en términos de reducción de precios; Excedente del consumidor, definido como la diferencia entre la voluntad de pago y los precios practicados; Beneficios indirectos como, por ejemplo, impacto de crecimiento en el ecosistema de TIC; Externalidades, concebidas éstas como crecimiento de empleo, impacto en la productividad, y beneficios sociales; en este caso, el estudio se apoya en las métricas calculadas por Ofcom (15% del valor del productor y consumidor) y por otro estudio (SCF Associates) que calcula beneficios económicos directos e indirectos. El valor del costo-beneficio resulta de la suma de estos efectos en un escenario de utilización de la banda de 800 MHz comparado con la no asignación a la telefonía móvil. El estudio construye numerosas sensibilidades en función de diferentes porciones de espectro a ser asignado. Por ejemplo, en caso de que la telefonía móvil reciba 80 MHz, esto resultaría en un valor para toda Europa de millones en términos de valor presente neto Estudio comisionado por la Asociación Australiana de Operadores Móviles En el año 2009, comisionadas por la asociación de operadores móviles de Australia, las firmas Spectrum Value Partners y Venture Consulting completaron un estudio de evaluación del Dividendo Digital para Australia. La metodología empleada es ampliamente consistente con aquella utilizada por Value Partners para su estudio realizado para la Unión Europea. El estudio concluye que la asignación de una porción de espectro en la banda de UHF (ésta varía por geografía pero en las zonas urbanas llega a 120 MHz) genera un beneficio en el orden de US $7.000 a $ millones, dependiendo del escenario de crecimiento de la demanda. Dadas las condiciones particulares de la geografía australiana, el estudio se enfoca en el beneficio en zonas rurales, donde la banda ancha será esencialmente inalámbrica, con lo que el beneficio económico máximo se obtiene como consecuencia de asignar a la telefonía móvil 140 MHz Simulación del valor social de atribución de espectro a las telecomunicaciones móviles Un ejercicio de simulación realizado por Avanzini y Muñoz (2010) refleja el alto valor social que tendría una asignación de 108 MHz para la banda ancha móvil en América Latina. Para calcular este valor los autores reestimaron el modelo de Hazlett y Muñoz (2009a) y consideraron que el Dividendo Digital se asignará a la provisión de servicios móviles digitales avanzados (AWDS, por sus siglas en inglés). Se supuso que el Dividendo Digital generado en cada país latinoamericano es de 108 MHz, localizados en el segmento de la banda de UHF alto, y que su reasignación, al menos parcialmente, no requiere el apagón analógico. Este estudio supone que el espectro se liberará antes de producirse el apagón analógico, puesto que las bandas que quedan en poder de los operadores de televisión abierta son suficientes para transmitir simultáneamente señales analógicas y digitales. De acuerdo a este supuesto, el valor social bruto es también una referencia para el valor social neto, sin necesidad de ajustar por el costo de los decodificadores INVESTIGACIONES REALIZADAS SOBRE EL IMPACTO ECONÓMICO DEL DIVIDENDO DIGITAL 49

100 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA necesarios para la reasignación de la televisión. Debido a restricciones de la base de datos, sólo se estimó el valor social para nueve países de la región obteniéndose un valor social promedio para el recurso de US $408.3 per cápita si se asignan los 108 MHz de la banda de UHF, siendo Brasil el país que menor bienestar obtendría (US $129,8 per cápita) y Venezuela el que potencialmente podría generar el mayor bienestar en la región (US $1.257,1 per cápita) (ver figura 2-2). Los autores concluyen que este alto valor en Latinoamérica puede deberse a que los operadores móviles se encuentran muy restringidos en el recurso espectro. FIGURA 2-2. Cálculo del valor social de reasignación de la banda de 700 MHz a la telefonía móvil País Población (MM) Var. Excedente del Consumidor (MM US $) Var. Bienestar (MM US $) Var. Bienestar per Capita (US $/capita) Valor MHz US $/MHz-pop Argentina 40, ,6 5,7 Brasil 191, ,8 1,2 Chile 16, ,4 3,8 Colombia 49, ,9 1,3 Ecuador 14, ,2 3,3 México 107, ,8 2,6 Perú 29, ,2 1,9 Uruguay 3, ,1 2,7 Venezuela 28, ,1 11,6 Fuente: Avanzini y Muñoz (2010) 2.5. Conclusión En primer lugar corresponde remarcar la consistencia entre los diferentes estudios respecto al rango de beneficio económico a ser generado si una fracción del espectro de UHF fuese asignado a la industria móvil (ver figura 2-3). El análisis de los estudios producidos hasta la fecha permite observar cierto nivel de consistencia metodológica combinada con una diversidad de enfoques y niveles de énfasis. En términos generales, se observa que aquellos estudios realizados en el año 2008 tienden a desarrollar modelos de proyección de la demanda de tráfico de datos (por ejemplo, el estudio realizado por Spectrum Value Partners para estudiar la situación europea). La necesidad de desarrollar modelos de tráfico se torna menos importante en este momento cuando comienzan a verse señales en el mercado de la tendencia explosiva de aumento de tráfico. Más allá de esta similitud, los niveles de énfasis en el enfoque en los diferentes estudios varían sustancialmente (ver figura 2-4). El presente estudio se diferencia de los anteriores en dos aspectos. En primer lugar, aplica de manera más integral la mayor parte de las metodologías que los trabajos mencionados arriba aplican parcialmente. Como se podrá observar en el siguiente capítulo, este estudio aplica diferentes análisis para evaluar costo-beneficio de escenarios de reasignación, evaluación de impacto socio-económico, cálculo de excedente del consumidor y del productor. En segundo lugar, todas las metodologías y cálculos de evaluación están basados en modelos y premisas desarrollados con base en el continente latinoamericano. INVESTIGACIONES REALIZADAS SOBRE EL IMPACTO ECONÓMICO DEL DIVIDENDO DIGITAL 50

101 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA 2-3. Evaluación comparada de valor en la asignación de espectro Indicador Televisión Telefonía - Banda ancha móvil Razón Unión Europea (Value Partners) VPN a 20 años millones 1,14 millones Asia (BCG) Contribución al PIB US $ millones US $ millones 9,3 Recaudación tributaria US $ millones US $ millones 3,8 Nuevos empleos Unión Europea (SCF) Ingresos generados millones millones 4,8 Gastos en ecosistema millones millones 2,9 Contribución al PIB 0,0% 0,6% Empleos directos ,25 Empleos indirectos 1.8 millones 2.3 millones 1,3 Fuente: compilación de TAS Figura 2-4. Niveles de análisis del Dividendo Digital Francia (ARCEP) Europa (CE) Europa (Operadores y fabricantes) Asia (Ericsson, Telenor, GSMA) Australia 118 (AMTA ) Europa (Deutsche Telekom) Autores Analysis Mason, Hogan & Hartson Analysis Mason, Hogan & Hartson Spectrum Value Partners BCG Venture Consulting, Spectrum Value Partners SCF Associates Modelo de tráfico Sí (escenarios cualitativos) Sí (modelo de tráfico) No Sí (modelo de tráfico) No Costo-beneficio para la industria móvil (con o sin UHF) No Sí (construyendo un modelo de red) No Sí (construyendo un modelo de red) No Excedente del productor Sí Sí No (supone benchmarks de reducción de costos) Sí Excedente del consumidor Sí (calculado sobre la reducción de precios en banda ancha móvil) No (usa disminución de precios para estimar penetración incremental) Sí (supone benchmarks internacionales para voluntad de pago) Sí (usa benchmarks) Externalidades (empleo, crecimiento del PIB, etc.) Sí (usa benchmarks) Sí (usa benchmarks) Sí (desarrolla modelos desagregados) Sí (usa benchmarks) Sí (usa benchmarks) Fuente: análisis TAS 118. Australian Mobile Telecommunication Association INVESTIGACIONES REALIZADAS SOBRE EL IMPACTO ECONÓMICO DEL DIVIDENDO DIGITAL 51

102 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA 3. METODOLOGÍA PARA LA ESTIMACIÓN DEL IMPACTO ECONÓMICO Y SOCIAL DEL DIVIDENDO DIGITAL EN AMÉRICA LATINA Con el objeto de estimar el impacto de asignar parte del espectro de UHF para la prestación de servicios de telecomunicaciones móviles (banda ancha) como consecuencia de la reasignación de espectro, es necesario definir escenarios de comparación, estimar el beneficio total en una serie de medidas (valor a la sociedad en términos de riqueza creada) de cada uno de los escenarios y proceder a efectuar algunas comparaciones relevantes. Es decir, se busca construir herramientas para poder responder a la siguiente pregunta: qué asignación de la banda de 700 MHz crea el mayor beneficio para la sociedad latinoamericana? Para responder a la pregunta anterior para América Latina, y de manera consistente con los estudios sobre el tema en otras regiones en el mundo analizados en el capítulo 2, se construyó una metodología que compara dos escenarios alternativos (ver figura 3-1). Figura 3-1. Esquema metodológico BANDA ANCHA MÓVIL Análisis de la situación actual Estimación de demanda Estimación de oferta Análisis Costo - Beneficio RADIODIFUSIÓN Comparación de banda ancha con radiodifusión Oferta de servicios de radiodifusión Estimación de valor creado 3.1. Análisis de la situación actual El estudio comienza por entender la atribución de la banda de 700 MHz en los países de la región, la disponibilidad de televisión terrestre y los escenarios posibles de reatribución. Las políticas anunciadas, tanto las referentes al apagón analógico como al Dividendo Digital, han variado sustancialmente, tanto en tiempos como en obligaciones y estándares. Anuncios oficiales recientes muestran que la región todavía está en el inicio de pensar de manera integral en el problema y definir una política de transición de la televisión a un mundo digital; la experiencia internacional está aportando ideas, muchas de las cuales están siendo consideradas para su incorporación. METODOLOGÍA PARA LA ESTIMACIÓN DEL IMPACTO ECONÓMICO Y SOCIAL DEL DIVIDENDO DIGITAL EN AMÉRICA LATINA 52

103 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Una vez completada la primera fase del estudio, se encaró la construcción de modelos de proyección de la demanda de tráfico de datos en la región Estimación de la demanda de servicios de banda ancha móvil A diferencia de países más desarrollados, ha sido la telefonía móvil la que ha permitido el acceso a los servicios tradicionales de telecomunicaciones a la mayoría de la población, especialmente a las personas con bajos recursos económicos (en algunas ocasiones referida como base de la pirámide ). Si esto sirve de experiencia, y viendo el crecimiento acelerado de la banda ancha móvil en la región, a pesar de la falta de despliegue de infraestructura, es de esperarse que el desarrollo de la banda ancha en la región responda al mismo patrón y sea el modo preferente de acceso, al menos para una parte importante de la población. La proyección de tráfico comienza por estimar la adopción de los terminales, que serán básicamente de dos tipos (terminal tipo telefónico [smartphones] y tarjeta de datos [por ejemplo, para computadoras, netbooks y tablets]), y el uso, que será medido en términos de datos transferidos por aplicación (internet, correo electrónico y otras comunicaciones personales sencillas, música, video, texto). Parte de esta utilización estará basada en la experiencia internacional, lo observado en la región actualmente y las proyecciones de estudios existentes (ver figura 3-2). Figura 3-2. Metodología para la proyección de tráfico de datos NÚMERO DE ABONADOS/UNIDADES 1. Terminales no 3G GSM idem 1x 2. Terminales 3G - 4G 1xEVDO 1xEVDO WCDMA HSPA 3. Periféricos Tarjetas PC Dongles 4. M2M Smart grids y metering Gestión de distribución Telemática Control remoto / monitoreo Feature phones Smartphones USO/CAPACIDAD MOU/terminal MB/terminal MOU/terminal MB/terminal MB/terminal MB/terminal Para estimar el uso, se supuso una tendencia de crecimiento en el tamaño de la unidad de servicio transmitido (página, video, música, etc.) y la variación que esta evolución puede tener dependiendo del escenario. Asimismo, fue necesario estimar la cantidad de unidades transmitidas por unidad de tiempo (por ejemplo, usuario-mes). Con datos de demanda para los diferentes escenarios, fue posible entonces estimar tanto la cantidad de datos transportados como la capacidad de red necesaria para poder satisfacer esta demanda Estimación del valor del Dividendo Digital desde la óptica del operador Desde el punto de vista de los operadores de redes móviles, el valor del Dividendo Digital para la industria móvil se basa en dos aspectos: menores costos y mayor cobertura. Para su cuantificación, el estudio parte de la premisa de que existirá en el mediano plazo la oferta necesaria para satisfacer la demanda. En este contexto, se calcula el costo de hacerlo bajo METODOLOGÍA PARA LA ESTIMACIÓN DEL IMPACTO ECONÓMICO Y SOCIAL DEL DIVIDENDO DIGITAL EN AMÉRICA LATINA 53

104 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA dos supuestos: con y sin la posibilidad de utilizar el espectro de 700 MHz para hacer el despliegue (ver figura 3-3). Para la estimación de la oferta, fue necesario hacer algunos supuestos sobre las asignaciones de espectro en el corto plazo en las bandas de 1,700/2,100 MHz, 1,900 MHz, 1,900/2,100 MHz y 2,600 MHz, 800 y 900 MHz e inclusive 1,800 MHz. Es importante resaltar que existe una menor armonización regional en la utilización de las diferentes bandas tal y como existe en Europa, por lo que es necesario analizar caso a caso la situación. Se supuso que será necesaria la posibilidad de hacer roaming durante el período de despliegue, por lo que las redes GSM deberán estar funcionando durante el período de estudio. Así como es el caso de las redes 3G, el lanzamiento de redes LTE probablemente sufrirá un rezago importante con respecto al resto del mundo, por lo que es de esperarse que el espectro actualmente utilizado permita satisfacer la demanda por un período mayor que lo esperado en países con sistemas de telecomunicaciones más avanzados. En ese momento, tanto para operadores como para los gobiernos que asignan el espectro, será necesario tomar una decisión de incrementar la oferta a través de la utilización del espectro de 2.6 GHz y de la banda de700 MHz para complementar los servicios. Con esta información y con estos supuestos, fue posible estimar los costos de hacer el nuevo despliegue (el complemento ) en la banda de 700 MHz. Figura 3-3: Estimación del valor del Dividendo Digital para la industria de las telecomunicaciones móviles Valor del beneficio por cobertura adicional Valor del beneficio sin UHF Valor del beneficio con UHF Beneficios Costos Beneficios Costos Statu quo Statu quo Utilización del espectro actual ( , 1.7/2.1, 1.9,2.5/2.6) Utilización del espectro actual ( , 1.7/2.1, 1.9,2.5/2.6) Radiobases para cobertura Diferencia en valor Radiobases para cobertura Fuente: adaptado de Spectrum Value Partners (2010) 3.4. Comparación del impacto económico y social de escenarios de atribución de espectro Para poder comparar alternativas, es necesario realizar un análisis de costo-beneficio evaluando los impactos esperados y el costo de alcanzarlos. Los impactos son generalmente de tres tipos: directos, indirectos y externalidades. Los directos, en el caso del despliegue de más capacidad y utilización de la banda de 700 MHz, están basados en un argumento de reducción de costos que se traducen en una reducción de precios, en uno de mayor velocidad de despliegue, que se traduce en una adopción más acelerada, y en uno de mayor cobertura, que se traduce en mayor posibilidad de adopción. En este caso, se crea tanto excedente del productor como excedente del consumidor". Más allá del análisis de costo-beneficio de utilización del Dividendo Digital para la telefonía móvil, se analiza el impacto económico y social resultante del acceso a este espectro. El análisis de impacto de la banda ancha móvil contempla el valor directo e indirecto a ser creado por el espectro reasignado (ver figura 3-4). METODOLOGÍA PARA LA ESTIMACIÓN DEL IMPACTO ECONÓMICO Y SOCIAL DEL DIVIDENDO DIGITAL EN AMÉRICA LATINA 54

105 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Figura 3-4. Impacto económico y social por reasignación de la banda de 700 MHz a la banda ancha móvil ESTIMACIÓN DE COSTOS CÁLCULO DE BENEFICIOS Estimación de la demanda Abonados por uso Estimación de la red BENEFICIOS DE LA DEMANDA Externalidades (PIB, empleo) Mejora de la productividad Competitividad de los sectores Excedente del consumidor EFECTOS DIRECTOS DE LA OFERTA Contribucion directa a la economia (PIB, empleo directo) Márgenes de operadores Ingresos de suministradores equipamiento, software, contenidos) Costos de despliegue y operaciones de red EFECTOS INDIRECTOS DE LA OFERTA Contribucion fondos publicos (impuestos y tasas) Empleo indirecto e inducido generado Asimismo, se estiman los efectos indirectos y las externalidades, que para los servicios de telecomunicaciones tienden a ser elevados. Las externalidades se observan tanto en generación de empleo en la economía como en productividad y en generación de riqueza (muchas veces cuantificada como crecimiento adicional en puntos porcentuales del PIB). El análisis de impacto económico y social parte de nuestra evaluación en la investigación hecha hasta la fecha en este 119 tema. Por otra parte, el análisis de impacto de la radiodifusión contempla el valor directo e indirecto a ser creado por el uso del espectro asignado. Para estimar la oferta, es necesario entender cuál es la política de migración, de oferta esperada de señales de alta definición (HD) y definición estándar (SD), de apagón digital (fecha), de televisión abierta, de televisión de paga y de multiplexación. El principal supuesto para efectos de comparación es que prácticamente todo el espectro de UHF será utilizado para alguna combinación de servicios digitales de radiodifusión terrestre (DTT), aunque variará de país de país dependiendo de las variables arriba mencionadas. Es importante mencionar que existe un escenario base (la política anunciada), sobre el cual deben estimarse los diferenciales de nuevos despliegues y transmisión, así como la creación de nuevas señales de televisión digital (costos de producción y programación). En general, una oferta mucho más diversa de señales de radiodifusión, sean de paga o libres, tiene externalidades importantes difíciles de estimar, tales como son la diversidad, el acceso a la información, la equidad, la inclusión, la democracia, y en general, una sociedad más libre "La contribución de la banda ancha al desarrollo económico", por Raul L. Katz: estudio completado en octubre de 2010 donde se analiza la contribución de la banda ancha al crecimiento del PIB latinoamericano y, específicamente, el impacto de la banda ancha en la creación de empleo en Chile (estudio a ser publicado en Galperín, H. y Jordan, V. (eds.) (2010). Acelerando la revolución digital: banda ancha en América Latina y el Caribe, Santiago: CEPAL). "The impact of broadband on the economy: research to date and policy issues", por Raul L. Katz: estudio donde se analiza la contribución de la banda ancha al crecimiento del PIB y la creación de empleo en numerosos países emergentes, incluyendo Brasil y Chile (estudio a ser publicado en Trends in Telecommunication Reform, Geneva: Internacional Telecommunication Union). "El impacto de la banda ancha en México", por Raul L. Katz: estudio donde se analiza el impacto de la banda ancha en el crecimiento del PIB y la creación de empleo en México (estudio publicado en el anuario 2010 de la Asociación Mexicana de Internet (AMIPCI)) METODOLOGÍA PARA LA ESTIMACIÓN DEL IMPACTO ECONÓMICO Y SOCIAL DEL DIVIDENDO DIGITAL EN AMÉRICA LATINA 55

106 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Adicionalmente, es importante mencionar dos tipos de efectos que afectan el mercado de televisión abierta. Por un lado, se tiene la competencia de redes alternativas de televisión (principalmente, televisión por cable y televisión por satélite), que compiten por la audiencia y por lo tanto por una fracción de los recursos de publicidad, que son la principal fuente de ingresos de las televisoras. Por otro, se tiene la competencia en el mercado de publicidad por otros medios diferentes a la televisión, tales como la prensa, el internet, e inclusive otros menores (como el mercadeo directo y la publicidad en exteriores). Finalmente, se tiene que monto total dedicado a la televisión abierta es poco probable que aumente sustancialmente con una mayor oferta de canales. En general, las empresas que se anuncian por televisión abierta están buscando alcance y pagan precios directamente relacionados a los puntos de audiencia; una mayor oferta implica que las audiencias serán menores, por lo que los precios deberán sufrir una tendencia a la baja para reflejar las menores audiencias. En otras palabras, al aumentar sólo marginalmente el monto total, nuevas señales de televisión estarán generando ingresos que provendrán principalmente de las señales existentes. Por lo tanto, para estimar el aumento en valor creado principalmente se utilizan dos medidas: cuánto más están dispuestos a pagar los consumidores por una mayor oferta de DTT (en términos de suscripción y equipos) y cuánto más se captura por parte de las televisoras de los recursos de publicidad (a pesar de la tendencia actual de migración a otros medios tales como internet). Figura 3-5. Impacto económico y social por asignación de la banda de 700 MHz a la radiodifusión ESTIMACIÓN DE COSTOS CÁLCULO DE BENEFICIOS Proyeccion de demanda de productos BENEFICIOS DE LA DEMANDA Estimulo de externalidad no existente Excedente del consumidor (dificil de estimar pero se debe mencionar) Abonados X tarifas Estimación de costos de nuevos productos EFECTOS DIRECTOS DE LA OFERTA Contribucion directa a la economia (PIB, empleo) Margenes de operadores (publicidad, TV paga) Ingresos de suministradores (equipamiento, productores de contenido audiovisual) Costos de utilizacion de espectro asignado EFECTOS INDIRECTOS DE LA OFERTA Contribucion fondos publicos (impuestos y tasas) Empleo indirecto e inducido generado Los cálculos anteriores costo-beneficio de la utilización del Dividendo Digital para servicios de banda ancha o para servicios de radiodifusión permiten entonces hacer un análisis de beneficio económico comparado (ver figura 3-6). METODOLOGÍA PARA LA ESTIMACIÓN DEL IMPACTO ECONÓMICO Y SOCIAL DEL DIVIDENDO DIGITAL EN AMÉRICA LATINA 56

107 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Figura 3-6. Comparación de Escenarios Banda Ancha Móvil Radiodifusión Beneficios de la demanda PIB Empleo por externalidades Productividad Excedente del consumidor Competitividad Excedente del consumidor Beneficios directos de la Oferta PIB Empleo directo Margenes de operadores Ingresos de suministradores PIB Empleo directo Margenes de operadores (publicidad, subscripcion, etc.) Ingresos de suministradores Beneficios indirectos de la oferta Contribución fondos públicos Empleos indirectos e inducidos Contribución fondos públicos Empleos indirectos e inducidos Costos de utilización espectro Inversión de capital Opex de red Transmisión Distribución Programación Costo - beneficio 3.5. Extrapolación del estudio detallado de cinco países al resto de América Latina Más allá del estudio detallado de la situación de Argentina, Brasil, Colombia, México y Perú, se abordó la generalización de resultados para el resto de América Latina. Para llevar adelante la extrapolación a nivel regional se utilizaron variables de generalización (en ocasiones referidas como multiplicadores ). Éstas están basadas en población urbana y rural, territorio, desarrollo económico (PIB) y el desarrollo actual de los sectores de telecomunicaciones, según la variable que se extrapola. Esto permite hacer una estimación del impacto socioeconómico a nivel regional de utilizar la banda de 700 MHz para la prestación de servicios de banda ancha móvil. Esta estimación entonces es comparada con un análisis de costobeneficio de utilizar la banda para la prestación de servicios de radiodifusión. La muestra de cinco países analizados en detalle fue escogida para tener suficiente representatividad en la estimación de los beneficios económicos y sociales de la utilización de la banda de 700 MHz y poder extrapolarla al resto de América Latina. Los países escogidos en su conjunto representan el 82,6% del PIB, 75,8% de la población y 79,1% del territorio continental. Para proceder a estimar el impacto a nivel regional, se seleccionaron todas las variables relevantes que fueron utilizadas en la estimación detallada para cada uno de los demás países de 120 América Latina Los países considerados en esta extrapolación son: Bolivia, Chile, Costa Rica, República Dominicana, Ecuador, El Salvador, Guatemala, Honduras, Nicaragua, Panamá, Paraguay, Uruguay y Venezuela. METODOLOGÍA PARA LA ESTIMACIÓN DEL IMPACTO ECONÓMICO Y SOCIAL DEL DIVIDENDO DIGITAL EN AMÉRICA LATINA 57

108 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Para la estimación de la red se buscó la extensión del área urbana y rural así como la cobertura y penetraciones actualmente observadas. Se estimaron las necesidades de red para el área urbana y posteriormente se procedió a calcular las necesidades de red para el área rural utilizando la media ponderada por la variable relevante para cada una de las variables utilizadas. Para estimar la recaudación por la licitación del espectro, se procedió a calcular el descuento a aplicar a los precios de las licitaciones observadas tanto en América Latina (Brasil, Colombia, México, Perú) como en Estados Unidos utilizando como deflactor el PIB en dólares nominales. Para extrapolar el resto del impacto en el ecosistema de la telefonía móvil, se utilizaron como parámetros la cobertura y penetración adicional esperada, la penetración actual, y el PIB per cápita. Para la radiodifusión, se utilizaron los parámetros normalizados por número de señales de televisión totales así como los demás valores en su relación por habitante. Para la generación de empleo directo en el caso de la telefonía móvil, se extrapoló el número de empleados por líneas nuevas como consecuencia de la cobertura adicional; para el empleo indirecto, se estimó la contribución al PIB por cada nuevo empleado generado en la muestra de países. En el caso de la radiodifusión, se estimó la contribución promedio al PIB por cada nuevo empleado en los cinco países y se extrapoló al resto de la región. Finalmente, el excedente del consumidor se estimó como una proporción de la contribución directa al PIB. Para la estimación de los impuestos, se partió de tomar una proporción de los nuevos ingresos generados. No se consideraron los impuestos específicos aplicados en cada uno de los países en el resto de la región. Con la metodología anterior fue posible calcular el valor estimado de la contribución económica y social para el resto de América Latina. Es importante destacar que el valor que estos países representan en el total del beneficio varía de manera importante para cada uno de los conceptos utilizados, ya que los parámetros de extrapolación que fueron utilizados también varían significativamente. Para ilustrar lo anterior, baste mencionar que la muestra de países utilizada representa tres cuartas partes de la población pero más del 82% de la riqueza. Asimismo, aunque la población urbana en la región es de 80%, en la muestra es de 82%, pero en el resto de América Latina es de tan sólo 72%. Esto explica que el número de estaciones radio base en estos países con la utilización de la banda de 700 MHz sea de más del 35% del total. Estas diferencias explican la variación en los valores de los parámetros estimados. METODOLOGÍA PARA LA ESTIMACIÓN DEL IMPACTO ECONÓMICO Y SOCIAL DEL DIVIDENDO DIGITAL EN AMÉRICA LATINA 58

109 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA 4. SITUACIÓN ACTUAL DE LA ASIGNACIÓN DE LA BANDA DE 700 MHz EN AMÉRICA LATINA 4.1. La situación actual de espectro de las telecomunicaciones móviles en América Latina En América Latina, el espectro radioeléctrico para telecomunicaciones ha seguido la tendencia de atribución del resto del mundo, pero se han observado dos diferencias muy claras: su concesionamiento siempre ha sufrido un rezago en términos internacionales y se ha concesionado mucho menos espectro que en países con mayores índices de desarrollo. En 2003, comparado con los países de la Organización para la Cooperación y el Desarrollo Económico (OCDE), a excepción de Paraguay, todos los países de la región tenían significativamente menos espectro asignado a operadores de telecomunicaciones móviles. La situación absoluta ha mejorado sustan- 121 cialmente, habiéndose aumentado el espectro medio concesionado de 104 MHz por país a 195 MHz del 2003 a mayo del Sin embargo, poco espacio se ha ganado en la situación relativa: en 2003 se tenía concesionado tan sólo el 38% del espectro en utilización en otros países, cifra que aumentó a tan sólo 46% ocho años después. Dada la explosión en la penetración de la telefonía móvil que la región ha observado en la última década, habiendo alcanzado casi el 100%, y en vista de que este indicador no es sustancialmente diferente al observado en la OCDE (marginalmente superior al 120%), existen pocos argumentos para defender la no asignación de más espectro para servicios móviles en América Latina. De manera general, este hecho ya ha sido aceptado por las diferentes autoridades en la región y se han tomado acciones para resarcirla. Así, varios países han licitado espectro en el pasado reciente y varios más tienen planeado hacerlo en el futuro cercano. Aunque hubo algunas licitaciones esporádicas en la región en la primera década de este siglo (por ejemplo, Brasil en 2001, México en 2005, Perú en 2007), no fue sino hasta 2010 cuando hubo un claro cambio en la tendencia. Tal como lo 122 señala la GSMA, tan sólo en ese año en la región se asignaron más de 300 MHz. Colombia licitó 50 MHz en la banda de MHz. Brasil licitó 20 MHz en la banda de 1.900/2.100 MHz y asignó bandas adicionales, llamadas sobras de licitaciones anteriores, en las bandas de 850 MHz, 900 MHz y MHz, subastando un total de 105 MHz. México licitó un total de 60 MHz en la banda de 1.700/2.100 MHz y 30 MHz en la banda de 1,900 MHz. Costa Rica dió un paso importante en la liberalización de su sector de telecomunicaciones y asignó 110 MHz en las bandas de 850 MHz, 1,800 MHz y MHz. Perú licitó 24 MHz en la banda de MHz en 2010 (ver figura 4-1). En el 2011 han sido anunciadas varias asignaciones adicionales de espectro en la región, que, aunque es posible que los procesos que las autoridades deben seguir para concesionarlo lleven a retrasos eventuales, se espera que se ponga a disposición de los operadores móviles una cantidad de espectro al menos equivalente a la liberada en 2010, pudiendo alcanzar hasta los 500 MHz en algunos países. México ha anunciado la intención de licitar más espectro (30 MHz) en la banda de 1.700/2.100 MHz, también conocida como la banda de AWS (Advanced Wireless Services) y ha comenzado el proceso de consulta pública para la licitación de la banda de 700 MHz. Argentina estará licitando 37.5 MHz en las bandas de 850 MHz y MHz y también está estudiando la posibilidad de licitar espectro de AWS. Colombia tiene planeado licitar 30 MHz en la banda de MHz, 90 MHz en la banda de MHz y 90 MHz en la banda AWS. Perú está en 123 proceso de licitar frecuencias en la banda de 900 MHz, esta 121. Fuente: Hazlett y Muñoz. (2009) Spectrum allocation in Latin America Fuente: Cabello (2010) El 14 de junio ProInversión lanzó el concurso para adjudicar por concesión única las bandas de MHz y MHz en las provincias de Lima y Callao, y las bandas de MHz y MHz para el resto del país. La concesión se extenderá por 20 años y el ganador podrá utilizar la tecnología que considere más apropiada para explotar esas frecuencias. La convocatoria contempla un plan de cobertura obligatorio además de proveer acceso a Internet gratuito en las escuelas dependientes del Ministerio de Educación, además de otros servicios de telecomunicaciones. El plazo para precalificar para el proceso sobre Nº1- vencerá el próximo 24 de junio. El contrato se publicará el 15 de junio y se recibirán propuestas hasta el 23 de junio. La publicación del contrato final será el 14 de julio. La presentación, recepción y apertura de los sobres Nº 2 (propuesta técnica) y Nº3 (propuesta económica) y adjudicación de la buena pro se concretará el 19 de julio próximo. SITUACIÓN ACTUAL DE LA ASIGNACIÓN DE LA BANDA DE 700 MHz EN AMÉRICA LATINA 59

110 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Figura 4-1. Cantidad de espectro disponible para operadores móviles en 2011 MHz 1T España EUA Noruega Brasila Chile 260 Reina Unido Australia Francia 265 Canadá México Colombia Guatemala 180 Paraguay 179 El Salvador 171 Perú 170 Argentina 170 Uruguay 170 Honduras 120 Nicaragua 110 Bolivia 110 Ecuador Notas: España: 310 MHz en las bandas de 800, 900, 1800 y 2600 MHz a asignarse en 2T2011; Chile 190 MHz en la banda de 2.6 GHz a asignarse en 2S2011; Colombia: 30 MHz en la banda de 1.9 GHz a asignarse en 1S2011 Fuente: GSMA (S. Cabello) buscando asignar 120 MHz en la banda de AWS, y ha anunciado la liberalización del espectro de 700 MHz. Chile pondrá a disposición 190 MHz en la banda de MHz. Brasil también ha anunciado la intención de reestructurar y subastar esta banda. Estas posibles asignaciones en su conjunto disminuirán la brecha de asignación de espectro en América Latina comparada con los países de la OCDE; la región tendrá concesionado el equivalente a aproximadamente dos terceras partes del espectro ya otorgado en países con mayor nivel de desarrollo. Sin embargo, como este documento argumentará más adelante, dada la explosión en la demanda no sólo de los servicios de voz y de datos básicos, sino de los servicios de banda ancha, este espectro no será suficiente para soportarla. Será necesario que los operadores de la región, para prestar un servicio con mayor cobertura y mejor calidad, cuenten con más espectro a su disposición. Aunque ha habido cambios importantes en las reglas que rigen la utilización del espectro, aún se tiene un legado de un sistema de comando y control importante en la región. Muchas de las licencias aún tienen limitaciones importantes en cuanto al tipo de servicios que pueden prestarse e inclusive condiciones contrarias a la neutralidad tecnológica. La mayoría de los países cuenta con topes de espectro, aunque éstos son revisados continuamente. Esta situación genera incertidumbre en asignaciones futuras. Muchos países cuentan con esquemas de cobros recurrentes por la utilización del espectro, que terminan siendo tasas equivalentes a impuestos adicionales a los usuarios y que muchas veces implican que el gobierno participa del riesgo en el negocio de las operadoras. No existe, excepto en contadas excepciones (Guatemala, El Salvador), la existencia o una definición clara de un mercado secundario de espectro, lo que agregaría flexibilidad en el mercado y en las operaciones, permitiendo que un recurso escaso sea canalizado a quien más valor puede generar con él. Y aunque la región ha observado una tendencia a contar con reglas que obligan la compartición de ciertos elementos de la red, no existen todavía reglas para la compartición del espectro. Es en esta nueva onda de liberalización de espectro en América Latina que estas cuestiones deberán ser abordadas, siguiendo, dentro de lo posible, las mejores prácticas internacionales. SITUACIÓN ACTUAL DE LA ASIGNACIÓN DE LA BANDA DE 700 MHz EN AMÉRICA LATINA 60

111 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA 4.2. La situación actual del espectro de 700 MHz en América Latina La utilización de la banda de700 MHz varía significativamente en la región. Por un lado, en Argentina, México y Perú está poco utilizada y por otro lado en Brasil muestra una mayor ocupación. Sin embargo, en Colombia existen varias transmisoras de televisión analógica. En diversas formas y con diferentes cronogramas, parece existir un consenso en la región para proceder al cambio de atribución de uso de la banda a servicios de telecomunicaciones móviles. En la XVI Reunión del Comité Consultivo Permanente II: Radiocomunicaciones Incluyendo Radiodifusión (CCP.II) acontecida en Bogotá (30 noviembre a 3 de diciembre de 2010), se decidió establecer un Grupo ad hoc para plantear el espectro del dividendo digital resultante de la transición a la televisión digital y oportunidades para aplicaciones convergentes para contar con un foro para compartir experiencias y planes para 124 la región. El foro, que tendrá una existencia limitada a dos años, se ha planteado considerar las cuestiones relativas al espectro relacionado con el dividendo digital y las oportunidades convergentes. Entre estas cuestiones se está discutiendo la utilización que esta porción de espectro podría permitir, cómo podría evaluarse el valor del espectro, cuáles podrían ser los mecanismos para facilitar el cambio de atribución de manera eficiente y oportuna, y cuáles son los posibles planes de bandas de frecuencias que se están desarrollando en las regiones 1, 2 y 3 de la UIT. Asimismo, se están discutiendo mecanismos de uso coordinado de las bandas de frecuencias en las áreas de frontera, las oportunidades para la prestación de servicios de seguridad y protección, así como servicios convergentes de radiodifusión y telecomunicaciones. El grupo reconoce que para una eficiente utilización del espectro será necesario armonizar al máximo posible a nivel regional los planes de bandas. Con respecto a este tema, la región 2 de la UIT (Americas) decidirá básicamente entre dos opciones de canalización, la adoptada en Estados Unidos (que incluye frecuencias para la utilización de seguridad pública) que asigna un total de 60 MHz para la banda ancha móvil, o la adoptada por la Telecomunidad Asia Pacífico (APT), que 125 considera tanto arreglos de 2x45MHz FDD (total 90MHz) con un bloque de separación de 10 MHz en el centro. 126 El informe publicado por la APT en septiembre de 2010 propone dos disposiciones de frecuencias para la banda de 700 MHz, considerando tanto aplicaciones FDD como TDD. Para garantizar suficiente protección a los servicios prestados en las bandas adyacentes, el Foro reconoció la necesidad de contar con medidas para mitigar las interferencias. Así, se decidió mantener una banda de guarda de 5 MHz en la parte inferior ( MHz) y una en de 3 MHz en la parte superior ( MHz). El arreglo FDD considera una estructura de 2 x 45 MHz con una banda central de 10 MHz, por lo que la banda disponible para servicios móviles sería de MHz y MHz. Para el arreglo TDD se consideró un ancho de banda de 100 MHz (703 a 803 MHz). 129 Publicada el 10 de agosto de 2007, la canalización adoptada en Estados Unidos consideró como punto de partida los desarrollos en el mercado de servicios de comunicaciones inalámbricas comerciales y las necesidades de servicios avanzados de banda ancha demandados por las entidades de seguridad pública. El plan considera, para usos comerciales, tres bloques de 12 MHz (2 x 6 MHz), un bloque de 22 MHz (2 x 11 MHz), y dos bloques no apareados de 6 MHz; para una red de seguridad pública en la modalidad de alianza público-privada, se asignó un total de 24 MHz (dos bloques de 12 MHz de 763 a 775 MHz y de 793 a 805 MHz). A pesar de que en teoría la canalización adoptada por la APT implica un uso más eficiente de dividir la banda de 700 MHz, como la propuesta es relativamente nueva, todavía no se tiene disponibilidad de equipamiento en el muy corto plazo. Sin embargo, es de esperarse que, dado que la región más poblada del mundo ha armonizado la banda de esta manera, el equipo rápidamente alcance las economías de escala necesarias para poder masificar los servicios; se contarían con 90 MHz disponibles para servicios comerciales y sería posible asignar el espectro a tres operadores diferentes. Por otro lado, la canalización adoptada en Estados Unidos permitiría un despliegue rápido con equipo disponible y que ya ha alcanzado importantes economías de escala, pero el aprovechamiento del espectro para fines comerciales no sería tan eficiente como en la propuesta de la región Argentina La banda de 698 a 806 MHz en Argentina tiene una ocupación parcial y con baja utilización, en gran parte debido a la alta penetración del servicio por cable, y que la televisión analógica se ha desplegado en la banda de VHF y marginalmente en el espectro UHF. (1) Debido a esto, los canales 52 a 69, que 124. Fuente: Informe Final de la XVI Reunión del Comité Consultivo Permanente II: Radiocomunicaciones incluyendo radiodifusión (2010, aprobado en 2011); Dúplex por división de frecuencia, por sus siglas en inglés ( Frequency división duplex ) APT Report on Harmonized Frequency Arrangements or the Band MHz, No. APT/AWF/REP-14, de septiembre de Dúplex por división de frecuencia, por sus siglas en inglés ( Frequency division duplex ) Dúplex por división de tiempo, por sus siglas en inglés ( Time division duplex ) Second Report and Order,FCC , aprobado el 31 de julio de 2007, publicado el 20 de agosto de SITUACIÓN ACTUAL DE LA ASIGNACIÓN DE LA BANDA DE 700 MHz EN AMÉRICA LATINA 61

112 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA conforman la banda de 700 MHz, tienen pocos ocupantes con base en licencias asignadas de manera provisional, por ciudad, por el ente regulador de los servicios de radiodifusión. Este espectro se encuentra atribuido a servicios complementarios de radiodifusión, principalmente para televisión codificada en 131 circuito cerrado. Hasta hace poco no existían licencias nacionales y varios de esos licenciatarios nunca habían efectivamente prestado servicios. Estos servicios de radiodifusión analógica tienen un muy bajo uso efectivo (unos clientes en el país aproximadamente). Dada la baja utilización efectiva y la implementación de la Televisión Digital Terrestre pensada principalmente por debajo del canal 51, se considera factible un escenario de atribución de los canales 52 al 69 al servicio móvil a título primario, y su asignación para servicios de banda ancha móvil previa al apagón analógico, El plan de digitalización de la televisión ya ha sido establecido y está siendo implementado. Al igual que Brasil, en julio de 2009, Argentina decidió adoptar el estándar japonés (ISDB-T). El gobierno argentino ha tomado el liderazgo del despliegue 132 de la Televisión Digital a través de un plan de rápido despliegue de las antenas en las principales localidades del país (50 antenas en la primera etapa, con un plan total de 400 antenas). Adicionalmente el gobierno ya ha comenzado a distribuir decodificadores a ciertos sectores, lo que podrá acelerar el proceso de digitalización. El plan incluye la provisión de TV Digital Satelital en escuelas rurales. Los primeros canales asignados a la Televisión Digital han sido los canales reservados para el Estado, que son los canales de UHF 22 a 25. A los radiodifusores privados también se le han entregado frecuencias por debajo del canal 40. En junio de 211 se determinó que a las provincias se les asignarían canales entre el 20 y el 31 y a las universidades del interior del 133 país entre el 31 y el 34. Sin embargo, a fines de Junio de 2011, la Autoridad Federal de Servicios de Comunicación Audiovisual (AFSCA) hizo una asignación de los canales analógicos del 55 al 66 en la Ciudad y Provincia de Buenos Aires a 15 universidades. Se cree que esta asignación es temporaria, ya que de persistir dificultaría el uso del espectro del Dividendo Digital para la promoción de la Banda Ancha Móvil, tal como está explicitado en los planes de gobierno y la propia Ley de Servicios de Comunicación Audiovisual. En julio de 2011 fue lanzado el concurso público de 220 licencias de televisión digital terrestre locales, entre 4 y 8 por en un total de 32 zonas, en definición estándar. Las señales serán asignadas a organizaciones sin fines de lucro y a empresas con fines comerciales a razón de uno a uno. Todos los licenciatarios contratarán los servicios de uso de infraestructura, multiplexado y trasmisión (denominada la Plataforma Nacional de TDT) de la Empresa Argentina de Soluciones Satelitales Sociedad Anónima (ARSAT). Se han establecido precios de referencia de a pesos (US$ US$ ) como pago de una vez más un pago recurrente de pesos (US$ 5.750) por mes por el servicio de multiplexeo. El Poder Ejecutivo Nacional, a través del Decreto 1552/2010, Plan Nacional de Telecomunicaciones "Argentina Conectada" destaca el espectro radioeléctrico como uno de los ejes centrales del Plan e indica como prioridad la definición de la 135 utilización del dividendo digital. La televisión analógica se emite actualmente en la banda VHF, y en la Ley de Servicios de 136 Comunicación Audiovisual, se indica que luego del apagón analógico, los radiodifusores deberán restituir dichas frecuencias, liberándose espectro adicional. Adicionalmente, la misma ley establece que el espectro debe ser asignado siguiendo las directivas de la UIT, considerando que el Dividendo Digital debe ser utilizado para promover la sociedad de la información. Para esto se requerirá un trabajo conjunto entre la Autoridad Federal de Servicios de Comunicación Audiovisual, la Secretaría de Comunicaciones, que es el órgano de política de telecomunicaciones y la Comisión Nacional de Comunicaciones, el ente regulador. El actual Secretario de Comunicaciones ha expresado públicamente que están trabajando en la Recomendación de la UIT con respecto a la banda de 700 MHz, para el mejor aprovechamiento del Dividendo Digital, 137 situación que están analizando desde junio de Brasil En Brasil, la banda de 698 a 746 MHz (canales 52-59) está atribuida al servicio de radiodifusión a título primario, con una gran ocupación de canales analógicos y digitales. La banda de 746 a 806 MHz (canales 60-69), está atribuida al servicio de repetición de televisión, tiene una ocupación parcial, pero está prevista para canales de televisión pública, lo que podría aumentar significativamente la utilización de esta porción de 138 la banda de 700 MHz por la radiodifusión. Los canales analógicos deberán ser restituidos al finalizar el período de transición que deberá ocurrir en No se ha tomado aún 131. CNC, Cuadro de Atribución de Bandas de Frecuencias, Argentina Resoluciones 687/2011 y 689/2011 de la AFSCA Dependiendo del fin y de la localidad. Fuente: Resoluciones 685/2011, 686/2011 y 812/2011 de AFSCA Esto aplica a toda la banda de UHF. SITUACIÓN ACTUAL DE LA ASIGNACIÓN DE LA BANDA DE 700 MHz EN AMÉRICA LATINA 62

113 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA alguna resolución respecto a la atribución futura de esta parte del espectro. Brasil adopto el estándar japonés (ISDB-T), con algunas variaciones, para la televisión digital terrestre. Se usará tanto la banda de VHF (parte alta) como la banda de UHF, ya que durante el período de transición todos los canales digitales se otorgarán en la banda de UHF debido a la alta utilización de la banda de VHF y para facilitar la recepción de terminales móviles. Se han definido áreas de servicio similares al servicio analógico para el servicio digital. La transición está siendo coordinada por el Foro Brasileño de Televisión Digital (Foro SBTVD) y el gobierno, y ya han sido inaugurados 102 canales digitales en 480 municipios. Existe una corriente de opinión dentro del gobierno que reconoce la necesidad de reasignar el espectro liberado por la transición a la televisión digital a la telefonía móvil, aunque siempre destacando la necesidad de esperar a que termine el período de transición. Por otra parte, la industria de la radiodifusión ha expresado su oposición al cambio de atribución de uso del espectro. Las empresas afiliadas a la ABERT (Asociación Brasileña de Radio y Televisión) han indicado su posición de no aceptar que la banda de 700 MHz sea asignada para servicios de telecomunicaciones. El gobierno ha dejado claro que la decisión de reasignación de la banda de 700 MHz tendrá un componente político importante y no se espera que suceda antes del Colombia La banda de 698 a 806 MHz en Colombia está relativamente ocupada, aunque las señales podrían ser reasignadas fácilmente. De hecho, las últimas resoluciones gubernamentales apuntan en esta dirección. A nivel nacional, existen dos canales privados (TV Caracol y RCN TV) y tres gubernamentales (Canal Uno, Señal Institucional y Señal Colombia) transmitiendo en señal analógica. Estos cinco canales nacionales, siete regionales y 48 estaciones locales están usando la banda de Dividendo Digital, por lo que deberán ser reasignados para poder utilizar la banda para telecomunicaciones móviles. El plan de digitalización de la televisión ya ha sido establecido y está siendo implementado; se estableció el 2019 como la fecha límite para concluirla. Se ha elegido el estándar europeo DVB-T. La Ley de Telecomunicaciones y TIC creó la Agencia Nacional del Espectro (ANE), con el objetivo de gestionar el soporte técnico para la gestión y la planeación del espectro radioeléctrico. Las primeras emisiones comerciales comenzaron en Bogotá en enero 29 de 2010 con la transmisión de las señales de Canal Uno, Señal Colombia y Canal Institucional, originadas en el canal de servicio público RTVC. Durante el 2010, cuatro sitios de transmisión adicionales fueron inaugurados, extendiendo el servicio de televisión digital terrestre a Cali, Medellín, Antioquia, Barranquilla y Santa Marta. Asimismo en diciembre de 2010, RCN y Caracol inauguraron el servicio digital en Bogotá y Medellín. Una resolución gubernamental publicada en noviembre de estipuló el compromiso para liberar la banda de 700 MHz para poder asignarla a servicios de telecomunicaciones móviles. Esta resolución anunció que la banda de 470 a 512 MHz, sería atribuida a la televisión para facilitar la transición. El Ministerio de Tecnologías de la Información y las Comunicaciones (Ministerio TIC) anunció que el Dividendo Digital en la 141 banda de 700 MHz se asignará en el Siguiendo la recomendación de la UIT, el Ministerio TIC ha anunciado que está coordinando la liberación de los canales 52 a 69 ( 700 MHz ) para asignar este espectro a la banda ancha móvil hacia principios de El Ministerio TIC ya efectuó el cambio de la atribución de la banda de MHz al servicio móvil a título primario. Sin embargo, a pesar de que se ha avanzado sustancialmente, el proceso de ejecución de la reasignación puede demorarse más allá de los tiempos especificados México La banda de 698 a 806 MHz en México, utilizada actualmente por servicios de televisión, ya cuenta con la atribución al servicio móvil a título primario en el Cuadro Nacional de Atribución de Frecuencias, y está relativamente poco utilizada, primordialmente en el norte del país y el Distrito Federal. De acuerdo al cuadro de utilización de frecuencias, en muchos casos múltiples frecuencias son utilizadas para transmitir la misma señal desde diferentes localidades dentro de un mismo estado. Al día de hoy, hay 20 transmisoras, la mayoría localizadas en regiones fronterizas. Adicionalmente, hay 8 emisoras 142 planificadas e inconclusas. La transición a la televisión digital comenzó a implementarse en el Después de haber optado por la norma ATSC, México definió un plan de transición a veinte años. La transición fue estructurada en seis fases definidas por geografía, en las que se puede migrar un vez que 20% de cobertura digital y repetición de 90% de la señal analógica ha sido alcanzada. Teóricamente, el plan estipula que se debería estar en la mitad de la fase 3, lo que implica 20% de cobertura en el Distrito Federal, Monterrey, Guadalajara, Tijuana, 140. Resolución 2623 del 8 de octubre de 2009, MINTIC, Colombia Ministerio TIC abrirá proceso de asignación de espectro para servicios de 4G en el cuarto trimestre del 2011, 16 de junio de Fuente: COFETEL, Unidad de Sistemas de Radio y Televisión. SITUACIÓN ACTUAL DE LA ASIGNACIÓN DE LA BANDA DE 700 MHz EN AMÉRICA LATINA 63

114 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Mexicali, Ciudad Juárez, Nuevo Laredo, Matamoros y Reynosa; también se debería tener transmisión comercial en por lo menos dos de estas ciudades y repetición de 90% de la señal analógica en todas. Sin embargo, hasta el momento sólo un número limitado de estaciones ha sido activado de acuerdo a las primeras dos fases. Dada la situación anterior, la Secretaría de Comunicaciones y Transportes (SCT) ha comenzado a buscar el acelerar el plan original de transición. El Presidente Calderón ha indicado que el apagón analógico debería ser concluido en el 2015, seis años antes de lo planeado; esto fue plasmado en un decreto 143 publicado en el Diario Oficial en septiembre El Presidente de la Cofetel, Mony de Swaan, declaró que el gobierno estaría dispuesto a otorgar un subsidio equivalente a 60 dólares para cada hogar para la migración a servicios digitales comenzando en el 2011; la primera partida de este 144 subsidio incluso fue presupuestada. Para coordinar las tareas de migración de la televisión analógica a la digital, fue creada una comisión integrada por las Secretarías de Comunicaciones, Gobernación, Hacienda y Crédito Público, Economía y Educación junto con la Presidencia. De acuerdo a declaraciones de la SCT, no se anticipan mayores dificultades en reasignar los pocos canales que están siendo utilizados. El factor más importante es la creación de impulso que permita superar los numerosos obstáculos burocráticos. Las acciones tomadas en Estados Unidos en la banda de 700 MHz han actuado como mecanismo de estímulo para acelerar el proceso. El decreto, que acelera el plan original, ordenó que se concluyera la liberación de la banda de 700 MHz en 2012, año en el que finaliza la presente administración. Existen planes para lanzar un programa de concesionamiento de licencias del Dividendo Digital para telefonía móvil, siguiendo las recomendaciones de la UIT. Tanto la SCT como la Cofetel consideran que no es necesario esperar el apagón analógico para cambiar a titulo primario la atribución de la banda y proceder a subastar la banda. Ya ha sido concluida la primera consulta pública al respecto Perú En 2005 fue aprobado el nuevo Plan Nacional de Atribución de Frecuencias (PNAF). En 2011 se elaboró un documento de trabajo que presenta alternativas de canalización de las bandas de 700 MHz y 1.700/2.100 MHz para ser próximamente subastadas para servicios inalámbricos avanzados. En dicho documento se reconoce la necesidad de seguir las recomendaciones de la UIT y se establece un plazo de un año para reasignar servicios de radiodifusión que operen en la banda de UHF. El gobierno ya ha cambiado la atribución del Dividendo Digital (desde 2008 ya había sido reservada para telecomunicaciones fijas y móviles la banda de MHz) y en marzo de 2011 lanzó la primera consulta pública sobre las alternativas de canalización de la banda, con la idea de proceder a fines de 2011 o principios de 2012 con la licitación pública, el cual recaería en manos de un nuevo gobierno DECRETO por el que se establecen las acciones que deberán llevarse a cabo por la Administración Pública Federal para concretar la transición a la Televisión Digital Terrestre. 2 de septiembre El presupuesto aprobado fue destinado a estudios de factibilidad. El concepto anunciado fue la utilización de recursos que proviniesen de la licitación de la banda de 700 MHz, con un estimado inicial de US $1.600 millones. SITUACIÓN ACTUAL DE LA ASIGNACIÓN DE LA BANDA DE 700 MHz EN AMÉRICA LATINA 64

115 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA 5. EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA Las comunicaciones móviles han alcanzado niveles de masificación tanto a escala mundial como en el continente latinoamericano. De acuerdo a las estadísticas de la Unión Internacional de Telecomunicaciones (UIT), la penetración de telefonía móvil a escala mundial ha llegado a finales del año 2010 a 76,2% En América Latina, el nivel de adopción era aproximadamente del 95,2%. Este proceso, combinado con la sustitución de terminales de voz por aquellos que permiten acceso a internet, conlleva un aumento acelerado en el volumen de tráfico, tanto de voz como de datos. El siguiente capítulo presenta un análisis del crecimiento de la telefonía móvil en América Latina en los últimos tres años y examina el consiguiente aumento en tráfico. Con base en esta tendencia histórica, y considerando los parámetros de difusión a futuro, se desarrolla una proyección de tráfico para los próximos diez años. El análisis del caso latinoamericano es precedido por una descripción de las grandes tendencias tecnológicas a nivel mundial Crecimiento del tráfico de datos móviles a nivel mundial El proceso de difusión de la telefonía móvil que ha llevado a estos niveles de adopción está conformado por varias tendencias que operan simultáneamente. En primer lugar, corresponde analizar la difusión de la telefonía móvil como tecnología de base. Este proceso representa tanto un proceso de sustitución de la telefonía fija por la tecnología inalámbrica (preponderantemente en países industrializados) como un proceso de "salto hacia delante", donde el proceso de adopción de telecomunicaciones es llevado principalmente por las tecnologías móviles (modelo dominante en países emergentes). Simultáneamente con la difusión de la telefonía móvil, se observa la migración hacia plataformas tecnológicas avanzadas que permiten la utilización más eficiente del espectro radioeléctrico y la introducción de terminales con una funcionalidad cercana a la de una computadora, lo que permite el fácil acceso a internet con base en normas tecnológicas como CDMA 1xEVDO, WCDMA, HSPA y LTE. Estas tecnologías, también llamadas de "banda ancha móvil", combinan tres subtendencias. La primera es la sustitución de terminales de funcionalidad básica (llamados feature phones) por smartphones. Estos últimos, al agregar una capacidad más alta de navegación y acceso a internet, resultan en un aumento acelerado de tráfico. Tal como se mostrará más adelante, el usuario de un smartphone tiende a acceder a internet más frecuentemente, tanto por utilizar más las redes sociales (como Facebook) como por bajar más aplicaciones de las múltiples tiendas de aplicaciones patrocinadas por fabricantes de terminales (Apple App store, Nokia Obi, etc.), operadores de red (AT&T AppCenter, Telcel Ideas APPSTORE, Telefónica mstore, Telecom Italia TIM Store, etc.) y otros (Amazon Appstore, etc.). La segunda subtendencia representa la conexión de computadores personales (laptops, tablets y netbooks) a banda ancha móvil. Ésta es impulsada principalmente por el sector corporativo en los países industrializados, mientras que en los países emergentes incluye también una demanda creciente de usuarios residenciales que no pueden acceder a la banda ancha fija, ya sea por capacidad de pago o por cobertura de la red. Finalmente, la tercera subtendencia es lo que se ha denominado "internet de los objetos", lo que incluye la instalación de terminales inalámbricos en instalaciones de telemetría o de monitoreo de infraestructura, lo que permite la transmisión a centrales de procesamiento. Las tendencias y subtendencias descritas arriba se combinan para generar un aumento acelerado del tráfico de acuerdo al modelo conceptual presentado en la figura 5-1. Como se observa en la figura 5-1, tanto la adopción de smartphones como de terminales para conexiones máquina a máquina no son consecuencia directa de la sustitución de tecnologías 2G por plataformas 3G o 4G. La adopción de smartphones y de sistemas de telemetría basada en telecomunicaciones inalámbricas comienza a desarrollarse bajo estándares 2G. Sin embargo, su crecimiento acelerado se produce a partir de la introducción de normas más apropiadas a la transmisión de datos. EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 65

116 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA 5-1. Modelo conceptual de aumento de tráfico inalámbrico Adopción de smartphones Crecimiento de la penetración de telefonía móvil Substitución de tecnología 2G por 3G y 4G Adopción de periféricos para conexión de PCs Aumento en tráfico de datos Adopción de terminales para conexiones M2M Penetración de la telefonía móvil Tal como se menciona arriba, la telefonía móvil está en camino a alcanzar niveles de universalización. Con una penetración mundial de 76,2% las telecomunicaciones móviles ya han sobrepasado tasas de penetración superiores al 100% no sólo en los países industrializados, sino en numerosos países emergentes (ver figura 5-2). Las proyecciones para los próximos cinco años próximos estiman que la penetración de telefonía móvil seguirá aumentando principalmente en los países emergentes, lo que determinará que a nivel mundial la estadística crecerá de 76,2% al 2010 a 104,6% en Como se puede observar en la figura 5-3, el único continente que se estima que no alcanzará un 100% de penetración de telefonía móvil en 2015 es África, aunque la tendencia de crecimiento muestra una clara dirección hacia la universalización en esta región. Finalmente, como es de esperar habiendo sobrepasado el índice de saturación, la tasa de crecimiento en la difusión de la tecnología se está ralentizando. FIGURA 5-2. Penetración mundial de las telecomunicaciones móviles (finales 2010) Continente África América del Norte América Latina Asia/Pacífico Europa Oriental Europa Occidental Medio Oriente Total Penetración 41,4% 98,1% 97,8% 67,8% 144,4% 123,5% 79,4% 76,2% Fuente: UIT (2011); Euromonitor (2011); Wireless Intelligence (2011) EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 66

117 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA 5-3. Proyección de penetración mundial de telefonía móvil ( ) 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% 1Q10 2Q10 3Q10 4Q10 África Europa Oriental América del Norte 1Q11 2Q11 3Q11 4Q11 1Q12 2Q12 3Q12 4Q12 1Q13 América Latina Europa Occidental 2Q13 3Q13 4Q13 1Q14 2Q14 3Q14 4Q14 1Q15 2Q15 3Q15 4Q15 Asia Pacífico Medio Oriente Fuente: Wireless Intelligence Migración a las plataformas 3G En el contexto de penetración acelerada de telefonía móvil, la necesidad de optimizar la utilización del espectro radioeléctrico ha determinado la migración de redes de estándares tecnológicos de digitalización de segunda generación (GSM, CDMA2000 1X) a tercera generación (WCDMA, HSPA, 1xEVDO). Esta migración facilita las comunicaciones de datos móviles en términos de velocidad de transmisión. Asimismo, los fabricantes de terminales han desarrollado equipamiento para usuarios más apropiado para acceder a internet mediante tecnologías móviles. La asignación de bandas de frecuencia que permiten el despliegue de redes 3G, proceso que se encuentra casi completado a nivel mundial, ha acelerado la sustitución de terminales 2G por 3G. La penetración de terminales 3G es hoy 17,1% proyectándose que ésta alcance 35.7% hacia el 2015 (ver figura 5-4). FIGURA 5-4. Comparación mundial de penetración de terminales 3G Continente Penetración de terminales 3G (1T2011) Penetración de terminales 3G (4T2015) África América del Norte América Latina Asia/Pacífico Europa Oriental Europa Occidental Medio Oriente Total 6,4% 53,7% 10,2% 12,5% 13,8% 40,2% 16,4% 17,1% 28,3% 63,8% 45,9% 29,1% 39,5% 59,2% 29,5% 35,7% Fuente: Wireless Intelligence EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 67

118 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA En paralelo con la migración hacia plataformas 3G y motivados principalmente por la necesidad de optimizar la utilización de espectro, los operadores de telecomunicaciones ya han comenzado a desplegar redes basadas en las normas de digitalización de señales de cuarta generación, principalmente LTE. Se estima que el impacto de esta tendencia hacia el 2015 va a estar principalmente limitado a los países industrializados (ver figura 5-5). FIGURA 5-5. Comparación mundial de penetración de terminales 4G Continente Penetración 4G (2011) Penetración 4G (2015) África América del Norte América Latina Asia/Pacífico Europa Oriental Europa Occidental Medio Oriente Total 0,00% 0,45% 0,00% 0,04% 0,06% 0,14% 0,05% 0,07% 0,5% 13,1% 1,6% 3,3% 2,7% 12,0% 4,1% 4,0% Fuente: Wireless Intelligence (2011) Adopción de smartphones Finalmente, incentivados por el acceso a plataformas de transmisión inalámbrica de datos más avanzadas, los fabricantes de equipos terminales han introducido smartphones y sus correspondientes sistemas operativos más sofisticados en términos de interfaces y gestión de aplicaciones (Blackberry, iphone, Samsung Galaxy, entre otros). La penetración de smartphones está considerablemente avanzada en los países industrializados. Por ejemplo, en Estados Unidos ésta era 20% en el 2010 y está proyectada a llegar a 59,5% en En Europa Occidental, la base instalada de smartphones sumaba ,65 millones, representando 14% de los abonados. La transición a smartphones, concebidos éstos como terminales de preferencia a los abonados de países industrializados, está reflejada en los volúmenes de entregas (ver figura 5-6). Como puede observarse en la figura 5-6, la tendencia a la adopción de smartphones se registra a nivel mundial. El volumen de entregas crece dramáticamente, estimándose que la base en uso alcanzará los 700 millones a nivel mundial para el Aumento en el uso de datos por terminales La adopción de terminales 3G, en particular smartphones, así como la instalación de periféricos de conexión de PCs, determina un aumento en el tráfico de datos en las redes móviles. Las estadísticas en países industrializados, generadas con base en series históricas fiables, indican de manera consistente no sólo el aumento en el tráfico de datos para todo tipo de terminal, sino también los saltos cuantitativos que ocurren cuando los usuarios adoptan unidades más apropiadas para el acceso a internet. A nivel global, se estima que el tráfico de datos generado en por terminales móviles sumó 810 Petabytes. De acuerdo a estas estimaciones, se proyecta que el tráfico promedio por terminal incrementará de 20 MB por mes en el 2010 a 155 MB en el Esto se traduce en un incremento de tráfico de diez veces, lo que representa un volumen de Petabytes. Estas proyecciones parten de que el número de abonados que adquieren planes de precio que incluyen el servicio de transmisión de datos crecerá de 225 millones a finales del 2010 a 470 millones en Noventa y cinco por ciento del tráfico estará generado por navegación de internet y descarga de contenidos de video IDATE. Mobile Year in review Fuente: Strategy Analytics. Handset Data Traffic Model. October Un Petabyte equivale a un millón de Gigabytes EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 68

119 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA 5-6. Volumen de entregas de smartphones 300,000, ,000, ,000, ,000, ,000,000 50,000, América del Norte Asia Pacífico Europa Oriental Europa Occidental América Latina Medio Oriente y África Fuente: IDC. Worldwide quarterly global phone tracker, December 1, En este contexto, el usuario promedio, que genera 15 MB por mes para navegar en internet y 1MB para descargar videos, necesitará 105 MB y 42 MB respectivamente en Estas proyecciones están ampliamente sustanciadas por las estadísticas históricas que muestran la evolución de tráfico promedio entre 2002 y 2011 (ver figura 5-7). FIGURA 5-7. Evolución comparada de tráfico de datos del usuario promedio (medido en MB por mes) Norteamérica 0,0 0,0 0,1 0,2 0,6 2,0 6,1 19,7 49,7 96,6 162,9 América Lat. 0,0 0,0 0,0 0,0 0,1 0,1 0,3 0,5 1,0 2,5 5,8 Asia/Pacífico 0,0 0,4 0,5 0,7 1,1 1,9 4,0 7,0 11,3 18,5 28,0 Japón 0,0 1,8 2,5 3,5 6,4 12,1 32,1 67,6 124,6 223,6 363,0 Corea 0,0 0,2 0,4 1,1 2,5 6,2 15,1 36,8 74,0 136,6 206,4 Europa Or. 0,0 0,0 0,0 0,0 0,1 0,1 0,3 0,6 1,5 3,3 6,8 Europa Occ. 0,0 0,0 0,0 0,1 0,3 0,8 2,0 6,1 16,7 40,0 76,3 M. Oriente 0,0 0,0 0,0 0,0 0,0 0,1 0,2 0,3 0,6 1,3 2,4 Total 0,0 0,2 0,2 0,3 0,6 1,1 2,5 5,5 10,8 19,9 32,5 Fuente: Strategy Analytics (2010) Los datos de la figura 5-7 muestran que el comportamiento en la generación de tráfico de datos está sustancialmente más desarrollado en los países como Japón y Corea del Sur y, en menor medida, América del Norte. Esto se explica en gran parte por el acceso temprano de los usuarios de estos países a smartphones y a las aplicaciones disponibles en las tiendas de aplicaciones patrocinadas por los diferentes fabricantes de terminales, operadores y otras empresas. En particular, el aumento en el tráfico de datos por los usuarios de smartphones se ve confirmado por el estudio publicado en 147 abril 2011 por Wireless Intelligence. De acuerdo con este trabajo, el usuario promedio de un smartphone accede aplicaciones (como redes sociales, juegos, video) generando 667 minutos de uso por mes, casi tanto como mensajería (671 minutos) y mucho más que las comunicaciones de voz (531 minutos). Con base en este comportamiento de uso, el 147. Wireless Intelligence. Smartphone users spending more 'face time' on apps than voice calls or web browsing. Abril EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 69

120 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA abonado promedio que posee un iphone genera 422 MB de tráfico por mes, mientras que el usuario de un terminal Android consume 133 MB. El rango de diferentes comportamientos de los usuarios de terminales Android es importante: 24% de los abonados consumen más de 500 MB por mes. * * * * En resumen, las tendencias de la telefonía móvil a nivel mundial indican no sólo una evolución hacia la universalización de terminales, llegando a una penetración de casi 105% en el 2015, sino una transición fundamental a terminales que facilitan la transmisión de datos y acceso a internet (llegando a 35,7% de los abonados a nivel mundial), y el consiguiente aumento dramático del tráfico de datos, alcanzando en cinco años un volumen diez veces mayor que el actual Tendencias de la adopción de terminales generadores de tráfico de datos en América Latina La siguiente sección detalla cómo las tendencias de aumento de tráfico de telecomunicaciones móviles, especialmente datos, se observan en el continente latinoamericano. Cada tendencia y subtendencia es descrita en términos de la región en su conjunto, luego presentándose detalles estadísticos para cada uno de los cinco países estudiados Penetración de la telefonía móvil Como se menciona arriba, la telefonía móvil ha alcanzado niveles masivos de penetración en el continente latinoamericano. El promedio continental de 97,8% al primer trimestre de 2011 representa una adopción relativamente similar a la observada en países industrializados (ver figura 5-8). Estos niveles de adopción se han producido en los últimos diez años y no muestran todavía señales de ralentización. Por ejemplo, Brasil incorporó 7,57 millones de líneas por mes en el 148 primer trimestre del Incluso Argentina, que aparentemente está en niveles de saturación, agregó más de abonados móviles por trimestre durante Corresponde mencionar también que, dados los altos niveles de adopción agregada de telefonía móvil, la penetración de la tecnología en los estratos socio-demográficos más desfavorecidos de la población latinoamericana también está alcanzando niveles altos (ver figura 5-9). FIGURA 5-8. Penetración comparada de telefonía móvil (1T2011) 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% 103% Estados Unidos 113% 93% 103% Australia Japón Corea 136% Argentina 59% Bolivia Brasil Chile Colombia 155% 120% 112% 118% 109% 106% 90% 68% 79% 79% Cosa Rica Ecuador México Panamá Perú Uruguay Venezuela Fuente: UIT; Wireless Intelligence; análisis TAS 148. Fuente: ANATEL (Agencia Nacional de Telecomunicaciones de Brasil) Fuente: Wireless Intelligence. EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 70

121 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA 5-9. Penetración comparada de telefonía móvil en la base de la pirámide socio-demográfica (2 deciles inferiores) (2009) 70% 60% 63% 60% 57% 56% 55% 50% 50% 48% 40% 30% 33% 20% 10% 22% 21% 19% 13% 10% 0% Colombia Chile México Costa Rica Uruguay Argentina Ecuador Brasil Perú R.Dominicana Venezuela Guatemala Bolivia Fuente: UIT; Wireless Intelligence; análisis TAS Como puede observarse, la penetración de las telecomunicaciones móviles en los dos deciles más bajos de la población en algunos de los países de la región ya excede el 50%. En otras palabras, la masificación de las telecomunicaciones móviles en la región ya es una realidad. Basados en la situación actual, la tendencia histórica, y una estimación de niveles de saturación esperados, proyectamos la evolución de la penetración hacia el 2020 (ver figura 5-10). FIGURA América Latina: Penetración proyectada de telecomunicaciones móviles País Penetración 2010 Penetración 2015 Penetración 2020 Argentina 141,7% 149,7% 157,0% Bolivia 74,7% 96,5% 99,0% Brasil 96,0% 140,6% 149,0% Chile 101,4% 165,4% 180,0% Colombia 96,6% 100,5% 104,0% Costa Rica 48,3% 66,4% 76,9% Ecuador 104,8% 116,5% 120,7% El Salvador 133,4% 128,1% 153,3% Guatemala 88,3% 103,1% 163,6% Honduras 119,0% 103,0% 140,1% México 81,1% 95,2% 110,0% Nicaragua 62,2% 73,4% 89,06% Panamá 134,7% 156,0% 211,1% Paraguay 93,1% 109,2% 110,3% Perú 88,4% 105,1% 112,0% R. Dominicana 91,6% 106,6% 109,1% Uruguay 126,7% 138,3% 142,6% Venezuela 101,2% 104,5% 111,9% Promedio 95,2% 116,8% 130,0% Fuente: UIT; Wireless Intelligence; Euromonitor; análisis TAS EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 71

122 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Tal como la figura lo indica, nuestros análisis muestran que la penetración esperada hacia finales de la década alcanzará alrededor de 130% A continuación, se presentan proyecciones específicas para los cinco países estudiados en detalle (ver figura 5-11). Las tasas anuales de crecimiento esperadas para cada uno de los países están proyectadas a estabilizarse hacia el 2015 entre 1% y 2%, lo que llevará a penetraciones de entre 104% (Colombia) y 157% (Argentina) (ver figuras 5-11 y 5-12). FIGURA Tasas de crecimiento de las telecomunicaciones móviles y penetraciones esperadas Tasa de crecimiento a partir de 2015 Penetración esperada en 2020 Base instalada en 2020 (en millones) (*) Argentina 1,0% 157% 69,82 Brasil 2,5% 149% 312,28 Colombia 1,5% 104% 54,65 México 2,0% 110% 130,65 Perú 1,5% 112% 37,67 (*) Abonados con terminales 2G, 3G feature phones, 3G Smartphones, y 4G Fuente: análisis TAS FIGURA Tasas de crecimiento de las telecomunicaciones móviles y penetraciones esperadas 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% 120% 100% 80% 60% 40% 20% 0% % 100% ARGENTINA 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% -10% COLOMBIA 120% 100% 80% 60% 40% 20% 0% -20% 160% 140% 120% 100% 80% 60% 40% 20% 0% 120% 100% 80% 60% 40% 20% 0% PERÚ 70% 60% BRASIL 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% MÉXICO 60% 50% 40% 30% 20% 10% 0% Penetración Tasa de Crecimiento 80% 60% 40% 20% 0% EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 50% 40% 30% 20% 10% 0% Fuente: Wireless Intelligence; análisis TAS 72

123 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Migración a las plataformas 3G De manera simultánea con la adopción acelerada de la telefonía móvil, los operadores de América Latina están migrando sus redes de tecnologías 2G a 3G. Hacia el 2015, se observarán también una migración hacia plataformas 4G, con base principalmente en el estándar LTE. En la actualidad, esta tecnología está siendo evaluada en pruebas piloto en Argentina, Brasil, Chile y Colombia. La migración a tecnologías 3G es importante en la medida que los terminales que operan en estas normas (por ejemplo, HSPA) son más adecuados para proveer un acceso de banda ancha eficiente a internet. Esta tecnología representa una respuesta extremadamente positiva adecuada a las necesidades de un mercado condicionado por los costos de adquisición de computadoras y los límites en el despliegue de banda ancha fija. La satisfacción de una necesidad del mercado, combinada con una utilización más eficiente del espectro radioeléctrico, determina que la transición de la masa de abonados en la región a plataformas 3G se completará en el curso de la década. Las proyecciones de la tasa de sustitución de terminales muestran que hacia el 2015, 46.2% de los abonados de la región estarán utilizando terminales 3G y 4G (ver figura 5-13). FIGURA América Latina: Penetración comparativa de terminales 3G y 4G País Penetración 3G (4T2011) Penetración 3G y 4G (4T2015) Argentina 15,1% 51,6% Bolivia 4,2% 29,6% Brasil 15,1% 48,8% Chile 9,0% 38,6% Colombia 8,9% 50,5% Costa Rica 15,9% 48,2% Ecuador 12,4% 53,4% El Salvador 6,7% 33,1% Guatemala 7,3% 42,6% Honduras 7,5% 40,6% México 11,0% 45,1% Nicaragua 9,2% 53,3% Panamá 4,3% 21,2% Paraguay 8,0% 51,5% Perú 18,3% 57,2% R. Dominicana 4,0% 18,1% Uruguay 7,0% 49,7% Venezuela 14,% 49,8% Total 12,% 46,2% Nota: Estas cifras se refieren a conexiones y no terminales como son reportadas por algunas autoridades regulatorias. Fuente: Wireless Intelligence; análisis TAS EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 73

124 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA En algunos países como Argentina, Brasil, Chile, Colombia, Perú y Venezuela, el porcentaje de terminales LTE ya estará comenzando a crecer. A continuación, se presentan proyecciones específicas para los cinco países estudiados en detalle (figura 5-14). En Argentina, el aumento dramático de abonados con terminales de tecnología HSPA y el ulterior despliegue LTE determina que hacia el 2020, 87% de la base será de tercera 150 o cuarta generación. Esa misma tendencia lleva a estimar que en Brasil y en México, la base de este tipo de terminales será de 76% y 73%, respectivamente. En Perú y Colombia, que comenzaron a crecer aceleradamente en un tiempo más tardío que los otros países latinoamericanos, las tecnologías 3G y 4G van a asumir un papel protagónico en el crecimiento del mercado después del Se estima que en estos dos países el 98% y el 93,5%, respectivamente, de los abonados estarán en posesión de terminales de estas tecnologías. FIGURA 5-14: Evolución tecnológica de la base instalada ( ) Argentina Brasil Colombia México Perú Fuente: Wireless Intelligence; análisis TAS Abonados no 3G Abonados 3G y 4G 150. Calculados de acuerdo a la siguiente formula: Crecimiento de HSPA & LTE( ) = Crecimiento Mercado( )-(1x+GSM+WCDMA) EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 74

125 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Adopción de smartphones La migración hacia terminales 3G incluye una subtendencia importante: la adopción de smartphones. La funcionalidad de estos terminales es más avanzada que la de los teléfonos básicos, en la medida que provee interfaces y formatos de pantalla más adecuados para acceder a internet. La adopción de smartphones representa una tendencia fundamental a ser estudiada ya que la conveniencia de estos terminales para el acceso a internet determina que los usuarios de smartphones tienden a utilizar la línea móvil de manera más intensa. Por ejemplo, en América Latina, un usuario típico de un teléfono básico tiende a generar tráfico de datos equivalente a 5,8 MB por mes, mientras que un usuario de un smartphone genera un promedio de 148 MB por mes. La figura 5-15 muestra la generación de tráfico para un usuario promedio en los cinco países considerados por el estudio. Comparativamente, se estima que un usuario de un smartphone en la región genera 25 veces más tráfico, mientras que una PC móvil o modem UBS genera 100 veces más tráfico por mes. El hecho de que América Latina todavía registre tráfico de datos considerablemente más bajo que otras regiones del mundo se debe a que la penetración de smartphones todavía es baja. De acuerdo a información de la industria, la base instalada en los países del estudio no supera los 21 millones (ver figura 5-16). Sin embargo, la adopción de smartphones está creciendo de manera acelerada. Esto puede deducirse de las cifras de embarques compiladas por la firma IDC, quien proyecta volúmenes de 33 millones para el año 2011, llegando a 80,6 millones en tres años (ver figura 5-17). FIGURA América Latina: Cuadro comparativo de tráfico de datos de abonados promedio (MB/mes) ( ) Argentina 0,4 1,0 2,7 6,2 12,2 21,6 35,8 56, Brasil 0,5 1,0 2,8 6,3 12,3 21,7 35,8 56, Colombia 0,4 0,8 2,0 4,7 9,5 18,2 31,9 53, México 0,7 1,4 3,2 6,9 12,9 22,5 37,2 59, Perú 0,4 0,8 2,0 4,7 9,5 18,2 31,9 53,2 89, Resto de Am. Latina 0,5 1,0 2,5 5,8 10,4 18,9 32,0 51,9 90,6 146,8 237, Fuente: Strategy Analytics; análisis TAS FIGURA América Latina: Base instalada estimada de smartphones (2011) (en miles) Base instalada Como porcentaje de abonados Como porcentaje de abonados 3G Argentina ,9% 26,2% Brasil ,9% 26,2% Colombia ,1% 45,9% México ,7% 44,8% Perú 966 3,7% 20,3% Total ,5% 31,0% Fuente: Ovum; IDC; análisis TAS EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 75

126 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA América Latina: Embarques de smartphones (en miles) Argentina Brasil México ,743 8, Resto de América Latina Total Fuente: IDC. Worldwide quarterly global phone tracker, December 1, 2010 Utilizando los volúmenes de entregas como punto de partida, y basándonos en la tasa de reemplazo de terminales, se ha proyectado la penetración de smartphones en los países considerados. De acuerdo a las estimaciones presentadas en la figura 5-18, la base instalada de smartphones representará 22% de los abonados móviles en los cinco países estudiados en el Con base en nuestras estimaciones, este porcentaje alcanzará el 56% en el FIGURA América Latina: Base instalada de smartphones (en miles) Argentina Brasil Colombia México Perú Total Fuente: análisis TAS Difusión de periféricos para la conexión inalámbrica tarjetas o dongles en América Latina es cercano al millón (ver figura 5-19) El acceso a banda ancha móvil para computadoras (principalmente PCs, laptops, tablets y netbooks) en América Latina todavía es un fenómeno embrionario. De acuerdo a las cifras compiladas por Ovum, el número de conexiones por Esto significa que de la base instalada de PCs en los cinco países estudiados, 0,8% tienen un periférico (tarjeta o dongle) para conectarse a la banda ancha (ver figura 5-20) En esta sección no se consideran las computadoras conectadas a redes WiFi; sólo se consideran aquellas conectadas a la red móvil Esto no incluye dispositivos con SIM embebidos como tablets y netbooks, los se espera que aumenten significativamente a futuro. EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 76

127 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA Base instalada estimada de periféricos para conexiones a banda ancha (en miles de unidades) País Argentina Bolivia Brasil Chile Colombia Costa Rica Ecuador El Salvador Guatemala Honduras México Nicaragua Panamá Paraguay Perú R. Dominicana Uruguay Venezuela Promedio Fuente: Ovum; análisis TAS FIGURA Porcentaje de PC conectadas a banda ancha inalámbrica (2010) Base instalada PC (miles) Base instalada de PC conectadas a BAM (miles) Porcentaje de PC conectadas a BAM Argentina ,1% Brasil ,7% Colombia ,9% México ,4% Perú ,1% Total ,8% Fuente: Euromonitor; Ovum; análisis TAS En el 2011, este porcentaje ha ascendido a 1,5%. De manera similar a la tendencia de los smartphones, el número de PCs conectadas por tecnología móvil también está en franco crecimiento proyectándose un incremento de 15%. De acuerdo a nuestras estimaciones, el porcentaje de PCs y laptops conectadas a la red móvil en los cinco países llegará a 4% en 2014 y 15% en EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 77

128 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA América Latina: Base instalada de periféricos para conexión a banda ancha móvil (en miles) Argentina Brasil , Colombia México Perú Total Fuente: análisis TAS Difusión de conexiones máquina a máquina La banda ancha móvil va a ser también utilizada para conectar terminales y equipamiento en la provisión de servicios de telemetría, monitoreo y servicios financieros. Como parte de nuestras proyecciones de tráfico de banda ancha, se han considerado las siguientes conexiones máquina a máquina: Telemática OEM Telemática servicio post-venta Telemática comercial AMI (Advanced Metering Infrastructure) Seguridad RMAC (Remote monitoring, automation, and control) Cajeros/POS Máquinas de venta RID (Remote info display) Tele-salud Otros Con base en estadísticas de la firma ABI Research, se estima que en América Latina existen 9,12 millones de conexiones máquina a máquina que utilizan la red móvil. La mayoría de las conexiones lo hacen a través de tecnologías 2G (7,7 millones) y 2.5G (1,4 millones). Las aplicaciones más comunes son las telemáticas. Se estima que a nivel continental, las conexiones máquina a máquina crecerán 28% llegando a 108 millones en el 2020 (ver figura 5-22) Evolución de la base instalada total de terminales generadores de tráfico de datos Sumando las proyecciones de la base instalada de terminales en América Latina, se estima que, en los cinco países considerados en este estudio, el número de terminales e instalaciones conectados a la red móvil evolucionará de 420,4 millones en el 2010 a 712,4 millones en 2020 (ver figura 5-24). FIGURA América Latina: Proyección de conexiones máquina a máquina por aplicación (en millones) Telemática OEM 1,27 2,55 5,00 8,36 11,58 14,89 19,16 24,65 31,72 40,81 52,50 Telemática postventa 2,06 2,56 3,07 3,58 4,04 4,45 4,89 5,38 5,92 6,51 7,16 Telemática comercial 1,82 1,91 2,03 2,21 2,45 2,74 3,07 3,44 3,84 4,30 4,81 AMI 0,52 0,70 0,89 1,14 1,40 1,91 2,60 3,53 4,81 6,54 8,89 Seguridad 0,95 1,10 1,32 1,72 2,26 3,02 4,04 5,41 7,23 9,68 12,94 RMAC 1,35 1,57 1,82 2,16 2,59 3,14 3,81 4,61 5,59 6,77 8,20 Cajeros/POS 0,88 1,05 1,21 1,36 1,56 1,82 2,13 2,49 2,91 3,40 3,97 Máquinas de venta 0,09 0,10 0,11 0,13 0,16 0,19 0,23 0,28 0,33 0,40 0,48 RID 0,00 0,00 0,01 0,01 0,02 0,04 0,08 0,15 0,28 0,53 1,00 Tele-salud 0,05 0,09 0,15 0,24 0,39 0,62 1,00 1,59 2,53 4,04 6,45 Otros 0,12 0,19 0,28 0,37 0,47 0,59 0,74 0,92 1,14 1,43 1,78 Total 9,12 11,81 15,88 21,29 26,93 33,43 41,74 52,44 66,30 84,40 108,19 Fuente: ABI Research; análisis TAS EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 78

129 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA América Latina: proyección de conexiones máquina a máquina por aplicación (en miles) Argentina ,064 1,400 1,763 2,188 2,732 3,432 4,339 5,524 7,081 Brasil 3,705 4,837 6,481 8,653 10,928 13,563 16,935 21,277 26,902 34,243 43,897 Colombia ,264 1,621 2,012 2,512 3,156 3,990 5,079 6,511 México 2,047 2,695 3,697 5,043 6,456 8,013 10,005 12,570 15,894 20,230 25,934 Perú ,034 1,291 1,622 2,051 2,611 3,347 Total 7,215 9,380 12,661 17,009 21,601 26,810 33,476 42,057 53,177 67,686 86,770 Fuente: ABI Research; análisis TAS FIGURA Argentina, Brasil, Colombia, México, Perú: Número de terminales e instalaciones conectados a la red móvil (en millones) Terminales 2G 373,0 394,2 389,8 365,2 321,8 275,1 229,5 195,0 165,1 138,4 114,7 Terminales 3G * 31,2 46,6 63,2 85,9 115,0 143,2 173,8 205,6 224,0 245,0 266,3 Smartphones 8,3 21,0 41,9 67,8 98,4 128,0 157,4 177,8 197,4 215,1 230,2 Tarjetas 0,7 1,3 2,3 3,5 4,7 6,9 9,0 11,1 12,6 13,4 14,4 M2M 7,2 9,4 12,7 17,0 21,6 26,8 33,5 42,1 53,2 67,7 86,8 Total 420,4 472,5 509,9 539,4 561,5 580,0 603,2 631,6 652,3 679,6 712,4 Fuente: análisis TAS (*) Excluyendo smartphones En Argentina, el número de terminales e instalaciones conectados a la red móvil evolucionará de 53,5 millones en el 2010 a 75,4 millones en 2020 (ver figura 5-25). En Brasil el número de terminales e instalaciones conectados a la red móvil evolucionará de 202,9 millones en el 2010 a 372,8 millones en 2020 (ver figura 5-26). FIGURA 5-25: Argentina: Número de terminales e instalaciones conectados a la red móvil (en millones) Terminales 2G 47,6 48,0 45,9 41, ,7 25,3 21,5 18,3 15,7 13,5 Terminales 3G * 4,3 6,2 8,3 10,7 13,2 13,2 15,4 19,1 20,0 21,7 22,7 Smartphones 0,9 2,2 4,7 8,4 12,9 16,7 20,6 23,3 25,8 28,2 30,1 Tarjetas 0,1 0,2 0,3 0,5 0,6 1,0 1,3 1,6 1,8 1,9 2,0 M2M 0,6 0,8 1,1 1,4 1,8 2,2 2,7 3,4 4,3 5,5 7,1 Total 53,5 57,4 60,3 62,2 63,7 62,8 65,3 68,9 70,2 73,0 75,4 (*) Excluyendo smartphones EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 79

130 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA 5-26: Brasil: Número de terminales e instalaciones conectados a la red móvil (en millones) Terminales 2G 178,6 199,4 199,8 188,9 167,9 146,0 126,9 109,6 94,9 82,1 71,1 Terminales 3G * 16,4 26,2 35,4 48,3 63,2 77,3 89,1 107,0 118,9 129,8 139,7 Smartphones 3,9 9,3 19,8 31,9 47,5 61,7 76,0 85,8 95,3 103,8 111,1 Tarjetas 0,3 0,6 1,2 1,7 2,3 3,4 4,4 5,4 6,2 6,6 7,0 M2M 3,7 4,8 6,5 8,6 10,9 13,6 16,9 21,3 26,9 34,2 43,9 Total 202,9 240,3 262,7 279,4 291,8 302,0 313,3 329,1 342,2 356,5 372,8 Fuentes: Anatel; Wireless Intelligence; análisis TAS (*) Excluyendo smartphones El número de terminales e instalaciones conectados a la red móvil en Colombia evolucionará de 43,9 millones en el 2010 a 62,9 millones en 2020 (ver figura 5-27). En México, el número de terminales e instalaciones conectados a la red móvil evolucionará de 91.4 millones en el 2010 a millones en 2020 (ver figura 5-28). En Perú, el número de terminales e instalaciones conectados a la red móvil evolucionará de 23,5 millones en el 2010 a 41,9 millones en 2020 (ver figura 5-29). FIGURA Colombia: Número de terminales e instalaciones conectados a la red móvil (en millones) Terminales 2G 40,8 41,8 41,3 39,0 33,9 26,9 19,0 12,6 7,9 3,5 0,1 Terminales 3G * 1,8 2,2 2,5 3,1 5,1 9,4 14,4 18,9 21,9 24,7 26,8 Smartphones 0,7 1,9 4,2 7,6 11,8 15,4 18,9 21,4 23,8 25,9 27,7 Tarjetas 0,09 0,2 0,3 0,4 0,6 0,9 1,2 1,4 1,6 1,7 1,8 M2M 0,5 0,7 0,9 1,3 1,6 2,0 2,5 3,1 4,0 5,1 6,5 Total 43,9 46,8 49,2 51,4 53,0 54,6 56,0 57,4 59,2 50,9 62,9 Fuente: análisis TAS (*) Excluyendo smartphones FIGURA México: Número de terminales e instalaciones conectados a la red móvil (en millones) Terminales 2G 79,9 83,7 83,1 78,1 68,7 58,1 49,3 42,1 36,1 31,1 27,0 Terminales 3G * 7,0 8,1 10,4 15,2 23,5 29,8 35,7 42,1 47,4 52,1 56,5 Smartphones 2,4 6,6 11,1 16,0 20,1 26,1 32,1 36,3 40,3 43,9 47,0 Tarjetas 0,09 0,2 0,3 0,6 0,8 1,2 1,6 1,9 2,2 2,3 2,5 M2M 2,0 2,7 3,7 5,0 6,4 8,0 10,0 12,6 15,9 20,2 25,9 Total 91,4 101,3 108,6 114,9 119,5 123,2 128,7 135,0 141,9 149,6 158,7 Fuente: análisis TAS (*) Excluyendo smartphones EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 80

131 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA 5-29: Perú: Número de terminales e instalaciones conectados a la red móvil (en millones) Terminales 2G 21,1 21,2 19,7 17,9 16,1 14,4 10,0 9,2 7,8 6,0 3,0 Terminales 3G * 1,7 3,8 6,5 6,6 9,9 11,3 13,9 15,9 16,7 17,9 20,4 Smartphones 0,4 1,0 2,1 3,9 6,1 7,9 9,7 11,0 12,2 13,3 14,2 Tarjetas 0.,5 0.,9 0,1 0,2 0,3 0,5 0,6 0,7 0,8 0,9 1,0 M2M 0,3 0,3 0,5 0,6 0,8 1,0 1,3 1,6 2,0 2,6 3,3 Total 23,5 26,4 28,9 29,2 33,2 35,1 35,5 38,4 39,5 40,7 41,9 Fuente: análisis TAS (*) Excluyendo smartphones 5.3. Estimación de tráfico de datos promedio por terminal Habiendo completado la estimación de la totalidad de la base instalada de terminales móviles, se debe considerar el tráfico de datos promedio generado por cada tipo de terminal. En primer lugar, se presentan las proyecciones de tráfico de voz en la medida de que éste influencia el tráfico a ser acomodado por la red móvil. En segundo lugar, se detallan las proyecciones de tráfico de datos por terminal Proyección de minutos de uso de voz El tráfico de voz en América Latina continúa creciendo. Las estadísticas de minutos de uso por abonado que son presentadas por los operadores revela una tendencia al aumento año a año (ver figura 5-30). FIGURA Tráfico de voz (medido en minutos de uso por mes) ( ) País Operador Argentina Movistar Personal Claro Promedio Brasil Claro TIM Vivo Promedio Colombia Comcel Movistar Tigo Promedio México Telcel (*) Movistar Promedio Perú Movistar Claro Promedio (*) Número ajustado por doble conteo de tráfico de entrada y salida intra-red; los números promedio reportados son 174, 194 y 210 para 2008, 2009 y 2010 Fuentes: Wireless Intelligence; Bank of America Merrill Lynch; análisis TAS EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 81

132 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Para proyectar el tráfico de voz al 2020 se ha calculado un promedio de datos actuales de tráfico y tendencias de los operadores para cada uno de los países estudiados y se ha extrapolado después del 2014 con base en el promedio de la tasa de variación de y Con base en esto, se calculan los siguientes volúmenes de tráfico de voz por abonado (ver figura 5-31). Estas proyecciones son usadas para estimar el tráfico de voz y aplicadas de manera indistinta a abonados con terminales 2G y 3G en la medida que reflejan promedios. FIGURA Tráfico de voz (medido en minutos de uso por mes) ( ) Argentina Brasil Colombia México Perú Fuente: análisis TAS Estimación de uso promedio de datos para terminales básicos El uso de datos varía por terminal e instalación con variación a lo largo del tiempo. Con excepción del tráfico de datos por teléfono básico, que varía por país, se ha optado por homogeneizar la información para los cinco países bajo estudio. Para los teléfonos 2G, 3G y 4G se utilizó información de Strategy 153 Analytics, con crecimiento extrapolado al En el caso de smartphones, dada la falta de información pública en la región, 154 se utilizó información de Estados Unidos. Para el tráfico generado por periféricos de PCs portátiles y tablets, se utilizó información de un operador de la región para el año A partir de ello, se uso la tasa de crecimiento de smartphones para proyectar el tráfico. Con respecto al tráfico generado por instalaciones que dependen de conexiones M2M, se utilizó 155 una estimación de Wirex para Estados Unidos. Los indicadores de tráfico utilizados fueron los siguientes (ver figura 5-32). FIGURA América Latina: Tráfico promedio generado por terminal (en MB/mes) Terminales no 3G 0,4 0,8 2,0 4,7 9,5 18,2 31,9 53, Teléfonos 3G y 4G 0,4 0,8 2,0 4,7 9,5 18,2 31,9 53, Smartphones 3G y 4G PC portátiles Instalaciones M2M Fuente: análisis TAS 153. Strategy Analytics Cell traffic report June 2010 ( ) 154. Para el año 2010, se utilizó información de Validas US data para abonados non-iphone ; para el año 2014 se utilizó información de Validas US para abonados iphone en 2010; a partir del 2015 se utiliza la tasa de crecimiento de teléfonos al uso de smartphones 155. Wirex. Esta estimación es un promedio dado que el tráfico M2M depende de la frecuencia de envío de señales de monitoreo. En algunos casos, el tráfico puede alcanzar 10x el promedio asumido en este estudio. Ver ACG Research. M2M: A Big Deal for Networks or Just another Flash in the Pan? 156. Esta estimación es consistente con las proyecciones de Cisco, que proyecta una tasa de crecimiento para América Latina del 111% entre el 2010 y el EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA 82

133 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Fuentes y supuestos: Terminales no 3G y teléfonos 3G y 4G: Strategy Analytics: Cell traffic report June 2010 ( ); crecimiento extrapolado al 2020 Smartphones: 2010: Validas US data para non-iphone; 2014: Validas US para iphone en 2010; 2015: se utiliza la tasa de crecimiento de teléfonos al uso de smartphones PC portátiles: 2010: operador real en Perú; se usa la tasa de crecimiento de smartphones al uso; : se usa la tasa de aumento del smartphones a tarjetas Instalaciones M2M: Estimación de Wirex para Estados Unidos en Impacto en el tráfico de voz y datos Habiendo estimado la evolución del parque de terminales al 2020 y el tráfico de voz y datos por terminal, es posible proyectar el tráfico total de la red para los cinco países estudiados en detalle. En términos agregados, el tráfico total de datos evolucionará de 11,906 terabytes en el 2011, y alcanzará los 180,214 terabytes en el 2016, equivalente a una 156 tasa de crecimiento del 117% (ver figura 5-33). FIGURA Proyección total del tráfico de datos ( ) (en gigabytes) Argentina Brasil México Colombia Perú Total % Fuente: análisis TAS Habiendo estimado la evolución del parque de terminales al 2020 y el tráfico de voz y datos por terminal, es posible proyectar el tráfico total de la red para los cinco países estudiados en detalle. En términos agregados, el tráfico total de datos evolucionará de 11,906 terabytes en el 2011, y alcanzará los 180,214 terabytes en el 2016, equivalente a una tasa de crecimiento del 117% (ver figura 5-33). Esta estimación de tráfico permitió estimar, a nivel agregado, las necesidades de red para poder satisfacer esta demanda. En la estimación del ahorro potencial que podría tenerse al construir una red en la banda de 700 MHz para servicios de banda ancha móvil comparado con una red en bandas superiores, la evolución de tráfico fue utilizada como uno de los parámetros que permitió estimar la inversión a lo largo del tiempo Esta estimación es consistente con las proyecciones de Cisco, que proyecta una tasa de crecimiento para América Latina del 111% entre el 2010 y el EL CRECIMIENTO DEL TRÁFICO EN LAS TELECOMUNICACIONES MÓVILES DE AMÉRICA LATINA` 83

134 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA 6. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL La asignación de espectro en la banda de 700 MHz a las telecomunicaciones móviles resultará en una serie de efectos positivos para las sociedades latinoamericanas. Estos efectos han sido categorizados en dos áreas: 1) el impacto en el ecosistema de la industria en términos de gastos a incurrir para hacer posible la oferta de servicios en la nueva banda, y 2) el impacto económico y social. La primera categoría está organizada de acuerdo a la siguiente cadena de valor (ver figura 6-1). FIGURA 6-1. Contribución al ecosistema de las telecomunicaciones móviles Adquisición de espectro Adquisición de bienes productivos Adquisición de servicios operativos Adquisición de servicios comerciales Inversión para la adquisición de espectro en subasta Sitios Equipamiento de acceso Red troncal Sistemas (OSS) Ingeniería civil Mantenimiento y reparación Distribución Logística Otros servicios Aplicaciones Publicidad Integración de sistemas comerciales (CRM, facturación, etc.) Otros bienes Esta dimensión de impacto incluye, en primer lugar, la contribución que la banda ancha móvil haría al tesoro nacional en términos del pago de licencias como consecuencia de la adquisición de espectro. La siguiente contribución al ecosistema incluye la inversión para la adquisición de bienes necesarios para construir y operar la red en la nueva frecuencia. Esto incluye torres, radio bases, equipamiento de acceso a la red troncal y bienes relacionados con la infraestructura. Las dos siguientes etapas en la cadena de valor incluyen la adquisición de servicios operativos y comerciales necesarios para la prestación de servicios en la nueva banda. La segunda categoría de efectos económicos está estructurada con base en el siguiente esquema (ver figura 6-2). EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 84

135 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA 6-2. Beneficios socio-económicos Contribución directa a la economía Contribución incremental al PIB Empleo directo generado Efectos de la oferta Contribución al crecimiento del PIB Contribución indirecta a la economía Empleo indirecto generado Impuestos adicionales Efectos de la demanda Excedente del consumidor Competitividad Esta categoría incluye los efectos de asignación del espectro sobre la oferta y la demanda de servicios. En la primera dimensión, se consideran las contribuciones directas (contribución al PIB con base en el ingreso generado por la oferta de nuevos servicios y la creación de fuentes de trabajo en el sector, incluyendo empleo en las empresas de telecomunicaciones y sus proveedores) e indirectas (externalidades positivas que resultan en una contribución al crecimiento del PIB por el impacto de la tecnología en otros sectores de trabajo y la correspondiente creación de empleo). Por el lado de la demanda, estudiamos, en particular, la generación de excedente del consumidor como resultado de la reducción de precios en banda ancha móvil. El efecto de competitividad no es evaluado cuantitativamente por la dificultad en parametrizar la variable. Este capítulo pasa revista a cada uno de los beneficios incluyéndose una estimación cuantitativa para cada uno de los cinco países estudiados en detalle. En primer lugar, se estiman l o s d o s i m p a c t o s r e s u l t a n t e s d e l a c c e s o d e l a s telecomunicaciones móviles al espectro en 700 MHz: el cerramiento de la brecha de cobertura en banda ancha móvil y el aumento de líneas de banda ancha móvil. A partir de esto, se estiman los impactos en el ecosistema y los beneficios económicos y sociales Cerramiento de la brecha de cobertura en zonas aisladas y rurales A pesar de la alta tasa de adopción de la telefonía móvil en América Latina, todavía existen zonas no cubiertas por la banda ancha móvil. Las estimaciones de este estudio, validadas con información de reguladores en cada país, indican que aún existen brechas importantes de cobertura en banda ancha 158 móvil (ver figura 6-3) El concepto de cobertura implica población que efectivamente reside en lugares donde existe señal de telecomunicaciones móviles para poder acceder a la red. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 85

136 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Figura 6-3. América Latina: Brechas de cobertura de la banda ancha fija y móvil País Brecha aproximada de cobertura de banda ancha fija Brecha aproximada de cobertura de telefonía móvil (1) Brecha aproximada de cobertura de banda ancha móvil (2) Argentina 10% 6% 25% Brasil 6% 4.5% 24,6% ( 3) Colombia 37% 17% 48% México 8% (6) 1% 61% (4) Perú 41% (7) 5% 35% (5) (1) Autoridades regulatorias reportando a UIT (2)Estimada en base a cobertura de 3G (3)Anatel y consolidación Teleco (Datos de Junio / 2011) (4)La penetración de 3G para México es calculada utilizando la población de ciudades en donde el servicio de 3G está habilitado. La información de las ciudades con servicio 3G está actualizada a enero del 2010 (5) Calculado en base de número de habitantes en distritos donde existe conexión 3G (348 distritos cubiertos con red 3G del total de 1833) para el caso de telecomunicaciones móviles. Diciembre 2010 (6) La cobertura de Banda Ancha a través de la tecnología ADSL es tomado del informe anual 2008 de Telmex (7) Calculado en base de número de habitantes en distritos donde existe al menos un cliente con conexión ADSL (14%), Diciembre 2010 Por sus características de propagación, la banda de 700 MHz es más apropiada que frecuencias superiores para garantizar la cobertura en zonas rurales, inclusive que la banda de MHz, aunque en este caso la diferencia sería menor. Por lo tanto, al asignar el Dividendo Digital para la banda ancha móvil, se podría aumentar la cobertura geográfica de la banda ancha en cada país, prestando servicio en zonas no cubiertas (ver figura 6-4). Figura 6-4. Cobertura incremental a ser alcanzada por la banda ancha móvil 90% 95% 94% 95% 90% 92% 94% 89% 75% 75% 63% 52% 59% 65% 39% Argentina Brasil Colombia México Perú Cobertura banda ancha fija Cobertura actual (3G) Cobertura estimada con la banda de 700 MHz Fuente: análisis TAS 159. Es de esperarse que para aquellos operadores que tengan frecuencias en 850 MHz (e inclusive 900 MHz), los servicios de banda ancha con tecnologías como HSPA sean primeramente ofrecidos utilizando estas bandas y sólo posteriormente a través de LTE que estará disponible mayormente en 700 MHz. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 86

137 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Esta cobertura adicional garantizará la provisión de servicio de banda ancha a una porción importante de la población residencial, así también como pequeñas empresas que no cuentan al día de hoy con acceso al servicio de banda ancha fija. Él numero adicional de abonados han sido estimados en función de evolución de la penetración de telefonía móvil en 160 los dos deciles inferiores de la pirámide sociodemográfica (ver figura 6-5). Estos abonados estarían siendo cubiertos por servicio de banda ancha móvil. FIGURA 6-5. Abonados adicionales creados como consecuencia de una cobertura total de los territorios nacionales País Penetración en la base de la pirámide (2009) Penetración en la base de la pirámide (2015) Abonados adicionales (millones) Argentina 50% 70% 0,91 Brasil 33% 56% 2,68 Colombia 63% 80% 2,60 México Perú 57% 22% 66% 40% 1,94 0,58 Total 8,70 Fuentes: Euromonitor; análisis TAS Así, es importante concluir que la asignación de la banda de 700 MHz a la banda ancha móvil permitirá la incorporación de un total de 8.7 millones de abonados que hasta la fecha han estado del servicio de banda ancha. Para el conjunto de América Latina este número llegaría a 20,1 millones Aumento de la penetración de banda ancha móvil como resultado de una reducción de precios La asignación de espectro en la banda de 700 MHz para el despliegue de redes con alta capacidad permitirá aumentar la oferta de banda ancha móvil, la cual sumada a la existente en las bandas convencionales y la basada en las redes fijas de cable y telecomunicaciones, resultará en una disminución de precios. En mercados competitivos como los latinoamericanos, la disponibilidad de espectro permitirá a la industria móvil posicionarse en situación de paridad competitiva con los oferentes de banda ancha fija. En este contexto, y dada la reducción de costos de despliegue como resultado de la utilización de una banda mas eficiente, una porción del excedente del productor (beneficio de costos mas bajos) será trasladado al consumidor. El servicio de la banda ancha móvil en América Latina, aún está siendo ofrecida a precios relativamente altos (ver figura 6-6) al compararla con mercados más maduros. El menor número de radio bases necesarias para el despliegue de una red con cobertura similar a la existente se traduce en una menor inversión de despliegue y en menores costos de operación y mantenimiento. Parte de estas economías se transferirán a los consumidores finales en términos de menores precios, que estimamos serán en el mediano plazo, entre 10 y 15% menores que los que se esperarían sin la utilización de la banda de 700 MHz. De manera conservadora, suponemos que, como resultado del aumento de la oferta a partir del acceso a espectro en la banda de 700 MHz, los precios de banda ancha 161 móvil disminuirán en un 10%. Convertidos a dólares estadounidenses con fines comparativos, la disminución de tarifas es presentada en la figura Calculado a partir de dividir el ingreso de hogares y adopción de telefonía móvil de acuerdo a la Encuesta Nacional de Hogares de cada país en diez niveles; la adopción es estimada a incrementarse en función de la penetración de telefonía móvil en niveles inferiores de ingreso; la tasa de aumento esta determinada por la extrapolación de la tendencia histórica Similar al supuesto utilizado en el estudio del impacto socio-económico de Dividendo Digital en Asia. Boston Consulting Group. Socio-economic impact of allocating 700 MHz band to mobile in Asia Pacific. A report to the GSMA. October EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 87

138 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA 6-6. Precios de banda ancha móvil en América Latina (2011) País Operador Pospago Precio Precio Prepago Precio Precio Restricción ML(*) US $ Duración (*) US $ Argentina Claro 512 MB $ 69 $17,69 1 hora $ 3 $0,77 Personal 1 GB 3 GB 500 MB $ 99 $ 129 $ 69 $25,38 $33,08 $17,69 2 hora 4 hora 6 hora 1 GB $ 5 $ 7 $ 10 $ 10 $1,28 $1,79 $2,56 $2,56 3 GB Ilimitado $129 $199 $33,08 $51,03 2 GB 3 GB $ 19 $ 59 $4,87 $15,13 Movistar 500 MB 69 $17,69 1 GB $ 10 $2,56 1 GB 99 $25,38 Brasil Claro 3 GB 500 MB 1 GB 129 R 69,90 R 79,90 $33,08 $38,83 $44, MB 700 MB R 9.90 R $5,50 $19,39 3 GB 5 GB 10 GB R 89,90 R 119,90 R 199,90 $49,94 $66,61 $111,06 TIM 20 horas R 39,90 $22, MB R 10 $5,56 60 horas R 59,90 $33,28 90 horas 120 horas R 79,90 R 109,90 $44,39 $61,06 Vivo 250 MB R 59,90 $33, MB 12 $6,67 1 GB R 89,90 $49, MB 35 $19,44 4 GB 8 GB 2 GB R 119,90 R 199,90 R 69,90 $66,61 $111,06 $38,83 1 GB 130 $72,22 Oi Comcel 5 GB 10 GB 500 MB R 89,90 R $49,94 $66,61 $13, MB 4,640 $2,44 Colombia 1 GB 2 GB 3 GB 5 GB 10 GB GB $16,80 $21,02 $27,87 $36,29 $48,94 $12,59 $16, MB 1.5 GB 3 GB 1 día 7 días 29,000 52,200 81,200 3,990 19,990 $15,28 $27,50 $42,77 $2,10 $10,53 Movistar 2 GB 3 GB 5 GB 10 GB 500 MB $M 249 $21,02 $27,87 $35,77 $52,62 $19,76 30 días 100 MB 69, $36,87 $3,89 1 GB $M 299 $23, MB 79 $6,27 Telcel 2 GB $M 349 $27, MB 199 $15,79 México Perú Claro 3 GB 5 GB 10 GB 2 GB 3 GB 5 GB 10 GB Ilimitado 2 GB $M 399 $M 449 $M 599 NS 78,34 NS 98,17 NS 127,09 NS 197,03 NS 197,33 NS 78,34 $31,67 $35,63 $47,54 $27,98 $35,06 $45,39 $70,37 $70,48 $27, GB 3 GB 100 MB 200 MB 700 MB 1.5 GB 3 GB 25 MB NS 10 NS 15 NS 45 NS 80 NS 130 NS 3 $23,73 $39,60 $3,57 $5,36 $16,07 $28,57 $46,43 $1,07 Movistar 3 GB NS 98,16 $35,06 40 MB NS 10 $3,57 5 GB NS 119 $42, MB NS 20 $7,14 10 GB NS 197,33 $70, MB NS 50 $17,86 Fuentes: Sitios de operadores (Mayo, 2011) (*) En moneda local; todos los precios excluyen promociones temporarias EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 88

139 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA 6-7. Proyección de tarifas de banda ancha móvil - Pospago de 3 GB (en US $) (*) País Argentina Brasil Colombia México Perú $ 32,25 $ 49,67 $ 27,52 $ 32,75 $ 46,76 $ 29,02 $ 44,70 $ 24,76 $ 29,48 $ 42,09 Fuente: análisis TAS (*) Se elige la tarifa de pospago de 3 GB porque éste es el paquete más común en usuarios de tarjetas de conexión de computadoras Con esta disminución de precios y suponiendo una elasticidad 162 de demanda de 0,6, la adopción de banda ancha móvil aumentará más allá de las proyecciones del mercado (ver figura 6-8). Como lo muestra la figura 6-8, la disminución de precios de banda ancha móvil en 10% aumentará la adopción de la misma en más de 1.4 millones de líneas Impacto en el ecosistema Como se mencionó arriba, si la banda de 700 MHz fuese asignada a la industria móvil, esto resultaría en una serie de contribuciones de la misma al ecosistema de la industria. Estos efectos han sido analizados a lo largo de la cadena de valor en tres áreas discretas: Contribución al tesoro nacional como resultado de la adquisición de espectro; Contribución a industrias como la construcción, electrónica y de telecomunicaciones, al invertirse en la construcción de una nueva red en el espectro asignado; Adquisición de servicios operativos y comerciales para poder ofrecer servicios en la nueva banda. FIGURA 6-8. Adopción de banda ancha como resultado de la disminución de precios ( ) (en miles) Argentina Brasil Colombia México Perú Fuente: análisis TAS IEsc. Base * Esc. Precio+ Esc. Base * Esc. Precio+ Esc. Base * Esc. Precio+ Esc. Base * Esc. Precio+ Esc. Base * Esc. Precio * Escenario base: Difusión de la banda ancha móvil de acuerdo a la proyección + Escenario precio: Aumento de la difusión de banda ancha como resultado de la disminución de precios 162. Se supone una tasa de elasticidad de demanda de Este valor puede ser conservador si se considera la externalidad recientemente calculada en banda ancha fija para América Latina, la que puede acercarse a 1.88, pese a que la elasticidad para todos los países de la muestra del estudio citado es de Ver Galperin, H. y Ruzzier, Ch. "Las tarifas de banda ancha: benchmarking y análisis", in Galperin, H., Jordán, V. y Peres, W. Acelerando la Revolución Digital: banda ancha para América Latina y el Caribe. Santiago: CEPAL y DIRSI, EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 89

140 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Contribución de recursos a gobiernos generados por la adquisición de espectro a ser reasignado en la banda de 700 MHz El posible cambio de atribución de uso del espectro en la banda de 700 MHz para que pueda ser utilizado para la prestación de servicios de telecomunicaciones móviles conlleva la necesidad de concesionarlo. Aunque existen varias modalidades de asignación y concesionamiento del uso del espectro, hoy la más común es el llevar a cabo un proceso de subasta. Existen ejemplos en la región donde todavía se ha concesionado de otras maneras, siendo la asignación directa y 163 concurso de belleza la más común. Sin embargo, muchos gobiernos han tomado como punto de partida el supuesto de que una de las mejores maneras de garantizar que el espectro termina en manos de quien más valor le otorga es a través de alguna modalidad de subasta, lo que ha llevado a que sea ésta la que domina en la región. Es importante resaltar que una subasta no necesariamente utiliza únicamente el criterio económico (principalmente la contraprestación pagada), sino que puede incorporar otro tipo de criterios que tienen un valor en el puntaje a ser considerado en las pujas durante la subasta. En América Latina se han utilizado criterios tales como promesa de monto de inversión, cobertura, tiempo de despliegue o precio máximo al usuario final. En comparación con el resto del mundo, en América Latina ha habido pocas licitaciones de espectro recientemente. Así, existe información limitada que pueda servir de guía para estimar el valor que podría ser recaudado al poner a disposición del mercado la banda de 700 MHz. Sin embargo, es posible utilizar la poca experiencia regional (recientemente, Brasil, México, Perú y Colombia) y las licitaciones conducidas algún tiempo atrás en Estados Unidos y recientemente Alemania para hacer comparaciones entre licitaciones de bandas iguales o parecidas, lo que permite estimar el posible valor que podría ser recaudado. Este ejercicio no busca estimar el valor del espectro de abajo hacia arriba, construyendo flujos de caja esperados; el objetivo es traducir la experiencia existente (benchmark) en valores aproximados que podrían ser recaudados al licitar el espectro. Bajo igualdad de condiciones en una licitación, el principal factor determinante del precio del espectro es la riqueza del país (PIB) y su población. Para efectos de normalización, la medida típica utilizada es el precio pagado por MHz por habitante (US $/MHz /POP). Esta sección busca hacer una estimación del precio esperado utilizando el histórico y los indicadores macroeconómicos. Es necesario sin embargo enfatizar que el tipo de subasta y las condiciones en las licencias afectan sustancialmente el precio a ser pagado. El limitar el número de posibles participantes le quita profundidad al mercado, lo que se traduce en precios menores. Existen varios casos en la región, tales como la asignación de la banda de 1.700/2.100 MHz en Chile (2009) y México (2010), en donde se impusieron restricciones a la participación de empresas existentes (por ejemplo, topes de espectro) o bien se consideraron criterios diferentes a la remuneración económica (Chile utilizó como criterio principal 164 el ofrecimiento de la mayor cobertura en el menor tiempo ); por lo tanto, sólo fue recaudado US $0,012 y US $0,004 por MHz por habitante en cada uno de estos países respectivamente. Asimismo, si se imponen condiciones de cobertura o 165 inversión, los flujos de caja esperados disminuyen, lo que tiene un impacto negativo en el precio que puede ser esperado. Finalmente, los pagos recurrentes por el pago del espectro que varían sustancialmente de país a país también tienen un impacto importante en los flujos de caja y por lo tanto en el valor recaudado. La Licitación 66 en Estados Unidos en 2006, de 90 MHz en la banda de 1.700/2.100 MHz, recaudó US $ millones, lo que se traduce en un pago por MHz por habitante de US $ 0,508. La Licitación 73, de 62 MHz en la banda de 700 MHz, 166 recaudó US $ millones, equivalente a US $1,006 por MHz por habitante. En Alemania en 2010 fueron licitados MHz en la banda de 800 MHz ; esta subasta recaudó millones, equivalentes a US $1,034 por MHz por habitante. Es decir, se pagó una cantidad normalizada similar a la obtenida 168 en Estados Unidos Los concursos de belleza o beauty contests implican una asignación a quien demuestre, con ciertos criterios, que es el mejor calificado para cumplir los objetivos buscados en la utilización del espectro. En general, estas asignaciones no exentan a los postores del pago de derechos Este concurso público, por sus características, es más apropiado considerarlo un concurso de belleza que una licitación en su acepción más común En Chile, Subtel adjudicó 60 MHz a Nextel y 30 MHz fueron asignados a VTR con un pago aproximado de US $17.8 millones. En México, parte del espectro subastado en la Licitación 21 fue asignado a Nextel con un pago de US $14.13 millones por 30 MHz nacionales, mientras que por los otros 30 MHz fueron recaudados US $405 millones Éste es el valor efectivamente recaudo ( net winning bids ) y no el valor obtenido al finalizar la subasta ( winning bids ). La diferencia es de aproximadamente US $ 160 millones El espectro del Dividendo Digital en Europa se encuentra en la banda de 800 MHz Para simplicidad en el ejercicio de estimación, por lo tanto, se tomó sólo el valor de referencia de Estados Unidos. El rango de precios es similar, ya que está cercano a la variación que ha presentado la moneda estadounidense frente al dólar en el último año. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 90

141 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA En 2010, México licitó 30 MHz en la banda de 1.700/2.100 MHz 169 por US $405 millones, equivalentes a US $0,12/MHz/POP, lo que implica que el descuento con respecto a la licitación de la banda similar en Estados Unidos cuatro años antes es de aproximadamente 76,3%. En Brasil en ese mismo año, la licitación de 20 MHz en la banda H (1.900/2.100 MHz) recaudó US $ 712 millones, equivalentes a US $0,187/MHz/POP, lo que implica un descuento comparado con Estados Unidos de 73,7%. En Perú en 2007 se licitaron 25 MHz en la banda B (850 MHz) y 35 MHz en las bandas D y E por US $22.2 millones y US $27 millones respectivamente, dando un pago de US $0,033 y US $0,28 por MHz por habitante; el descuento implícito respecto a Estados Unidos es, en promedio, por lo tanto, de 93,5%. En Colombia en 2010 fueron 20 MHz en la banda de MHz licitados por US $220 millones, equivalentes a US $0,022 por MHz por POP (ver figura 6-9). FIGURA 6-9. Licitaciones de espectro recientes País Año Referencia Frecuencia Oferta Recaudación (en millones US $) Precio por MHz por POP (US $) Estados Unidos 2006 Auction /2.100 MHz Estados Unidos 2008 Auction MHz Alemania Venezuela MHz MHz Perú 2007 Banda B 850 MHz Perú 2007 Bandas D y E MHz Panamá Chile MHz MHz /2.100 MHz 90 MHz $ 18 México 2010 Licitación MHz /2.100 MHz 90 MHz 62 MHz 60 MHz 60 MHz 25 MHz 35 MHz 80 MHz 30 MHz $ $ $ $ 240 $ 22 $ 27 $ 229 México 2010 Licitación MHz 30 MHz $ 217 México 2010 Licitación MHz /2.100 MHz 30 MHz $ 14 Brasil 2007 Bandas F, G, I y J MHz/2.100 MHz 90 MHz $3,096 Brasil 2010 Banda H MHz/2.100 MHz 20 MHz $ 712 Colombia MHz 20 MHz $ 22 Colombia MHz 50 MHz $42 Costa Rica 2010 Varias MHz $170 Nicaragua MHz 50 MHz $ 12 $ 405 $ 0,508 $ 1,006 $ 1,034 $ 0,137 $ 0,033 $ 0,028 $ 0,813 $ 0,012 $ 0,120 $ 0,080 $ 0,004 $0.180 $ 0,187 $ 0,024 $ 0,018 $0,338 $ 0,041 Fuente: Autoridades regulatorias Nota: Para la Licitación 21 en México fueron separadas las asignaciones a Nextel y las de los demás operadores. La relación entre el precio normalizado y el PIB per cápita es bastante estrecha. Es decir, el descuento aplicado a los precios entre Estados Unidos y los países de América Latina es parecido a la diferencia en niveles de riqueza. En México y Brasil el descuento pagado por el espectro en esta comparación es marginalmente menor a la diferencia en riqueza, mientras que en el resto de los países es marginalmente mayor (entre 5 y 10 puntos porcentuales). Por tanto parece razonable estimar el precio esperado utilizando como deflactor el PIB per cápita en cada uno de los países. Siendo así, la experiencia muestra que podría esperarse un monto algo superior tanto en México como en Brasil, y algo inferior en el resto de los países. Esto dependerá, claramente, de las condiciones de competencia, de potencial de mercado, de percepción de estabilidad, y, sobre todo, de la estructura y condiciones de la subasta. De lo anterior, en la región podría esperarse un promedio global de US $0,185 por MHz por habitante, lo que arroja una recaudación estimada total de US $9,449 millones en el caso de que fuesen licitados 90 MHz en todos los países. Estos números esconden las diferencias regionales, ya que México y Brasil representan tan sólo 55% de la población de América Latina, pero podrían esperar recaudar dos terceras partes del total regional (ver figura 6-10) No se considera el espectro que fue asignado a Nextel dado que no fue contestado por varias empresas al haber tenido condiciones de reserva para un nuevo entrante a la telefonía móvil. Utilizarlo en el cálculo se traduciría en valores sustancialmente menores de recaudación. Es de esperarse que en la banda de 700 MHz, en el caso de licitarse en México para la telefonía móvil, existan varios contendientes en la licitación, por lo que se decidió para la presente estimación considerar únicamente los 30 MHz que fueron pujados por más de una empresa El haber considerado como criterios principales la cobertura y el tiempo de despliegue, aunado a la existencia de topes de espectro, permite considerar esta asignación como un concurso de belleza. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 91

142 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA Estimación de la recaudación potencial por la licitación del espectro de 700 MHz Factor de descuento Recaudación esperada (millones de US $) US $/MHz/habitante) 60 MHz 60 MHz Argentina -80,6% 0, Brasil -77,1% 0, Colombia -86,7% México -79,7% 0, Perú -93,9% 0, Resto de A.L -87,2% 0, América Latina -81,6% 0, Fuente: análisis TAS Nota: El factor de descuento fue calculado como PIB de cada uno de los países dividido por el PIB de Estados Unidos en dólares corrientes Adquisición de bienes productivos Una de las principales características de la banda de 700 MHz es que permite una propagación más amplia de las señales. Una radio base en esta frecuencia en terreno plano tiene un 171 alcance aproximado de 13 kilómetros de radio, dando una cobertura superior a los 500 kilómetros cuadrados. Aun en terrenos con topografía montañosa, la cobertura posible supera los 30 km2 (ver figura 6-11). Esto se traduce en que esta frecuencia es ideal para prestar servicios en zonas densas en las que la demanda por servicios de datos no sea excesiva, o bien, para cubrir zonas rurales, que tienen poca densidad poblacional. Específicamente, en el caso de zonas rurales, el alcance posible con una radio base en 700 MHz hace que una inversión donde la principal palanca de costos es la construcción de la torre y la instalación de la antena, se pueda dar servicio a un número mayor de personas, lo que implica que la inversión, la operación y el mantenimiento de la radio base será amortizada entre más usuarios, viabilizando así el servicio. En gran parte del territorio latinoamericano eso no sería posible con la utilización de frecuencias superiores, ya que el pequeño número de usuarios no permite rentabilizar la inversión y los costos en que se incurre para proporcionarles el servicio. FIGURA Cobertura alternativa Radio de cobertura (km) Cobertura aproximada (km2) Cobertura de una radio base en 700 MHz Valles y llanuras (terreno plano) Lomerío Sierra Terreno montañoso Cobertura de una radio base en zonas urbanas 700 MHz 900 MHz 1,700/2,100 MHz 2,600 MHz ,8 0,5 0, ,6 0,4 Fuente: FCC; análisis TAS 171. FCC. The Broadband Availability Gap, OBI Technical Paper No.1. Abril 2010 EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 92

143 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA De esta manera, desde el punto de vista de red y cobertura, existen dos valores asociados con el Dividendo Digital: Un número menor de estaciones radio base para lograr una cobertura similar a la que podría aspirarse utilizando bandas de frecuencias más altas; Una mayor cobertura de habitantes, lo que no es factible económicamente con bandas superiores. Para cada uno de los países estudiados en detalle, se estimó la construcción de dos redes hipotéticas. Por un lado, la primera red supuso la utilización de la frecuencia de 700 MHz; la segunda red buscó una cobertura similar a la actual con una combinación de radio bases en las bandas de 900 MHz, MHz, 1.700/1.900 MHz y MHz. Para ello, se analizó en detalle la situación demográfica y topográfica de cada uno de los cinco países estudiados. Primeramente fueron separadas las áreas urbanas; posteriormente, para poblados menores, se analizó su cercanía con otros poblados y el tipo de terreno (valle/llanura/planicie, lomerío, sierra o montaña) en la zona, verificando que podría conseguirse cubrir con una sola radio base un número suficiente de personas para hacer viable la inversión para poder prestar el servicio. El número mínimo de personas necesarias rondó en todos los casos entre 3,000 y 4,000, que, con una penetración conservadora de alrededor de 10% y una generación de ingresos recurrentes de entre 10 y 15 dólares al mes por suscriptor, sería posible recuperar el costo de instalar, operar y mantener una estación radio base. Con estas premisas se estimó el número de radio bases necesarias para conseguir una cobertura urbana cercana al 100% y aumentar al máximo la cobertura rural sujeto a que toda la inversión es rentable. Actualmente, la cobertura rural, 172 dependiendo del país, varía entre 30% y 65%. Con la utilización de la banda de 700 MHz, esta cobertura podrá verse incrementada entre 16 y 31 puntos porcentuales (figura 6-11). Dada la dispersión de la población y la topografía poco favorecedora para la instalación de redes celulares, especialmente en Perú, Colombia y México, es difícil conseguir una cobertura casi universal. Sin embargo, este aumento de la cobertura, por pequeño que parezca, implica que 26, millones más de latinoamericanos podrán estar en la posibilidad de adquirir servicios de banda ancha móvil porque 174 estarán disponibles. Esto equivale a uno de cuatro habitantes rurales, muchos de ellos personas de menores recursos económicos; la disponibilidad de la banda ancha podrá ser un elemento importante de disminución de la brecha digital y de la inclusión social de esta parcela de la población. Así, la cobertura de 3G, más adecuada para el uso en banda ancha móvil por su capacidad de transmisión, no llega a una cobertura adecuada. Esta es la razón por la cual la banda de 700 MHz permite llegar a zonas no cubiertas, potenciando la masificacion de la banda ancha móvil. En toda América Latina, suponiendo que es posible utilizar la banda de 700 MHz para la prestación de servicios de banda ancha móvil, el número de estaciones radio base necesarias hacia el año 2020 para conseguir los resultados anteriores es de aproximadamente De éstas, serían estaciones en zonas rurales, permitiendo aumentar la cobertura a 26,5 millones de personas, equivalentes a 4,8% de la población (figura 6-12). En caso de no contar con la banda de 700 MHz, el número de estaciones radio base necesarias sería de aproximadamente Con este número, 150% superior, no sería posible aumentar de manera significativa la cobertura actual, ya que sería o físicamente imposible o económicamente inviable (ver figura 6-12). El resultado anterior trae además consigo una necesidad de inversión sustancialmente menor. El hecho de requerir un número de estaciones 60% menor implica que tanto los requisitos de capital invertido como los gastos recurrentes de operación y mantenimiento serán menos cuantiosos. Por un lado, el despliegue de las redes será en un menor lapso de tiempo, anticipando así la disponibilidad del servicio y la adopción por parte de la población. Por otro, habrá un impacto en el costo de prestación de los servicios, que estimamos que será de mínimamente un 10% menor a los precios que podrían practicarse en el caso de no utilizase el Dividendo Digital. Es decir, el capital ahorrado en el despliegue de una red en la banda de700 MHz no sólo será liberado para otras inversiones, sino que se verá reflejado en menores precios que generan un excedente del consumidor, que tendrá un derrame en el resto de la economía, como se describirá en las secciones posteriores Esta cifra fue estimada utilizando la información pública en los sitios de los reguladores y en los departamentos de geografía y estadística de cada uno de los países. Como las definiciones de población urbana, rural, dispersa, así como de aglomeraciones, varían de país en país, se buscó homogeneizar al máximo la información, utilizando como punto de partida los datos reportados por el Banco Mundial (http://data.worldbank.org/data-catalog#tables) para cada uno de los países y ajustándolos con informaciones censales y de los órganos reguladores ,9% de la población total de la región El número estimado es 24,2%. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 93

144 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA Número de estaciones radio bases en redes alternativas Sin 700 MHz Con 700 MHz Estaciones sin la utilización de la banda de 700 MHz Estaciones utilizando la banda de 700 MHz Estaciones rurales Diferencia en el número de radio bases Argentina Brasil Colombia México Perú Resto de América Latina Total Fuente: análisis TAS Considerando los flujos hasta el 2020, una tasa de descuento de 10% y un despliegue de acuerdo a la demanda, es posible estimar el menor dispendio de capital y menores costos de operación en los que incurrirán las empresas de telecomunicaciones si tuvieran acceso a la banda de 700 MHz. Por un lado, una red sin la utilización de esta banda costaría aproximadamente US $ millones. Esta cantidad se vería reducida a US $ millones con la red utilizando el espectro del Dividendo Digital. En otras palabras, el valor del Dividendo Digital en términos de construcción y operación de red es de aproximadamente US $ millones (figura 6-13). Este valor es una estimación del excedente del productor, que será al menos parcialmente transformado en excedente del consumidor bajo condiciones razonables de competencia. En resumen, desde el punto de vista de despliegue y operación de la red de telecomunicaciones, es incuestionable el valor del Dividendo Digital para América Latina: supera los US $5.000 millones de dólares, permite una cobertura adicional de 31.5% para banda ancha móvil, e impacta los precios de los servicios a la baja, permitiendo así una mayor adopción de los servicios de banda ancha. 175 FIGURA Costo de redes alternativas (en millones de US $) Costo de la red sin 700 MHz Costo de la red con 700 MHz Valor del Dividendo Digital en construcción y operación de la red Argentina Brasil Colombia México Perú Resto de América Latina Total a Fuente: análisis TAS 175. La estimación del impacto de los precios en la adopción y penetración de los servicios es abordado en la sección 6.2. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 94

145 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Adquisición de servicios operativos y comerciales El despliegue de una red en la banda de 700 MHz requiere la adquisición de servicios operativos y comerciales. Estos han sido categorizados en las siguientes áreas y calculados con base en benchmarks de operaciones móviles: Servicios operativos (adquisición de sistemas de información): Sistemas de información (6% de ingresos) Servicios comerciales Mercadeo (7% de ingresos) Estos benchmarks han sido aplicados a los ingresos adicionales acumulados a ser generados por las telecomunicaciones móviles de cada uno de los países estudiados si la banda de 700 MHz fuese atribuida a esta industria (ver figura 6-14). Extrapolando al resto de América Latina, estimamos que los servicios operativos podrán equivaler a aproximadamente US $ 194 millones y los servicios comerciales a US $ 44 millones. De acuerdo a la figura 6-14 y a esta generalización, el impacto total anual de la industria móvil en el ecosistema de servicios operativos y comerciales si ésta se viese atribuida la banda de 700 MHz sería de US $ 474 millones. FIGURA Adquisición anual de servicios operativos y comerciales ( ) (en millones de US $) Argentina Brasil Colombia México Perú Servicios operativos 10,91 52,71 16,08 24,45 4,83 Servicios comerciales 12,73 61,49 18,76 28,64 5,64 Fuente: análisis TAS 6.4. Contribución directa a la economía Si las telecomunicaciones móviles se viesen asignadas de espectro en la banda de 700 MHz, ésta contribuiría directamente a las economías latinoamericanas, tanto en términos de crecimiento de ingresos resultantes de la oferta de bienes y servicios, como por la correspondiente creación de empleo. Ambos efectos serán analizados a continuación Contribución directa al PIB La contribución directa de las telecomunicaciones móviles al PIB si ésta se viese asignada espectro en la banda de 700 MHz está medida en términos de los ingresos adicionales a ser generados a partir de los bienes y servicios ofrecidos al mercado por la nueva disponibilidad de espectro. Ésta incluye abonados adicionales de servicios móviles de voz y datos en zonas rurales (resultante de la cobertura adicional lograda con las redes de 700 MHz) y conexiones adicionales derivadas de una reducción de precios de 10% con base en una elasticidad de demanda de 0,6. El punto de partida para la estimación de la primera es la porción de la población (generalmente residente en áreas rurales) a la que no llega al día de hoy la telefonía móvil (ver figura 6-3). Con esta base, y considerando que la banda de 700 MHz permite mejorar de manera importante la cobertura nacional, se estimó el número de abonados que gradualmente podría acceder a la telefonía móvil, en particular la banda ancha móvil (ver figura 6-7). Utilizando el ARPU compuesto de voz y datos (que incluye el servicio de banda ancha) proyectado de cada uno de los mercados nacionales al 2012, se calculó el ingreso adicional en términos de líneas rurales. Para la estimación de ingresos adicionales de banda ancha móvil, se estimó el número de nuevas líneas adquiridas como resultado de la disminución de precios con base en una elasticidad de 0.6 (ver figura 6-6). Con esta base, y considerando un ARPU medio para cada mercado equivalente al promedio de 176 las tarifas actuales de servicio pospago por 3 GB, se calcularon los ingresos adicionales. Los resultados están incluidos en la figura Se elige esta tarifa dado que las estadísticas promedios de uso de tarjetas de conexión de PCs en América Latina es 2.5 GB. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 95

146 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Figura Ingresos adicionales generados por el Dividendo Digital asignado a las telecomunicaciones móviles ( ) (en miles de US $) Argentina Brasil Colombia México Perú Total Líneas rurales B. ancha móvil Total Líneas rurales B. ancha móvil Total Líneas rurales B. ancha móvil Total Líneas rurales B. ancha móvil Total Líneas rurales B. ancha móvil Total Líneas rurales B. ancha móvil Total Total 13,380 9,056 22, ,461 38, ,584 48,124 6,835 54,959 90,597 5,660 96,257 6,296 5,555 11, ,858 65, ,087 36,814 12,539 49, ,379 54, ,095 88,944 9,510 98, ,368 9, ,273 15,934 8,586 24, ,439 95, ,695 80,451 16,370 96, ,685 74, , ,145 13, , ,146 13, ,589 29,027 11,111 40, , , , ,058 24, , ,310 99, , ,152 19, , ,494 18, ,536 45,685 16,666 62,352 1,111, ,182 1,289, ,058 32, , , , , ,580 24, , ,961 24, ,017 59,361 21,716 81,078 1,220, ,961 1,458, ,058 39, , , , , ,580 30, , ,961 32, ,507 59,361 26,767 86,128 1,220, ,502 1,532, ,058 44, , , , , ,580 34, , ,961 43, ,828 59,361 30,302 89,663 1,220, ,468 1,620, ,058 47, , , , , ,580 37, , ,961 58, ,686 59,361 32,322 91,683 1,220, ,762 1,729, , ,993 50, , ,910 1,193, ,310 3,190, ,215 1,628, ,526 4,808, ,580 1,909,058 39, , ,359 2,125, ,961 2,359,412 79, , ,204 2,644,897 59,361 59,361 34, ,873 94, ,622 1,220,270 8,435, ,617 2,594,330 1,877,208 11,353,930 Fuente: análisis TAS Para efectos de comparación con el escenario de asignación de espectro a la radiodifusión, se suman los ingresos anuales a ser generados una vez que la cobertura de telefonía rural es completada (alrededor del 2016) con los ingresos generados por la banda ancha móvil en el 2020: Argentina: $ 181,9 millones Brasil:$ 878,5 millones Colombia: $ 313,4 millones México: $ 409,2 millones Perú: $ 94,2 millones Resto de América Latina: $ 643,0 millones Total: $ 2.520,2 millones Contribución directa a la creación de empleo La provisión de nuevas líneas de telefonía móvil en medios rurales (sección 6.1) y de conexiones adicionales de banda ancha móvil (sección 6.2) requerirán el reclutamiento de nuevos empleados. El número de empleos a ser creados en las empresas de telecomunicaciones y sus proveedores (considerados como un efecto directo) ha sido estimado a partir del número de empleos de la industria móvil de los países estudiados en detalle en función del número de líneas móviles (ver figura 6-16). Con la base de los ratios por línea, y sumando el número de líneas rurales y de banda ancha móviles adicionales a ser generadas por el Dividendo Digital, se estimó el número de empleos directos e indirectos a ser generados por la industria (ver figura 6-17) Para la estimación se considera un efecto de economías de escala del 80% (ver Katz et al., 2002) EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 96

147 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA Empleo directo e indirecto de las telecomunicaciones móviles (2006) País Número de líneas móviles (millones) Empleos Directos Indirectos Total Argentina 31, Brasil 99, Colombia 29, México 57, Perú 8, Fuente: Convergencia Research. Contribución de la telefonía móvil a las economías de América Latina y el Caribe, 2007 FIGURA Creación de empleos en la industria de las telecomunicaciones móviles País Líneas adicionales (millones) Empleos directos Empleos indirectos Total Argentina Brasil ,195 Colombia México Perú Resto de América Latina Total 17, Fuente: análisis TAS Así, de acuerdo a las estimaciones presentadas en la figura 6-16, el número de empleos a ser creados en la industria móvil de los cinco países estudiados como consecuencia de la reasignación de espectro alcanza 2,988, llegando a un total de 4,100 para toda América Latina. El número reducido de empleos creados se debe a que las industrias móviles en la región ya están operando a niveles de masa crítica de acuerdo a las economías de escala, después de la vasta creación de puestos de trabajo de la industria de las telecomunicaciones 178 en los últimos años Contribución indirecta a la economía La industria de las telecomunicaciones móviles contribuye indirectamente a tres áreas de la economía: crecimiento del PIB como resultado de los efectos de derrame de las conexiones adicionales de banda ancha móvil al crecimiento de otros sectores económicos, la resultante creación de empleo en otras áreas de la economía, y la contribución impositiva generada por los ingresos adicionales originados en los nuevos servicios que son ofrecidos al mercado Contribución indirecta al crecimiento del PIB La investigación sobre la contribución de la banda ancha al crecimiento del producto interno bruto ha generado resultados que indican un impacto positivo. Más allá de los resultados generados con base en investigación de países desarrollados, los autores han generado numerosos estudios en países emergentes con resultados consistentes (ver figura 6-18) Este análisis supone que la estructura de mercado seguirá siendo, en términos generales, la misma. Si existen nuevos entrantes que intenten ofrecer banda ancha móvil como complemento de su línea de productos (operadores de cable, operadores fijos), éstos entrarán en el mercado mediante mecanismos virtuales, como MVNO, adquiriendo servicio al mayoreo de los operadores existentes. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 97

148 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA Estudios realizados sobre la contribución de la banda ancha al crecimiento del PIB Estudio Región/País Impacto en crecimiento del PIB en puntos porcentuales por cada punto adicional de penetración en banda ancha Observaciones estadísticas Koutroumpis (2009) 5 países de la OCDE con penetración superior al 30% 0,023 Coeficiente significativo al 1% 132 Observaciones 8 países de la OCDE con penetración entre 20% y 30% 0,014 Coeficiente significativo al 1% 132 Observaciones 8 países de la OCDE con penetración inferior al 17% 0,008 Coeficiente signficativo al 5% 132 Observaciones Katz et al. (2010a) ondados de desarrollo avanzado en Alemania Condados de desarrollo retrasado en Alemania 0,0256 0,0238 Coeficiente significativo al 1% 214 observaciones Coeficiente significativo al 1% 210 observaciones Qiang et al. (2009) Países de desarrollo medio y bajo 0,138 Coeficiente significativo al 10% Capital humano explica muy poco; la mayor parte de la varianza es explicada por las variables "dummy" América Latina y el Caribe 0,0158 Coeficiente significativo (estadísitico t = 1.98) 49 observaciones Katz (2010b) Brasil Chile 0,008 0,009 Coeficiente no significativos (estadísitico t = 0,16) 27 observaciones Coeficientes significativos (estadísitico t = 2,04) 13 observaciones India 0,031 Coeficientes significativos (estadísitico t = 1,97) 19 observaciones Posible endogeneidad Malasia 0,077 Coeficientes significativos (estadísitico t = 5,81) 15 observaciones China Coeficientes no significativos 30 observaciones Indonesia Coeficientes no significativos 93 observaciones Fuente: análisis TAS En este caso, para evaluar la contribución indirecta del efecto de asignación de espectro en la banda de 700 MHz al crecimiento del PIB, se ha utilizado un modelo econométrico construido para evaluar el impacto de la banda ancha en América Latina y el Caribe. La figura 6-19 presenta las variables utilizadas para el análisis. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 98

149 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA Variables utilizadas para medir el impacto de la banda ancha en el crecimiento económico en América Latina y el Caribe Variable Serie Fuente Observaciones Crecimiento económico Crecimiento del PIB para (2004-6) y (2007-9) Banco Mundial y bancos centrales Variable dependiente Crecimiento en la penetración de banda ancha Crecimiento en la penetración de banda ancha para los períodos ( ) y ( ) Unión Internacional de Telecomunicaciones y agencias nacionales regulatorias Variable independiente Control de nivel de inversión) Promedio inversión/pib para los períodos ( ) y ( ) Banco Mundial Medida para diferenciar niveles de inversión de capital Control de crecimiento de la población Crecimiento de la población para los períodos ( ) y ( ) Banco Mundial Medida para diferenciar el tamaño de la población Control de capital humano Nivel de educación terciaria (2002) Unesco, Earthtrends, University of West Indies, Euromonitor, Government of the Commonwealth of Dominica Medida para determinar diferencias en capital humano Control del nivel desarrollo PIB per cápita en el inicio de los períodos 2003 y 2006 Banco Mundial Medida del punto de partida de crecimiento Control del nivel desarrollo PIB per cápita en el inicio de los períodos 2003 y 2006 Banco Mundial Medida del punto de partida de crecimiento Fuente: análisis TAS Los resultados son los siguientes: FIGURA Impacto de la banda ancha en el crecimiento económico de América Latina y el Caribe Crecimiento PIB Coeficiente Error estándar Estadístico t P>[t] Crecimiento en la penetración de banda ancha para los períodos ( ) y ( ) [Intervalo de confianza al 95%] 0, , ,8 0,54-0, , Promedio inversión/pib para los períodos ( ) y ( ) Crecimiento de la población para los períodos ( ) y ( ) Nivel de educación terciaria (2002) PIB per capita en el inicio de los períodos 2003 y 2006 PIB per capita en el inicio de los períodos 2003 y 2006 Promedio del índice de globalización ( ) y ( ) -0, , ,28 0,82-0, , , , ,32 0,752-3, , , ,93 0,060-0, , , , ,85 0,000-0, , , , ,85 0,000-0, , , , ,34 0,737 0, , Constante 13, ,4659 1,08 0,286-11, ,33982 EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 99

150 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Número de observaciones F(6,42) Prob>F R 2 Root MSE 49 7,18 0,0000 0,3814 7,024 Fuente: análisis TAS Los resultados de la regresión muestran que cuando se controla estadísticamente por el nivel de educación y el punto de partida de crecimiento económico, un aumento de 1% en la penetración de banda ancha contribuye 0,0158 puntos porcentuales al crecimiento del PIB. Con este valor del coeficiente de impacto, se calculó la contribución que ejercería la banda ancha móvil adicional adoptada como consecuencia de la reducción de precios presentada arriba en la sección 6.2. La fórmula empleada establece: Líneas de banda ancha móvil * coeficiente de impacto al PIB = Efecto marginal Considerando las proyecciones de crecimiento del PIB estipuladas por el Fondo Monetario Internacional, se calculó el impacto adicional a ser generado por las nuevas líneas de banda ancha móvil resultantes de la disminucion de precios y de la cobertura adicional (ver figura 6-21). FIGURA Contribución indirecta acumulada de la banda ancha móvil adicional (en miles de millones US $) Argentina Brasil Colombia México Perú Total Fuente: análisis TAS PIB * Crec. BAM+ PIB * Crec. BAM+ PIB * Crec. BAM+ PIB * Crec. BAM+ PIB * Crec. BAM+ PIB * Crec. BAM Total 363,9 0, ,9 0, ,1 0, ,1 0, ,2 0, ,2 0, ,8 0, ,1 0, ,2 0, ,2 0, ,1 0, ,4 0, ,1 0, ,3 0, ,8 0, ,3 0, ,7 0, ,2 0, ,7 0, ,8 0, ,9 0, ,9 0, ,7 0, ,0 0, ,2 0, ,5 0, ,7 0,008 1,177,9 0, ,6 0, ,9 0, ,7 0, ,5 0, ,1 0, ,3 0, ,0 0, ,6 0, ,0 0, ,9 0, ,5 0, ,8 0, ,9 0, ,1 0, ,2 0, ,0 0, ,9 0, ,4 0, ,5 0, ,0 0, ,9 0, ,8 0, ,1 0, ,5 0, ,6 0, ,9 0,239 * Proyección de PIB de acuerdo al crecimiento real proyectado del FMI, a precios del Crecimiento asociado con la penetración adicional de la banda ancha móvil (conexiones a computadoras portátiles) 0,119 0,515 0,073 0,132 0,037 0,876 De acuerdo a la figura 6-22, la contribución acumulada al PIB de la banda ancha adicional resultante de la disminución de precios suma US $877 millones para los cinco países considerados en el estudio. Extrapolando al resto de la región, se esperaría una contribución adicional de US $185 millones. Esto daría un acumulado total para toda América Latina de US $1.062 millones Contribución indirecta a la creación de empleo De manera similar a la contribución al crecimiento del PIB, la banda ancha contribuye al crecimiento del empleo. Nuevamente, este efecto ha sido ampliamente comprobado en numerosos estudios (ver figura 6-22). EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 100

151 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA Estudios sobre la contribución de la banda ancha al crecimiento del empleo Estudio Región/País Impacto en la creación de empleo por cada aumento de 1% en la tasa de crecimiento de la banda ancha Observaciones estadísticas Empleo Desempleo Katz et al. (2010a) Condados de desarrollo avanzado en Alemania 0,0061 Coeficiente significativos al 1% 214 observaciones Shideler et al. (2007) Kentucky, EUA 0,14-5,32 0,008 Coeficientes significativos al 1% (empleo total) Los niveles significativos varían dependiendo del sector industrial analizado 114 observaciones (total). La muestra varía de acuerdo al sector i Brasil -0,0069 Coeficiente significativo (estadístico t = 1,73) 27 observaciones Katz (2010b) Chile 0,00181 Coeficiente significativo (estadístico t = 3,85) 324 observaciones India 0,02825 Coeficientes significativos (estadístico t = 1,86) 19 observaciones Arabia Saudita -0,2434 Coeficientes significativos (estadístico t = -8,29) 13 observaciones Indonesia -8,6163 Coeficientes significativos (estadístico t = -2,58) 99 observaciones Katz (2011) Condados rurales de Estados Unidos -0,1953 Coeficientes significativos al 1% 140 observaciones Fuente: análisis TAS En este caso, para evaluar la contribución indirecta del efecto de asignación de espectro en la banda de 700 MHz al crecimiento del PIB, se ha utilizado un modelo econométrico construido para evaluar el impacto de la banda ancha en el crecimiento del empleo en Chile. Para ello, se construyó un pánel de datos utilizando datos trimestrales por región del país, recopilando información de cada una de las regiones de Chile (excepto la Región Metropolitana por falta de datos trimestrales) desde el 2001 hasta el cuarto trimestre del La base contiene la siguiente información (figura 6-23): Las características específicas de cada región que tienen un impacto en el mercado de mano de obra (sectores industriales, nivel educativo) son controladas por los efectos fijos de los datos del pánel. Los resultados del modelo son los siguientes (ver figura 6-24): EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 101

152 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA Variables utilizadas para medir el impacto de banda ancha en la creación de empleo de Chile Tipo de variable Series Fuente Observaciones Ocupación Tasa de empleo (2002-9) Institutos regionales de estadística Variable dependiente Nivel de producción o actividad económica Índice de actividad económica (2001-9) 179 Institutos regionales de estadística Variable independiente Crecimiento en la penetración de banda ancha Crecimiento en la penetración de banda ancha (2002-9) Subtel Variable independiente FIGURA Chile: Impacto de la banda ancha en la creación de empleo Tasa de ocupación Coeficiente Error estándar Estadístico - t P>[t] Índice de actividad económica (-1) 0, , ,90 0,000 Crecimiento en la penetración de banda ancha (-1) 0, , ,85 0,000 Constante 0, , ,03 0,000 Número de observaciones F(2,310) Prob>F R Fuente: Katz (2010) Así, de acuerdo al modelo, un aumento de 1 por ciento en la penetración de la banda ancha determina 0,18% de aumento en la tasa de ocupación. Con este valor del coeficiente de impacto, se calculó la contribución que ejercería la banda ancha móvil adicional adoptada como consecuencia de la reducción de precios presentada arriba en la sección 6-2. La fórmula empleada establece: Líneas de banda ancha móvil * coeficiente de impacto al empleo = Efecto marginal Considerando las proyecciones del porcentaje de la población económicamente activa empleada generadas por la Organización Internacional del Trabajo y Euromonitor, se calculó el impacto adicional a ser generado por las nuevas líneas de banda ancha móvil (ver figura 6-25) El Índice de Actividad Económica Regional (Inacer) es una síntesis de estadísticas económicas de la región producidas por el INE y otros organismos en el sector productivo. Corresponde a un indicador de tendencia de la actividad económica agregada regional, que busca estimar los ritmos de aceleración o estancamiento en doce de las trece regiones del país, (la Región Metropolitana es excluida). En el Inacer se toman como base los sectores económicos agropecuario, silvícola, pesca, minería, industria manufacturera, electricidad, gas y agua, construcción, comercio, transportes, almacenamiento y comunicaciones, servicios financieros, otros servicios (personales y administración pública) y propiedad de vivienda. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 102

153 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA Creación acumulada de empleos de la banda ancha móvil adicional (en miles de empleos/año) Argentina Brasil Colombia México Perú Total Fuente: análisis TAS Pobl. Empleada* Crec. BAM+ Pobl. Empleada* Crec. BAM+ Pobl. Empleada* Crec. BAM+ Pobl. Empleada* Crec. BAM+ Pobl. Empleada* Crec. BAM+ Pobl. Empleada* Crec. BAM Total , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,390 * Proyección de la población empleada de acuerdo al crecimiento real proyectado de la OIT + Crecimiento asociado con la penetración adicional de la banda ancha móvil (conexiones a computadoras portátiles) 0,753 3,119 0,581 0,670 0,358 5,481 De acuerdo a la figura 6-26, la contribución acumulada al número de empleos / año de la banda ancha adicional resultante de la disminución de precios suma para los cinco países considerados en el estudio y para el resto de la región, dando un total de para toda América Latina Contribución impositiva La contribución tributaria adicional a ser generada por la industria móvil en caso de que ésta se viese atribuida la banda de 700 MHz fue estimada con base en la recaudación marginal sobre las ventas resultantes de la utilización de este espectro. En este sentido, se excluyen pagos de impuestos especiales como tasa radioeléctrica, pagos por derechos e impuestos municipales. Así, los ingresos considerados en esta estimación incluyen a aquellos generados por la venta de líneas adicionales en las nuevas áreas cubiertas en las zonas rurales y las nuevas líneas de banda ancha móvil adquiridas como resultado de la disminución de precios de 10% mencionada arriba en la sección 6.2 (ver figura 6-8). Con base en estos ingresos adicionales, y de acuerdo a la tasa impositiva a las ventas en cada país, la contribución tributaria acumulada de las telecomunicaciones móviles entre el 2012 y el 2020 sería de US $ millones (ver figura 6-26). Extrapolando para el resto de los países de América Latina, se esperaría una recaudación adicional de US $882 millones, llevando el total regional a US $3.420 millones. FIGURA Contribución tributaria adicional acumulada ( ) (en miles US $) Argentina Brasil 180 Colombia México Perú Ingresos Tasa= 21% Ingresos Tasa= 27,5% Ingresos Tasa= 20% Ingresos Tasa= 16% Ingresos Tasa= 20% Total impuestos Total Fuente: análisis TAS 180. El impuesto sobre telecomunicaciones en Brasil se calcula sobre el total de los ingresos brutos, lo que equivale a que la tasa efectiva comparable al resto de los países con una estructura tributaria tradicional es tasa de impuestos / (1 tasa de impuestos). Para la tasa de 27.5% promedio en Brasil, correspondiente al tributo conocido como ICMS ( Imposto sobre circulação de mercadorías e serviços ), se obtiene una tasa efectiva de 37.9%. Como el ICMS varía por estado, se utilizó una tasa promedio (27.5%) y no se consideraron otro impuestos incidentes sobre telecomunicaciones (PIS/Cofins y contribuciones a los fondos de universalización y desarrollo tecnológico). EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 103

154 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA 6.6. Generación de excedente de consumidor El excedente del consumidor creado por la utilización del Dividendo Digital está calculado con base en el beneficio generado por la disminución de precios. Así, el mismo fue calculado con base en la reducción de precios en banda ancha móvil resultante de la reducción de tarifas del 10% (ver figura 6-27). FIGURA Cálculo del excedente del consumidor Disminución de la subscripción mensual de banda ancha de 10% como consecuencia de una reduccion del capex de red P Excedente del consumidor Subscripción mensual de banda ancha móvil: $ (3GB) Nueva subscripción mensual: $ Ingresos Q Incremento de ingresos por la elasticidad de 0.6 Excedente del consumidor = (Abonados a BAM nuevos abonados por elasticidad) * (Precio actual (o voluntad de pago) Nuevo precio) De acuerdo a este concepto, se aplica la disminución de precios a las proyecciones de conexiones de banda ancha móvil (ver figura 6-28). Así, el excedente del consumidor para los cinco países analizados en detalle suma US $3,842 millones. La extrapolación al resto de la región arroja una cantidad adicional de US $1,315 millones, dando un total para América Latina de US $5,157 millones Beneficios Sociales de la banda ancha móvil Más allá de los beneficios económicos cuantificables, la asignación del espectro de 700 MHz a la banda ancha móvil en América Latina podrá tener una contribución social positiva en numerosas áreas. Por ejemplo, al expandir la provisión de banda ancha inalámbrica a zonas rurales, la utilización del espectro permitirá a la población residente en áreas hoy no cubiertas por banda ancha y que, por definición, están desfavorecidas respecto a calidad de vida imperante en medios urbanos, acceder a mayores recursos educativos, mejores servicios de salud y poder recibir servicios financieros. Al mismo tiempo, la banda ancha inalámbrica pasible de ser introducida en zonas rurales permitirá la provisión eficiente de servicios públicos, mejorando la interrelación entre la sociedad civil y la administración. A continuación, detallaremos aspectos específicos de estos beneficios sociales El impacto en la educación América Latina confronta todavía problemas significativos en lo que se refiere al acceso a la educación y desempeño de estudiantes. El 84% de la población en el mundo tiene, en promedio, un desempeño escolar mejor que los países de América Latina (ver figura 6-29). EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 104

155 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA Cálculo del excedente acumulado del consumidor para los cinco países ( ) (en millones US $) Argentina Brasil Colombia México Perú Total Dem. ex DD* Descuento** Excedente Dem. ex DD* Descuento** Excedente Dem. ex DD* Descuento** Excedente Dem. ex DD* Descuento** Excedente Dem. ex DD* Descuento** Excedente Total $ 3,23 $ 3,23 $ 3,23 $ 3,23 $ 3,23 $ 3,23 $ 3,23 $ 3,23 $ 3, $ 4,97 $ 4,97 $ 4,97 $ 4,97 $ 4,97 $ 4,97 $ 4,97 $ 4,97 $ 4, $ 2,75 $ 2,75 $ 2,75 $ 2,75 $ 2,75 $ 2,75 $ 2,75 $ 2,75 $ 2, $ 3,28 $ 3,28 $ 3,28 $ 3,28 $ 3,28 $ 3,28 $ 3,28 $ 3,28 $ 3, $ 4,33 $ 4,33 $ 4,33 $ 4,33 $ 4,33 $ 4,33 $ 4,33 $ 4,33 $ 4, (*) Demanda de conexiones de banda ancha móvil excluyendo la demanda adicional resultante de la posible reatribución de la banda de 700 MHz. (**) Reducción de 10% sobre la tarifa promedio de 3 GB pospago. Fuente: análisis TAS FIGURA Resultado Comparado de la prueba estandarizada PISA (2006) % Uruguay Chile Mexico Argentina Colombia Brasil China India * Argentina, Australia, Austria, Azerbaiján, Bélgica, Brasil, Bulgaria, Canadá, Chile, Taiwán, Colombia, Croacia, República Checa, Dinamarca, Estonia, Finlandia, Francia, Alemania, Grecia, Hong Kong, Hungría, Islandia, Indonesia, Irlanda, Israel, Italia, Japón, Jordania, Corea del Sur, Kyrgyzstan, Latvia, Liechtenstein, Lituania, Luxemburgo, Macao, México, Montenegro, Holanda, Nueva Zelanda, Noruega, Polonia, Portugal, Qatar, Rumania, Rusia, Serbia, Eslovaquia, Eslovenia, España, Suecia, Suiza, Tailandia, Tunisia, Turquía, Reino Unido, Estados Unidos, Uruguay Fuentes: OCDE, FMI EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 105

156 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Entre los problemas mas importantes se pueden mencionar la deserción escolar en medios rurales, y los recursos educacionales limitados en zonas rurales y aisladas. En este contexto, la banda ancha móvil desplegada en base a la utilización de la frecuencia de 700 MHz puede contribuir de manera significativa a resolver algunos de estos problemas. Sin desconocer efectos de otra índole, nuestra investigación (Flores-Roux, 2010) muestra que si la escuela pública está dotada con aula de medios (computadoras con acceso a banda ancha), la probabilidad de desempeñarse mejor que la media de escuelas privadas aumenta de 2,5% a 6,5% (ver figura 6-30). FIGURA México: Distribución de Resultados de Prueba Enlace Públicas Privadas Públicas con aula de medios Probabilidad 6.5 % vs. 2.5% públicas Calificación (2009) Nota: La muestra de escuelas públicas con medios son las escuelas receptoras de donativos de la Fundación Únete, enfocada en la dotación de este tipo de infraestructura a escuelas públicas. Al día de hoy han conectado aproximadamente 6,000 escuelas Fuente: Secretaría de Educación Pública, Antonio Purón (2010) El impacto de banda ancha en educación cubre numerosas áreas de aplicación desde la educación a distancia a la utilización de terminales portátiles y herramientas informáticas en las escuelas. En el caso de la educación a distancia, la investigación muestra, en primer lugar, que alumnos en áreas remotas cuyo vehiculo primario de instrucción es este tipo de plataforma tienden a desempeñarse de manera similar a aquellos que se benefician del medio presencial (instrucción en aula) (Hudson; 2006). Los estudios de Witherspoon et al. (1993) muestran que, pese a la distancia y a la separación física con el maestro, los alumnos de sistemas a distancia tienden a estar más motivados para el aprendizaje, son más maduros mientras que el diseño de material educativo tiende a ser más sistemático y orientado a eficientizar el proceso de enseñanza. En segundo lugar, Hudson muestra que cursos a distancia complementados con programas de tutoría difundidos por telecomunicaciones tienden a reducir la deserción escolar (Hudson, 1990). Asimismo, Lemke et al. (2006) concluyen, después de haber compilado investigación realizada en los últimos veinte años, que existe una relación estadísticamente significativa entre alumnos que usan computadoras y su nivel de desempeño académico. En particular, áreas donde las TIC tienen un impacto positivo son la educación de ciencias naturales, algebra, geometría y semántica gramatical. Finalmente, pese a que los resultados son altamente dependientes de la calidad del equipo y del acceso de telecomunicaciones, cuando estos factores son controlados, el acceso a Internet mediante la banda ancha puede compensar problemas como bajo nivel de EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 106

157 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA capacitación de educadores o falta de material educativo 181 (Puma et al., 2002). De acuerdo a los éxitos puntualizados en la investigación académica, la banda ancha móvil podría proporcionar acceso a herramientas en línea para educación básica. Por ejemplo, el Proyecto K-Nect desplegado en Carolina del Norte (EE.UU.) demostró que el desempeño de estudiantes en pruebas de matemática mejoro 30% a partir del uso de aplicaciones educativas desplegadas gracias a smartphones. A nivel universitario, la banda ancha móvil también podría ser utilizada para ofrecer cursos a distancia que podrían completar la actividad presencial El impacto en los servicios de salud El sector sanitario en América Latina enfrenta numerosos desafíos. En primer lugar, los cambios demográficos, combinados con el envejecimiento progresivo de la población, la migración y la movilidad social representan un reto estructural a la capacidad de entrega de servicios sanitarios. Esto se combina con el desarrollo de nuevas enfermedades y pandemias generadas tanto por el estilo de vida moderno como por la globalización. En este contexto de mayores retos, el sistema sanitario debe responder garantizando su sostenibilidad económica y eficiencia en la entrega de servicios. Los obstáculos son numerosos. En primer lugar, la entrega de servicios sanitarios está organizada en torno a hospitales, centros de salud, centros privados y distribuidores de medicinas, cada uno de los cuales posee su propia información y sistemas a los que no se puede acceder desde el exterior, con lo que cada centro se transforma en una isla informativa. Este problema se magnifica con la movilidad de ciudadanos en el sentido de que la falta de interconexión informativa asume una dimensión geográfica. En segundo lugar, las plataformas tecnológicas para la entrega de servicios son, en muchos casos, anti-económicas en relación con la propuesta de valor de las telecomunicaciones. La banda ancha representa una plataforma indispensable para resolver estos desafíos, especialmente en relación con el paciente, los cuidados clínicos y la gestión administrativa. En términos generales, se pueden identificar dos áreas de acción prioritarias donde la banda ancha ejerce un impacto sustancial en la eficiencia de servicios sanitarios. En primer lugar, la comunicación con pacientes puede ser realizada de manera más eficiente mediante la instalación de registros en línea, la telemedicina (en áreas como la tele-psiquiatría, tele-cardiología, tele-radiología, y tele-cirugía), el tele-monitoreo remoto, y la creación de comunidades y redes sociales de pacientes y profesionales de la salud. La aplicación de la banda ancha móvil en los procesos tanto de atención primaria como especializada tiene un impacto positivo no sólo en la calidad de la asistencia sino también en la flexibilidad de entrega de dichos servicios. Por ejemplo, el programa Alokito en Bangladesh permite la comunicación por video-conferencia basada en banda ancha móvil entre equipos de enfermeras en zonas rurales y doctores sitos en zonas urbanas para la realización de diagnósticos. En segundo lugar, la comunicación inter-profesional tendiente a mejorar el nivel de capacitación y actualización puede ser acrecentada mediante la creación de redes profesionales, programas de formación a distancia y videoconferencia que dependen de banda ancha móvil. Así, los profesionales del sector que residen en zonas rurales o aisladas pueden continuar su formación y actualización en base a plataformas de e-learning. En tercer lugar, la banda ancha móvil puede contribuir, mediante la utilización de redes sociales, a la educación para la prevención de problemas de salud. Por ejemplo, la organización internacional UNAIDS ha estado utilizando redes sociales como Facebook y Youtube para difundir información sobre HIV y SIDA. De la misma manera, se podría distribuir información sobre prevención de enfermedades infecciosas en niños. El impacto de estas acciones es múltiple. A nivel social, se observa una mayor eficiencia en la calidad a la atención de pacientes, la reducción en tiempos de entrega de servicio, un incremento en la calidad de la asistencia, una agilidad en el intercambio de información entre profesionales y el beneficio de su formación continua. A nivel económico, la contribución de la banda ancha al sector sanitario se manifiesta en una mayor eficiencia en gestión de recursos tantos materiales como humanos, una reducción en los costos de atención y movilización, y la consiguiente reducción en costos de manejo de información El impacto en la inclusión financiera La población latinoamericana esta generalmente poco servida por servicios financieros (ver figura 6-31) Véase el estudio de caso del sistema Aprendaris. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 107

158 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA América Latina versus países desarrollados: Inclusión financiera País Cuentas Corrientes por 1,000 adultos(1) Tarjetas de crédito per capita Argentina Brasil Chile Costa Rica Colombia Ecuador El Salvador México Paraguay Perú R. Dominicana Uruguay Venezuela 906, ,35 729, , ,44 569,24 825, ,78 108,74 783,37 529, ,39 0,775 (*) 0,9758 (*) 1,3206 (*) 0,25 0,14 (*) 0,0855 0,1091 0,2364 (*) 0, ,21017 (**) 0,1182 0, ,2186 (*) Países avanzados ,9662(***) Notas: (1) En bancos comerciales (*) 2010 (**) 2008 (***) Promedio de EE.UU., Francia, Italia, España, y Reino Unido. Fuentes: Banco Mundial. Financial Access Factsheet; Euromonitor. Consumer Finance; Mato, M. Balance del sistema Financiero Peruano para 2008; Felaban Como se observa en la figura 6-31, el número de cuentas corrientes en América Latina es significativamente inferior al promedio de países avanzados. Asimismo, con la excepción de Argentina, Brasil y Chile, la adopción de tarjetas de crédito también es relativamente muy baja. Así, una porción importante de la población no está suficientemente informada de las posibilidades de servicios financieros para afrontar gastos personales o no tienen acceso al crédito necesario para desarrollar nuevos emprendimientos. La banda ancha móvil puede ayudar a resolver estos problemas. Terminales con acceso a banda ancha podrían proveer acceso a servicios bancarios a poblaciones en medios rurales, así como servicios de compañías de seguros para afrontar gastos de emergencia. El acceso a crédito podría ayudar al desarrollo de micro-emprendimientos. Asimismo, la banda ancha móvil podría poner a disposición de la población en zonas rurales una gama de aplicaciones financieras como presupuestacion y control de inventario, lo que ayudaría a micro-empresas con base en hogares de zonas rurales El impacto en el acceso a servicios públicos En términos generales, la contribución de la banda ancha a la administración pública se materializa en cuatro dimensiones: Social: la introducción de la banda ancha en el aparato del estado implica una mejora en la calidad de servicio a los ciudadanos en la prosecución de tramites administrativos (por ejemplo, declaración de renta, renovación de licencias, y acceso al aparato administrativo) y un mejoramiento de las relaciones del gobierno con la comunidad EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 108

159 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Administrativa: el impacto de la banda ancha en la administración pública conlleva una simplificación de procesos administrativos, la agilización en la presentación de información, una disminución de la redundancia en procesos administrativos como resultado de la integración de bases de datos, y como consecuencia de ello, un mejoramiento de la 182 imagen de la administración pública Transparencia: la incorporación de la banda ancha a la administración pública permite el flujo libre de información sobre la gestión gubernamental, con el consiguiente incremento en la participación de la sociedad civil en los asuntos públicos y la posible influencia sobre políticas y acciones del gobierno Económica: los impactos económicos de la banda ancha en la administración pública incluyen la posibilidad de reducir costes materiales requeridos para atender a ciudadanos, así como en la gestión misma de servicios, y la posibilidad de aumentar ingresos por recaudación fiscal El impacto significativo de la banda ancha en la eficiencia de la administración pública ha sido demostrado numerosas veces. El gobierno australiano, por ejemplo, ha medido tres tipos de efectos positivos en la administración pública de ese país: Acceso a la administración: 80% de los usuarios de Internet recurren al acceso informático para informarse sobre trámites administrativos. Eficiencia administrativa: 86% de los ciudadanos perciben un beneficio social o económico una vez implantados los sistemas de gobierno electrónico. Costo/beneficio: la adopción de la banda ancha en la administración pública genera ratios del orden de 5:1 con respecto a la inversión destinada la instalación y los beneficios económicos estimados. En este contexto, la banda ancha móvil constituye una tecnología esencial para la provisión de servicios públicos en zonas aisladas o rurales. Así, la población de estas regiones podría acceder a servicios generalmente disponibles en núcleos metropolitanos. Esto resultaría en beneficios desde el punto de reducción de tiempo empleado en trámites y el ahorro de tiempo y energía en desplazamiento a oficinas públicas. FIGURA Impacto social de la banda ancha móvil Área de impacto Ejemplos Radiodifusión Educación Conectividad a recursos educacionales Educación a distancia Salud Telediagnóstico Comunicación interprofesional de salud Información sanitaria Inclusión financiera Acceso a plataformas de micropagos Educación para acceso a microfinanzas Acceso a servicios públicos Acceso a programas de gobierno electrónico Inclusión informativa Acceso a información estatal Programas de relación enrre ciudadanos y gobierno Fuente: análisis TAS Impacto Alto Impacto Nulo 182. Véase el estudio de caso de ChileCompra en Chile. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 109

160 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Conclusión sobre los beneficios sociales de la banda ancha móvil El análisis de áreas de aplicaciones realizado arriba nos permite concluir que la banda ancha móvil representa una plataforma más eficiente para la provisión de servicios que aumentan el bienestar de los ciudadanos (ver figura 6-32) Conclusión En resumen, si la banda de 700 MHz fuese utilizada para prestar servicios de banda ancha móvil se genera valor en áreas como contribución directa a la economía derivada de la oferta adicional de banda ancha móvil y cobertura rural, creación de empleo, impacto en el ecosistema de las telecomunicaciones como resultado de la adquisición de espectro, la construcción y operación de redes de y gastos en servicios comerciales, además de una contribución impositiva adicional. Los valores para los cinco países estudiados en detalle son los siguientes (ver figura 6-33). La siguiente sección evalúa el impacto económico de mantener la atribución de la banda de 700 MHz a la radiodifusión para su uso en la televisión. FIGURA América Latina: Impacto económico acorde con el uso de 700 MHz para telecomunicaciones móviles (en millones de US $ excepto empleos) Área de impacto Argentina Brasil Colombia México Perú Resto de América Latina Total América Latina Contribución al ecosistema 183 Adquisición de espectro Bienes productivos Servicios operativos Servicios comerciales ,9 12, ,7 61, ,6 28, ,8 5, Contribución directa 184 Ingresos adicionales Empleos directos Contribución indirecta 185 Contribución al PIB Empleos indirectos Contribución impositiva Excedente del consumidor Fuente: análisis TAS 183. Contribución anual no recurrente ("one-time cost") Contribución anual recurrente ("ingresos anuales" y "empleos-año) Contribución acumulada de ocho años Contribución acumulada de ocho años. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ATRIBUCION DE ESPECTRO A LA BANDA ANCHA MÓVIL 110

161 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA 7. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN La atribución de la banda de 700 MHz a la radiodifusión para su uso en la transmisión de señales de televisión también resulta en una serie de efectos positivos para las sociedades latinoamericanas. De la misma manera que en el capítulo anterior, los efectos han sido categorizados en dos áreas: 1) el impacto en el ecosistema de la industria en términos de gastos a incurrir para hacer posible la oferta de señales de televisión digital en la nueva banda, y 2) el impacto económico y social. Antes de evaluar cuantitativamente el impacto de esta política, comenzaremos pasando revista a la situación de la industria de televisión en América Latina Situación actual de la industria de televisión en América Latina Los ingresos totales anuales de la industria de televisión en los cinco países estudiados en detalle son US $ millones, lo que representa 0,56% del PIB (ver figura 7-1). El componente más importante de los ingresos televisivos es la publicidad de televisión terrestre. Esta ultima suma US $ millones en los cinco países estudiados en detalle, esperándose que llegue a US $ millones en 2020 (ver figura 7-2). La suscripción de televisión de paga representa la siguiente categoría de ingresos en términos de volumen, sumando US $ 8,497 millones, y esperándose que alcance US $ 17,530 millones en 2020 (ver figura 7-3). La publicidad en la televisión paga representa la siguiente categoría de ingresos en términos de volumen. Ésta suma en estos momentos US $ 1,377 millones, esperándose que llegue a US $ 4,987 millones en 2020 (ver figura 7-4). FIGURA 7-1. América Latina: Ingresos de la industria de la televisión (en millones de US $) País Argentina Televisión PIB , Brasil Ingresos/PIB Televisión PIB Ingresos/PIB 0,68% ,51% 0,76% ,56% 0,83% ,52% 0,87% ,53% Colombia Televisión PIB Ingresos/PIB ,79% ,85% ,82% ,81% México Televisión PIB Ingresos/PIB ,44% ,56% ,52% ,50% Perú Televisión PIB Ingresos/PIB ,16% ,20% ,26% ,32% Total Televisión PIB Ingresos/PIB 0,51% 0,59% 0,56% 0,56% Fuentes: Pricewaterhouse Coopers; FMI; análisis TAS EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 111

162 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA 7-2. América Latina: Ingresos de la publicidad de televisión terrestre (en millones de US $) Argentina Brasil Colombia México Perú Total Fuentes: PricewaterhouseCoopers; Wilkofsky Gruen Associates; Urbano et al. (2005); análisis TAS FIGURA 7-3. América Latina: Ingresos de abono de TV por subscripción (en millones de US $) Argentina Brasil Colombia México Perú Total Fuentes: PricewaterhouseCoopers; Wilkofsky Gruen Associates; Urbano et al. (2005); análisis TAS FIGURA 7-4. América Latina: Ingresos de publicidad en televisión de paga (en millones de US $) Argentina , Brasil , Colombia , México , Perú. Total , Fuentes: PricewaterhouseCoopers; Wilkofsky Gruen Associates; Urbano et al. (2005); análisis TAS (*) Números difíciles de estimar segmentados de los ingresos por suscripción Finalmente, el servicio de video bajo demanda (video on demand), sin contar el mercado peruano, está generando tan sólo US $2,57 millones en los cinco países estudiados (ver figura 7-5). En resumen, con ingresos totales de US $ 19,890 millones en los cinco países estudiados, la televisión representa 0,56% del PIB agregado. Al 2015, este negocio alcanzará US $ millones, lo que representará 0,5% del PIB de los cinco países. A continuación se analizara en detalle la situación de la televisión abierta en cada uno de los países estudiados en detalle. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 112

163 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA 7-5. América Latina: Ingresos de Video on Demand (en millones de US $) Argentina , Brasil , Colombia , México , Perú (*). Total , Fuentes: PricewaterhouseCoopers; Wilkofsky Gruen Associates; Urbano et al. (2005); análisis TAS (*) Números difíciles de estimar segmentados de los ingresos por suscripción La televisión terrestre en Argentina La televisión abierta en Argentina incluye estaciones públicas y privadas. El principal canal estatal, Canal 7, transmite a partir de 293 repetidoras a todo el país. Entre los otros canales estatales se cuenta con el canal Encuentro, que es una emisora educacional operada por el Ministerio de Educación, Ciencia y Técnica. Los canales privados incluyen América TV, TeleArte (canal 9), Telefe (canal 11) y Artear (Canal 13). Más allá de las estaciones nacionales, existen un número importante de estaciones de bajo poder operando en todo el país (ver figura 7-6). El ingreso de publicidad de la televisión abierta en Argentina llega a US $ 759 millones, habiendo crecido a una tasa compuesta de 9,9% desde el Al ingreso de la publicidad de la televisión abierta se suman las ventas publicitarias de la televisión por subscripción, la que alcanza en el 2011 US $ 184 millones. La serie histórica desde el 2005 al 2011 y una proyección muestran que el gasto publicitario como proporción del PIB es relativamente estable (ver figura 7-7). FIGURA 7-6. Estaciones de televisión abierta por región Región Estaciones públicas Estaciones privadas AMBA (Buenos Aires) Cuyo Nordeste Pampas Patagonia Fuente: Aegis (2010) Nota: algunas de la estaciones no cubren la región entera FIGURA 7-7. Argentina: La publicidad televisiva versus PIB ( ) (en US $ millones) TV abierta TV subscripción Total PIB Publicidad/PIB Fuentes: PricewaterhouseCoopers; Wilkofsky Gruen Assoc; Banco Mundial; FMI; análisis TAS EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 113

164 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA La serie histórica indica que el volumen publicitario como proporción del PIB se mantiene a un nivel relativamente estable. Más allá de la publicidad, existen estaciones de televisión terrestre que cobran una subscripción por la señal inalámbrica usando frecuencias en la banda de UHF. Por ejemplo, el operador Antina transmite 72 canales en el área entre Campana y La Plata (excluyendo Buenos Aires) por un costo de a$ 94,90 (abono mensual) y a$ 24,00 por canal Premium como HBO. El plan de digitalización de la televisión ya ha sido establecido y está siendo implementado. Al igual que Brasil, en julio de 2009, Argentina decidió adoptar el estándar japonés (ISDB-T). El gobierno argentino ha tomado el liderazgo del despliegue 187 de la Televisión Digital a través de un plan de rápido despliegue de las antenas en las principales localidades del país (50 antenas en la primera etapa, con un plan total de 400 antenas). Adicionalmente el gobierno ya ha comenzado a distribuir decodificadores a ciertos sectores, lo que podrá acelerar el proceso de digitalización. El plan incluye la provisión de TV Digital Satelital en escuelas rurales. Los primeros canales asignados a la Televisión Digital han sido los canales reservados para el Estado, que son los canales de UHF 22 a 25. A los radiodifusores privados también se le han entregado frecuencias por debajo del canal 40. En junio de 2011 se determinó que a las provincias se les asignaron canales entre el 20 y el 31, a las universidades del interior del 188 país, entre el 31 y el 34. Sin embargo, la Autoridad Federal de Servicios de Comunicación Audiovisual (AFSCA) también hizo una asignación de los canales analógicos del 55 al 66 en la Ciudad y Provincia de Buenos Aires a 15 universidades. Se cree que esta asignación es temporaria, ya que de persistir dificultaría el uso del espectro del Dividendo Digital para la promoción de la Banda Ancha Móvil, tal como está explicitado en los planes de gobierno y la propia Ley de Servicios de Comunicación Audiovisual. En julio de 2011 fue lanzado el concurso público de 220 licencias de televisión digital terrestre locales, entre 4 y 8 por en un total de 32 zonas geográficas, en definición estándar. Las señales serán asignadas a organizaciones sin fines de lucro y a empresas con fines comerciales a razón de uno a uno. Todos los licenciatarios contratarán los servicios de uso de infraestructura, multiplexado y trasmisión (denominada la Plataforma Nacional de TDT) de la Empresa Argentina de Soluciones Satelitales Sociedad Anónima (ARSAT), propiedad del Estado. Se han establecido precios de referencia de a pesos (US$ US$ ) como pago de una vez más un pago recurrente de pesos (US$ 5.750) por mes por el servicio de multiplexeo. Cada múltiple dará cabida a 4 señales de televisión La televisión terrestre en Brasil La mayor parte de las estaciones de televisión terrestre en Brasil está en manos privadas. Existen siete redes terrestres - Red Globo, Red Bandeirantes, Red Gazeta, Sistema Brasileiro de Televisão, Red TV!, MTV Brasil, y Red Record - que son transmitidas en la gran parte de los centros urbanos. Al mismo tiempo, Brasil cuenta con un número significativo de estaciones locales y regionales, especialmente en las grandes ciudades. Las siete redes nacionales operan en frecuencias dedicadas de acuerdo a la figura 7-8. FIGURA 7-8. Frecuencias de televisión para las redes nacionales en Sao Paulo Analógica Digital Red Record Red Globo Red Bandeirantes Red TV! SBT MTV Brasil Red Gazeta Fuente: Aegis (2010) Resoluciones 687/2011 y 689/2011 de la AFSCA Dependiendo del fin y de la localidad. Fuente: Resoluciones 685/2011, 686/2011 y 812/2011 de AFSCA. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 114

165 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA El Sistema Brasileiro de Televisão Digital (SBTVD) fue creado 190 por decreto en 2003; su plan de implementación fue establecido, también por decreto, en 2006, y reglamentado por el Ministerio de las Comunicaciones. Los planes contemplan el calendario de transmisión en analógica y digital y la fecha de apagón analógico, así como el mantenimiento de canales de 6 MHz por licenciatario, con el objetivo de que se puedan explotar cabalmente los beneficios del estándar. El Plano Básico de Distribui ão de Canais de TV Digital (PBDTV) consiguió viabilizar canales en 296 localidades brasileiras. De acuerdo a los cronogramas establecidos, actualmente Anatel ha colocado en consulta pública algunas de las piezas de esta implementación, con el objetivo de alterar los planos básicos en diferentes estados. Las reglas originales contemplan el que cuatro canales sean reservados para el Estado Poder Ejecutivo, educación, cultura y ciudadanía. El estándar del SBTVD es el japonés (ISDB-T) con algunas variaciones. Como se ha comentado anteriormente, se usará tanto la banda de VHF (parte alta) como la banda de UHF, ya que durante el período de transición todos los canales digitales se otorgarán en la banda de UHF debido a la alta utilización de la banda de VHF y para facilitar la recepción de terminales móviles. Una de las variaciones respecto al ISDB-T original es la utilización de Ginga, que es el middleware abierto de desarrollo local y que permite la existencia de aplicaciones interactivas para la TV digital independiente del equipo. Actualmente ya se instala en varios modelos de televisión, pero existe presión por parte de la industria de software de que la incorporación del software se vuelva obligatoria. La transición está siendo coordinada por el Foro Brasileño de Televisión Digital (Foro SBTVD) y el gobierno. No parece estar sufriendo mayores retrasos. A la fecha ya han sido inaugurados 102 canales digitales en 480 municipios. Con respecto a los aspectos económicos de la televisión abierta comercial, el ingreso de publicidad en Brasil llega a US $ millones, habiendo crecido a una tasa compuesta de 8,6% desde el Al ingreso de la publicidad de la televisión abierta se suman las ventas publicitarias de la televisión por subscripción, la que alcanza en el 2011 US $ 569 millones. La serie histórica desde el 2005 al 2011 y una proyección muestran que el gasto publicitario como proporción del PIB oscila entre 0,26% y 0,34% (ver figura 7-9). FIGURA 7-9. Brasil: La publicidad televisiva versus PIB ( ) (en US $ millones) TV abierta , TV subscripción Total PIB Publicidad/PIB Fuentes: PricewaterhouseCoopers; Wilkofsky Gruen Assoc; Banco Mundial; FMI; análisis TAS Pese a la estabilidad relativa del gasto publicitario respecto del PIB, se debe notar también que el gasto publicitario por televisión creció casi un 10% entre el 2010 y el 2011, potenciado por el crecimiento económico. El plan de digitalización de la televisión ya ha sido establecido y esta camino a su implantación. Brasil ha adoptado el estandard japonés (ISDB-T) para la televisión digital terrestre. La televisión digital usará tanto la banda VHF como la de UHF; durante el período de transición todos los canales digitales serán ofrecidos en la banda de UHF debido a la alta utilización de la banda de VHF. Las áreas de servicio digital son similares a las de servicio analógico. La transición está siendo coordinada por un Foro de SBTVD. Canales digitales ya han sido inaugurados en 27 ciudades; Sao Paulo y Rio de Janeiro ya tienen 22 canales cada una Decreto del 26 de noviembre de Decreto del 29 de junio de Portaria MC 652/2006 del 10 de octubre de EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 115

166 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA La televisión terrestre en Colombia La industria de la televisión colombiana está compuesta de dos canales privados (Caracol y RCN TV) y tres de propiedad estatal (Canal Uno, Canal Institucional y Señal Colombia), todos transmitiendo en sistema analógico. Por otro lado, existen ocho cadenas regionales y 48 canales locales. La Red Televisión Nacional de Colombia (RTVC) es una entidad pública independiente que produce, programa y emite los canales públicos de la televisión colombiana (Canal Uno, con carácter público-comercial; Señal Colombia, con carácter cultural y educativo; y Canal Institucional, con carácter oficial), además de estaciones de radio nacional (Radio Nacional de Colombia y Radiónica). Por otro lado, los canales privados no son quienes directamente operan su red de transmisión a nivel nacional, sino que la contratan a un consorcio conformado por RCN y Caracol denominado Consorcio de Canales Nacionales 193 Privados (CCNP). Los canales nacionales, y en menor medida los regionales, operan compartiendo la infraestructura de transmisión. El plan de digitalización de la televisión ya ha sido establecido y está en proceso de implementación. Colombia eligió en 2008 el estándar europeo (DVB/T / MPEG-4) para los servicios de televisión digital terrestre, aunque el tema, aparentemente cerrado, fue discutido de nuevo durante La Ley de Telecomunicaciones y TIC ha llevado a la creación de una agencia nacional par a la gestión del espectro radioeléctrico (Agencia Nacional del Espectro ANE), con mayor poder que la disuelta Comisión Nacional de TV, el regulador de medios de comunicación. El plan de digitalización de la televisión establece como fecha de apagón analógico el Las primeras emisiones de televisión digital comenzaron en Bogotá en enero 29 del 2010 con las tres señales de la RTVC. Durante el 2010, cuatro sitios de transmisión adicionales fueron inaugurados extendiendo el servicio de televisión digital terrestre a Cali, Medellín, Antioquia, Barranquilla y Santa Marta; se está en proceso de invertir 20 millones de dólares adicionales durante el 2011 para el aumento de la cobertura a través de 14 estaciones. En diciembre del 2010, las redes privadas Caracol y RCN comenzaron la transmisión digital en Bogotá y Medellín. El plan de reorganización de frecuencias (Resolución de 2010), a pesar de haber sufrido modificaciones, está llevándose a cabo, con la migración a la banda de 470 a 512 MHz. En enero de 2011se dio un paso más, al haber sido publicado un anteproyecto sobre las condiciones técnicas para la prestación de la TV digital. El acuerdo reconoce la posibilidad de transmisión en los 67 canales de VHF y UHF (del 2 al 69), pero deja claro que la banda de 700 MHz está atribuida a otros servicios. Es decir, el plan de digitalización avanza, pero, dada la estructura del sector y la ocupación intensiva del espectro, es posible que sufra aún más alteraciones y retrasos. Con respecto a los aspectos económicos de la televisión abierta colombiana, el ingreso de publicidad de la televisión abierta llega a US $ 698 millones, habiendo crecido a una tasa compuesta de 6,8% desde el Al ingreso de la publicidad de la televisión abierta se suman las ventas publicitarias de la televisión por subscripción, las que alcanzan en el 2011 US $ 264 millones. La serie histórica desde el 2005 al 2011 y una proyección muestra el gasto publicitario como proporción del PIB (ver figura 7-10). En este caso, se puede observar una tendencia decreciente del gasto publicitario como proporción del PIB: en el 2005 este alcanzaba 0,4850%, mientras que en el 2009, el porcentaje cayó a 0,3749%; la proyección al 2014 estima que esta proporción bajara a 0,3487%. FIGURA Colombia: La publicidad televisiva versus PIB ( ) (en US $ millones) TV abierta TV subscripción Total ,058 1,156 1,266 PIB , , Publicidad/PIB 0,4850 0,3814 0,4117 0,3667 0,3749 0,3180 0,3125 0,3252 0,3360 0,3487 Fuentes: PricewaterhouseCoopers; Wilkofsky Gruen Assoc; Banco Mundial; FMI; análisis TAS 193. Ministerio de Comunicaciones de Colombia, Resumen Ejecutivo. Impacto socioeconómico de la implementación de la televisión digital terrestre en Colombia, agosto EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 116

167 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA La televisión terrestre en México México cuenta con dos redes nacionales de televisión (Televisa y TV Azteca) y un número de servicios regionales. Televisa opera con cuatro canales - 2 (El canal de las estrellas), 4 (Foro TV), 5 (Canal 5) y 9 (Galavisión). TV Azteca opera dos estaciones (Azteca 13 y Azteca 7). La cobertura 194 de todos estos canales es prácticamente nacional. Las estaciones regionales cubren generalmente estados. Así, en cada estado, dependiendo de la cobertura que puede ser alcanzada, pueden existir un número de frecuencias atribuidas a cada operador. Además de estos, existen varios operadores locales; 38 están operando bajo la figura de permiso. Asimismo, México cuenta con un canal que está buscando cobertura nacional (Canal 11), en manos del Instituto Politécnico Nacional (IPN), una universidad estatal. El ingreso de publicidad de la televisión abierta en México suma US $ millones, habiendo crecido a una tasa compuesta de 1,8% desde el Al ingreso de la publicidad de la televisión abierta se suman las ventas publicitarias de la televisión por subscripción, las que alcanzan en el 2011 US $ 330 millones. El volumen publicitario fue afectado por la crisis económica del 2008 con recuperación progresiva después del La crisis afectó más la publicidad de TV terrestre que la de TV por subscripción. Se proyecta que la publicidad por TV terrestre aumentará con Olimpiadas (2012) y la Copa del Mundo (2014). Por otra parte, la aprobación de COFETEL del servicio HD para TV Azteca genera un aumento en la publicidad en servicios multicanales por subscripción. La serie histórica desde el 2005 al 2011 y una proyección muestra el gasto publicitario como proporción del PIB (ver figura 7-11). FIGURA México: La publicidad televisiva versus PIB ( ) (en US $ millones) TV abierta TV subscripción Total PIB Publicidad/PIB , , , , , , , , ,2462 2, ,2436 2, ,2425 Fuentes: PricewaterhouseCoopers; Wilkofsky Gruen Assoc; Banco Mundial; FMI; análisis TAS Al comparar el gasto publicitario televisivo mexicano con el de los otros países latinoamericanos se observa que México registra una intensidad publicitaria inferior a la de Brasil, y Colombia. La transición de la televisión abierta analógica a la TV digital comenzó a implementarse en el Después de la decisión de adoptar la norma ATSC, México definió un plan de transición a 20 años. La transición fue estructurada en seis fases definidas por geografía, en las que se puede migrar una vez que 20% de cobertura digital y repetición de 90% de la señal analógica ha sido alcanzado. Teóricamente, el plan estipula que se debería estar en la mitad de la fase 3, lo que implica 20% de cobertura en el D.F., Monterrey, Guadalajara, Tijuana, Medical, Ciudad Juárez, Nuevo Laredo, Matamoras y Reins, transmisión comercial en por lo menos dos de estas ciudades, y repetición de 90% de la señal analógica en todas. Hasta el momento, sólo un número limitado de estaciones han sido activadas, de acuerdo a las primeras dos fases. La Secretaría de Comunicaciones y Transportes ha comenzado a trabajar en la transición de la televisión digital. El Presidente de México confirmó que el apagón analógico deberá ser concluido en el 2015, seis años antes que lo planeado. El Presidente de la COFETEL, mencionó que el gobierno estaba dispuesto a otorgar un subsidio de US $60 para cada hogar para la migración a servicios digitales, comenzando en el año 2011; incluso llegó a ser presupuestado. Asimismo, se ha creado una comisión integrada por los ministerios de comunicaciones, interior, finanzas, economía y educación y el presidente de la COFETEL para coordinar las tareas de migración de la televisión a servicios digitales. Sin embargo, los nuevos planes, establecidos por decreto, han sido detenidos por el Poder Judicial y el Ejecutivo no ha emitido planes nuevos. Recientemente se ha anunciado que se licitarán dos cadenas nacionales de TDT, pero los estudios requeridos por ley aún no han sido concluidos. Se espera que se presente un plan de implementación de TDT en el último trimestre de El canal 4 opera en modalidad local, pero la cobertura es superior al 80% de la población. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 117

168 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA La televisión terrestre en Perú Actualmente existen 934 estaciones autorizadas para la transmisión de televisión por señal abierta en las frecuencias VHF y UHF. De estas, 73% están en manos privadas, mientras que 27% pertenecen al estado. A nivel nacional, el sistema de televisión abierta peruano cuenta con siete canales privados (Andina TV, Frecuencia Latina, América TV, y Panamericana) y un canal estatal (Televisión Nacional de Perú). Se estima que también existen trece canales privados locales cuya señal solo llega a una localidad o provincia determinada. De éstos, se encuentran canales religiosos, municipales, y universitarios, la mayoría de los cuales emiten su señal en UHF. La transición a la televisión digital terrestre ha venido 195 avanzando. Perú fue dividido en cuatro territorios y se planteó una implementación progresiva. Una primera convocatoria para la asignación de 3 canales, en julio de 2010, terminó declarándose desierta. Sin embargo, ya han sido adjudicados 11 de los 21 canales disponibles (7 por territorio). Lima cuenta con 5 canales transmitiendo. Se ha planteado que cerca del 50% de las transmisiones deberán ser en alta definición. A pesar de que ha habido problemas con el financiamiento prometido por Japón durante el proceso de adopción del estándar ISDB-T, las televisoras han ido lanzando los servicios conforme a los cronogramas establecidos en el Plan Maestro de Implementación de la Televisión Digital Terrestre. Los ingresos por la venta de espacios publicitarios de la televisión abierta en Perú suma US $ 301 millones, habiendo crecido a una tasa compuesta de 6,0% desde el Al ingreso de la publicidad de la televisión abierta se suman las ventas publicitarias de la televisión por subscripción, las que alcanzan en el 2011 US $ 30 millones. La serie histórica desde el 2006 al 2011 y una proyección muestra el gasto publicitario como proporción del PIB (ver figura 7-12). FIGURA Perú: La publicidad televisiva versus PIB ( ) (en US $ millones) TV abierta TV subscripción Total PIB Publicidad/PIB 0,1478 0,1548 0,1616 0,1646 0,1812 0,1972 0,2189 0,2434 0,2716 Fuente: Urbano et al. (2005); Banco Mundial; FMI; análisis TAS El gasto publicitario televisivo como proporción del PIB peruano ha crecido constantemente desde el 2006, proyectándose que alcanzara niveles comparables con el de los otros países estudiados. Este crecimiento se fundamenta en el crecimiento del consumo de artículos del hogar y bienes personales, las elecciones regionales, municipales y nacionales y en un aumento de tarifas publicitarias resultante de medidas tomadas por los canales para resolver un desfasaje tarifario. Ahora se procederá a analizar cuál será el impacto económico y social si la televisión continuara contando con la banda de 700 MHz en América Latina Contribución al ecosistema Como se mencionó arriba, si la banda de 700 MHz continuase atribuida a la radiodifusión y se utilizara principalmente para el desarrollo de la TV digital terrestre, esto resultaría en una serie de contribuciones de la misma al ecosistema de la industria (ver figura 7-13). Bajo este esquema, suponemos que la radiodifusión no contribuirá al tesoro nacional por pago de licencias dado que este espectro en algunos casos ya ha sido asignado industria servicios de TV y porque de haber nuevas asignaciones, éstas suelen no tener contraprestación alguna por el uso del espectro. No se conocen casos anteriores donde el gobierno haya licitado espectro a la radiodifusión. Sin embargo, algunos países están 196 considerando la asignación a través de licitación, pero no es claro que así será, especialmente porque existe una tendencia clara a dar un mayor peso a criterios no económicos. La siguiente contribución al ecosistema incluye la inversión para la adquisición de bienes necesarios para desplegar una red de retransmisoras destinadas a transmitir las nuevas señales de 195. Territorio 1 (Lima y Callao), Territorio 2 (Arequipa, Cusco, Trujillo, Chiclayo, Piura y Huancayo), Territorio 3 (Ayacucho, Chimbote, Ica, Iquitos, Juliaca, Pucalpa, Puno y Tacna); el resto del Perú fue asignado al Territorio Obviamente, las radiodifusoras pagan por la concesión del servicio u obtención de la licencia. Sin embargo, esta contribución no es contabilizada en este estudio porque tampoco fue incluida en el caso de caso del sector de las telecomunicaciones que también requiere de una licencia. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 118

169 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA Contribución al ecosistema de la radiodifusión Inversión para la adquisición de espectro en subasta Sitios Equipamiento de acceso Red troncal Sistemas (OSS) Ingeniería civil Otros bienes Mantenimiento y reparación Distribución Logística Otros servicios Aplicaciones Publicidad Integración de sistemas comerciales (CRM, facturación, etc) televisión digital en la banda de UHF, arriba del canal 51. Esto incluye torres, transmisores, caminos de acceso a las retransmisoras y bienes relacionados con la infraestructura. Esta contribución al ecosistema es independiente de la fuente de los recursos financieros, ya sea el Estado (como el caso de Argentina) o la iniciativa privada. Las dos siguientes etapas en la cadena de valor incluyen la adquisición de servicios operativos y comerciales necesarios para el lanzamiento de señales de televisión digital adicionales en la nueva banda. Estos efectos han sido analizados a lo largo de la cadena de valor en dos áreas discretas: Contribución a industrias como la construcción, electrónica y de equipamiento al invertirse en la construcción de una nueva red en el espectro asignado; Adquisición de servicios operativos y comerciales para poder ofrecer señales digitales de televisión Adquisición de bienes productivos La adquisición de bienes productivos necesarios para el lanzamiento de nuevas señales digitales adicionales en la banda de 700 MHz suma US $ 1.214,1 millones en los cinco países estudiados en detalle. Esta suma ha sido estimada con base en los siguientes presupuestos (ver figura 7-14). Para el resto de los países de América Latina serían necesarios US $ 333,5 millones, dando un total regional de US $1.548 millones Adquisición de servicios operativos y comerciales La adquisición anual de servicios operativos necesarios para el lanzamiento de señales digitales de televisión adicionales en la banda de 700 MHz suma US $174,8 millones en los cinco países estudiados en detalle, mientras que la adquisición de servicios comerciales ha sido estimada en $ millones. Estas sumas han sido estimadas con base en los siguientes presupuestos (ver figura 7-15). Esta categoría incluye los efectos sobre la oferta y la demanda. En la primera dimensión, incluimos las contribuciones directas (contribución al PIB con base en el ingreso generado por la oferta de nuevas señales digitales de televisión en términos de suscripción y publicidad, así como la creación de fuentes de trabajo en el sector, incluyendo empleo en las empresas de televisión y sus proveedores de contenido). En el caso de la atribución de espectro a la radiodifusión consideramos que los efectos indirectos (externalidades positivas que resultan en una contribución al crecimiento del PIB por el impacto de la tecnología en otros sectores de trabajo y la correspondiente creación de empleo) no existen en la medida de que estos son capturados en tanto efectos directos (por ejemplo, necesidad de EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 119

170 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA Estimación de costos de adquisición de bienes productivos por la radiodifusión Argentina Brasil Colombia México Perú Total Número de señales digitales Cantidad de retransmisoras (nacional) Razón señales/retransmisoras Número de retransmisoras Costo por retransmisora (en miles de US $) $ 500 $ 500 $ 500 $ 500 $ 500 Costos productivos totales (en millones de US $) $ 93,0 $ 834,0 $63,0 $ 135,0 $ 90,00 $ 1.215,0 Fuente: análisis TAS FIGURA Estimación de costos de adquisición de servicios operativos y comerciales por la radiodifusión (en miles de US $) Argentina Brasil Colombia México Perú Total Número de señales digitales Cantidad de retransmisoras Razón señales/ Número de retransmisoras Electricidad por retransmisora $ 60 $ 60 $ 60 $ 60 $ 60 Mantenimiento por retransmisora $ 12 $ 12 $ 12 $ 12 $ 12 Total $ 72 $ 72 $ 72 $ 72 $ 72 Costos operativos totales $ $ $ $ $ $ Costos comerciales por señal/mes $ $ $ $ $ Costos comerciales totales $ $ $ $ $ $ Fuente: análisis TAS FIGURA Beneficios socio-económicos Contribución directa a la economía Contribución incremental al PIB Empleo directo generado Efectos de la oferta Contribución al crecimiento del PIB Contribución indirecta a la economía Empleo indirecto generado Impuestos adicionales Efectos de la demanda Excedente del consumidor Competitividad 197. Se están adjudicando 220 licencias en 34 áreas (4 u 8 dependiendo del área; 16 en Gran Buenos Aires). Esto da un promedio de licencias a nivel nacional. A pesar de que en algunas áreas se están licitando sólo 4, no existen limitaciones para que en un futuro sean adjudicadas hasta 8 por área. Un múltiple en Argentina da lugar a cuatro señales. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 120

171 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA contenidos genera un impacto directo en la industria de producción pero no crea un impacto indirecto en otros sectores económicos. Por el lado de la demanda, existe la generación de un excedente del consumidor como resultado de la oferta de nuevas señales. Este capítulo pasa revista a cada uno de los beneficios incluyéndose una estimación cuantitativa para cada uno de los cinco países estudiados en detalle. Si la banda de 700 MHz continuase asignada a los servicios de radiodifusión existente o al despliegue de nuevos servicios de TV digital, ésta contribuiría directamente a las economías latinoamericanas, tanto en términos de crecimiento de ingresos resultantes de la oferta de bienes y servicios de publicidad y suscripción de señales digitales premium, como por la correspondiente creación de empleo. Ambos efectos se analizan a continuación Contribución directa al PIB La contribución directa de la televisión al PIB debe ser estimada en términos de los ingresos a ser generados por suscripción y publicidad. Ahora bien, es importante considerar que, de acuerdo al modelo previsto de televisión digital en Brasil y Argentina, éste se plantea principalmente para señales de organizaciones sin fines de lucro o de tipo social como Universidades, Estatales, ONGs, donde no existen ingresos por subscripción; los ingresos publicitarios, en caso de existir, 198 tienden a ser relativamente marginales. En ninguno de los países estudiados, a pesar de que ya existe desde hace 199 décadas oferta de canales ligados al gobierno, en caso de que capturen recursos publicitarios, éstos son mínimos. A manera de ejemplo, puede tomarse México, en donde menos del 1% de la publicidad de televisión abierta está destinado a 200 este género de canales. Es de esperarse que con una mayor oferta exista una mayor pulverización de la audiencia, lo que se verá reflejado en una mayor dificultad para capturar recursos provenientes de la publicidad; esta situación podrá cambiar sólo en el caso de que consigan ratings considerables, lo que es poco probable. Estos canales, por lo tanto, estarán sustentados con transferencias de recursos, la mayoría de ellos 201 públicos, aunque algunos provendrán de donaciones privadas. Por lo tanto, pese a que en Argentina ya existe un modelo de televisión digital por subscripción, dada la reciente adjudicación de canales por arriba del 52 a 15 universidades 202 de Capital Federal y del Gran Buenos Aires, se decide no considerar ingresos publicitarios o por subscripción. Lo mismo se considera en el caso de Brasil. Utilizando como supuesto que los ingresos por suscripción y por publicidad se ubicarán en niveles similares a los observados actualmente (ver sección 7.1), estimamos que los ingresos 203 anuales de las señales, ya sea por publicidad o suscripción, en Colombia, México y Perú podrán sumar US $282,40 millones (ver figura 7-17). En el resto de los países de la región estos ingresos podrán ascender a US $231 millones, dando un total regional de US $ 513,4 millones. FIGURA América Latina: Contribución directa anual de la televisión en 700 MHz al PIB (en millones de US $) País Número de señales digitales de televisión Suscripción Publicidad Ingresos totales Colombia México Perú Resto de América Latina Total $ 26,8 $ 54,6 $ 46,4 $ 95,0 $ 26,6 (*) $ 90,0 $ 38,0 $ 136,0 $ 53,4 $ 144,6 $ 84,4 $231,0 $ 513,40 Fuente: análisis TAS (*) Nota: esta estimación es una simulación dado que se reconoce que la televisión por suscripción en Colombia no puede cobrar publicidad En los países estudiados, a pesar de que existen señales gubernamentales (sean de los poderes ejecutivo, legislativo o judicial, locales o nacionales) o de universidades, éstas han capturado una proporción mínima de los recursos publicitarios (por ejemplo, en México es menos del 1% del total de la publicidad en televisión abierta) A cualquiera de los poderes (ejecutivo, legislativo o judicial) a nivel nacional o regional Fuente: LAMAC (2011) con base en IBOPE Media, México La publicidad oficial es una transferencia de recursos públicos Ver Ésta es una sobreestimación de la generación de ingresos, ya que, por definición, las audiencias se verán fragmentadas, lo que disminuye el valor de la publicidad por señal. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 121

172 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA A continuación, se presentan los presupuestos utilizados para calcular estas proyecciones en cada uno de los tres países estudiados en detalle. Con respecto al número de señales de televisión digital, para Colombia, la preservación de la banda de UHF para la radiodifusión permitirá la introducción de 7 canales de alta definición. De acuerdo al esquema actual de ocupación de la banda UHF, existen 6 canales nacionales, 7 regionales y 48 locales transmitiendo actualmente en 700 MHz. Asumiendo que la capacidad máxima de la banda permite de acuerdo al modelo de multiplexacion, un máximo de 32 canales (promedio Europeo), se estima que existiría la posibilidad máxima de introducción de 21 canales. Sin embargo, dada la oferta existente, un escenario más probable es que cada canal nacional y regional planee introducir un canal de alta definición, lo que requeriría como máximo 7 multiplexadores. Para México el máximo número de señales digitales que podrían ser lanzados en la banda disponible es de 30. Aunque aún es prematuro estimar el número de señales que piensan lanzarse en México, de manera conservadora estimamos 15 señales de televisión digital 13 abiertas y 2 que podrán ser licitadas. El plan de utilización de múltiples no ha sido definido aún. Indicios en prensa, por declaraciones de personal de la COFETEL, indican que existe una tendencia a asignar, por proceso licitatorio, dos canales a nivel nacional. El proceso licitatorio, sin embargo, deberá considerar criterios adicionales a los económicos. Para Perú, a pesar de que la implementación de la TDT está avanzando y la banda de 700 MHz está atribuida a telefonía móvil y se espera que se licite en un futuro cercano, estimamos de igual manera cuál sería el valor si se utilizara para servicios de radiodifusión. Es importante destacar que, aunque aparentemente está decidido, el cambio en la administración y el retraso a los planes anunciados por el gobierno anterior, podrían existir nuevas consideraciones. Por lo tanto, para fines de comparación, se supuso que serían lanzadas 10 señales de televisión digital por encima de la que ya están siendo planeadas. Para las estimaciones se supuso que se tendrían 5 señales por multiplexador excepto en el caso de Argentina, que ya ha sido definido. En Brasil se planea que pueda haber alternancia de formatos dependiendo del horario, por lo que hemos de manera conservadora 5 señales, que es el máximo esperado. Como punto de comparación, es importante resaltar que la oferta promedio de señales digitales en los países de la Unión Europea es de 32. Para la estimación de los ingresos por subscripción, se tomaron para Colombia y México las cuotas de canales premium en cada uno de estos países. Un modelo en estas líneas ya fue implementado en México, pero tuvo cuestionamientos legales, por lo que ha sido suspendido. Es posible que en el futuro, una vez que avance la digitalización de la televisión, este modelo pueda ser relanzado, por lo que no lo descartamos en nuestro análisis. De igual manera, tanto en Perú como en Colombia existe demanda por servicios de televisión de paga que podrán ser satisfechos con modelos similares. Como los avances de la implementación de la TV digital en estos países aún son incipientes, no hemos descartado estos modelos de nuestro análisis. En general, los precios varían relativamente poco de país a país ya que gran parte del contenido proviene de empaquetadoras internacionales y tienden a hacer pocos ajustes por mercado. Por suscriptor por mes, los canales más caros son vendidos en aproximadamente US $0,50, mientras que los canales marginales tienen un precio alrededor de US $ 0,05. La estimación se basa en el precio promedio global, que está entre US $0.10 y US $0,11. Para Perú se hizo un ajuste tomando los precios actuales en el mercado local, suponiendo el precio por mes en promoción de un canal premium en alta definición, lo que equivale a US $1,24. Para la estimación de los ingresos por suscripción, se consideró la adopción de paquetes de canales en igual proporción que las penetraciones actuales de televisión de paga en cada uno de los países y el tamaño del paquete (ver figura 7-18). Para los ingresos de publicidad, se tomaron los valores de México por cada punto de rating normalizado por habitante y ajustado por nivel de riqueza (PIB). En México, la publicidad multicanal genera anualmente US $285 millones por 30 canales locales. Haciendo los ajustes relevantes por PIB y por población, estimamos que los ingresos anuales generados por el concepto de publicidad por señal digital son de US $3,8 millones, US $6,0 millones y US $3,8 millones para Colombia, México y Perú, respectivamente Contribución directa a la creación de empleo La creación de empleos directos de la televisión ha sido estimada en términos del número de empleados necesarios para operar una señal de televisión (equivalente a un canal actual). Éstos han sido estimados entre 60 y 80 dependiendo de las economías de escala de las empresas de cada uno de 204. Plano Básico de distribuição de canais digitais PBTVD, Decreto 5.820/2006, Portarias MC 652/2006 y 256/ La suscripción de TV Movistar es de nuevos soles por 8 canales. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 122

173 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA América latina: Supuestos de adopción de servicios de televisión de paga Colombia México Perú Adopción de paquetes de canales 33% 35% 23% Compra promedio de canales por paquete ,5 Fuente: SNL Kagan, Anatel (Brasil), Ministerio TIC (Colombia) FIGURA América Latina: Contribución directa a la creación de empleo de la televisión en 700 MHz al PIB País Número de señales digitales Empleos por señal Número de empleos Argentina Brasil 30 (20 locales) Colombia 7 locales México Perú Resto de América Latina Total Fuente: análisis TAS países considerados. Así, los empleos directos a ser creados por la televisión en los cinco países estudiados suman (ver figura 7-19), con empleos adicionales generados en el resto de la región, danto un total en América Latina de El número reducido de empleados por señal de televisión está determinado por las economías de escala y la compra de producción sindicada que no requiere plantilla adicional. De hecho, solamente las señales de televisión digital locales requieren personal adicional. Por ejemplo, en el caso de Brasil, dado que un grupo controla la mayoría de la producción televisiva nacional, se supone un factor de sinergia, reduciéndose el empleo por señal adicional a 60. Por otra parte, 20 de las 30 señales adicionales son locales, requiriendo la contratación de personal Contribución indirecta a la economía En teoría, la radiodifusión debería ejercer tres tipos de impacto indirecto en la economía: externalidades positivas en el resto de la economía, contribución tributaria y creación de excedente del consumidor. En este análisis, y de manera 207 consistente con investigaciones realizadas hasta la fecha, las externalidades han sido estimadas en cero. Si bien la televisión contribuye a la creación de ingresos y empleo, éstos han sido incluidos dentro del ecosistema y, por lo tanto, considerados como efecto directo (ver arriba sección 7.1). Por otro lado, si 206. La posible existencia de 220 licencias indica que se alcanzarán pocas economías de escala, por lo que supusimos un número cercano al nivel superior de las estimaciones existentes Ver Boston Consulting Group (2010), p. 113; Bohlin, et al. (2007), p. 12 EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 123

174 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA bien la creación de un excedente del consumidor como medida de beneficio al usuario existe, ésta es extremadamente difícil de estimar. La introducción de señales de televisión digital adicionales puede crear beneficios vinculados al servicio público y la mejor información de ciudadanos, pero éstos son difíciles de cuantificar. Por otra parte, es importante mencionar que, pese a esta dificultad, la asignación de espectro a la radiodifusión genera un efecto de segundo orden que se traduce en mayor espacio publicitario (creado por las señales adicionales) con el consiguiente posible excedente del productor (anunciantes) y consumidor (transferencia de parte del excedente del productor a menores precios en productos y servicios de los anunciantes) Contribución impositiva La contribución impositiva de la televisión ha sido estimada en términos del impuesto al ingreso marginal a ser generado por el lanzamiento de las señales adicionales. Esta contribución varía de acuerdo al número de señales digitales a ser lanzadas por país y a la tasa del impuesto a las ventas. Obviamente, en el caso de canales públicos, la contribución impositiva no es considerada. Así, los impuestos a ser generados por la televisión en los cinco países estudiados suman US $817,82 millones (ver figura 7-20) Impacto económico acumulado Si la banda de 700 MHz continuase atribuida a la radiodifusión para su explotación televisiva, ya sea en el modelo existente o de desarrollo de la TV digital, se genera valor en áreas como contribución directa a la economía como consecuencia de la oferta de señales digitales adicionales (en términos de suscripción y publicidad), creación de empleo, impacto en el ecosistema de la radiodifusión como resultado de la construcción y operación de redes de retransmisora nacionales y gastos de producción de contenido, y contribución impositiva. Los valores para los cinco países estudiados en detalle y para el resto de la región latinoamericana son los siguientes (ver figura 7-21). FIGURA América Latina: Contribución impositiva adicional de la radiodifusión (en millones de US $) País Ingresos adicionales Tasa de contribución tributaria Contribución impositiva anual Contribución acumulada (*) Argentina $ 0 21,0% $ 0 6$ 0 Brasil $ 0 27,5% $ 0 $ 0 Colombia $ 53,4 20,0% $ 10,69 $ 85,49 México $ 144,6 16,0% $ 23,13 $ 185,04 Perú $ 84,4 19,0% $ 16,04 $ 128,29 Resto de América Latina $ 231,0 $419,00 Total $ 513,40 $ 817,82 Fuente: análisis TAS (*) Por ejemplo, calidad de vida, inclusión social, ciudadanía bien informada, pertenencia a una comunidad. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 124

175 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA FIGURA América Latina: Impacto económico acorde con el uso de 700 MHz para televisión (en millones US $ excepto empleos) Impacto Área Argentina Brasil Colombia México Perú Resto AL Total Contribución al ecosistema Bienes productivos $ 93,0 $ 833,1 $ 63,0 $ 135,0 $ 90,0 $ 334,0 $ 1.548,1 Servicios operativos $ 13,4 $ 119,9 $ 9,1 $ 19,4 $ 12,9 $ 48,0 $ 222,7 Servicios comerciales $ 180,0 $ 540,0 $ 126,0 $ 270,0 $ 180,0 $ 441,0 $ 1.737,0 Contribución directa Ingresos adicionales $ 0 $ 0 $ 53,4 $ 144,6 $ 84,4 $ 231,0 $ 513,40 Empleos adicionales , Contribución indirecta Contribución impositiva $ 0 $ 0 $ 85,49 $ 185,04 $ 128,29 $ 419,0 $ 817,82 Fuente: análisis TAS 7.7. Impacto social y beneficios públicos: Es importante, finalmente, discutir cuales son los beneficios sociales a ser generados a partir del uso de la banda de 700 MHz para la radio difusión. Si bien estos son extremadamente difíciles de cuantificar, esto no impide su evaluación, en particular si la consideración de criterios puramente económicos puede esconder beneficios sociales. Oliver et al. (2008) definen "valor publico" como aquel generado por un servicio que crea beneficios que van mas allá del excedente de consumidores y productores. Así, los "beneficios públicos" generados por la radiodifusión pueden incluir: Generación de contenidos educacionales Promoción de una ciudadanía informada y por lo tanto receptiva de valores democráticos Creación de una identidad cultural y sentido de pertenencia a una comunidad Inclusión social El beneficio social fundamental de la televisión, en tanto medio de comunicación, es la transmisión de contenidos, especialmente relevante en el terreno educativo e informativo. La educación a distancia en términos de complemento a la educación presencial ya es utilizada en numerosos países en vías de desarrollo. Su eficacia ha sido probada como recurso complementario a la presencia de instructores en áreas como lenguaje, y gramática. Asimismo, la televisión, en tanto medio de comunicación de masas juega un papel fundamental en la difusión de información y noticias lo que resulta en una población más dispuesta a participar en el terreno democrático. Finalmente, la televisión puede ayudar a la difusión de programas de educación sanitaria, como por ejemplo, en la prevención de enfermedades infecciosas. Por otra parte, en la medida de que la televisión no incluye un canal de distribución de datos punto a punto, su poder en las áreas de inclusión financiera y acceso a servicios públicos es extremadamente limitada (ver figura 7-22). Sin negar la relevancia de la televisión en tanto medio con beneficios sociales, corresponde preguntarse hasta que punto estos beneficios pueden modificar nuestro análisis comparado de escenarios de asignación de espectro. En este sentido, la pregunta a considerar es: hasta qué punto estos beneficios reales en el caso de la asignación a la radiodifusión no existen también en el caso de la banda ancha móvil? EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 125

176 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA La respuesta es que estos beneficios sociales son generados en ambos escenarios. Por ejemplo, los programas de educación a distancia han sido desplegados con base en ambas plataformas (televisión y banda ancha) y los beneficios tienden a variar acorde con los contenidos educativos. Por ejemplo, la televisión es más efectiva en la enseñanza de idiomas y gramática, mientras que plataformas y aplicaciones con banda ancha son más útiles en la enseñanza de geometría, matemática, y ciencias, donde la interactividad es un factor decisivo. De manera similar, existe una relación verificada entre una ciudadanía informada y conciencia democrática (Haltz-Bacha et al, 2000). El problema es que estudios que prueban esta relación, lo hacen estudiando acceso a noticias e información. Es indudable que, en este caso, la banda ancha, implica una oportunidad más amplia de acceso a información que la distribución vertical implícita en la televisión. FIGURA Beneficios sociales de la televisión Área de impacto Ejemplos Radiodifusión Educación Conectividad a recursos edicacionales Educación a distancia Salud Telediagnóstico Comunicación interprofensional de salud Información sanitaria Inclusión financiera Acceso a plataformas de micropagos Educación para acceso a microfinanzas Acceso a servicios públicos Acceso a programas de gobierno electrónico Inclusión informativa Acceso a información estatal Programas de relación enrre ciudadanos y gobierno Fuente: análisis TAS Impacto Alto Impacto Nulo Respecto a la creación de identidad cultural, si bien es cierto que la televisión puede crear el reforzamiento de pertenencia a un grupo, esto también puede ser generado por aplicaciones y servicios distribuidos por la banda ancha. En resumen, sin desconocer los beneficios públicos pasibles de ser generados por la televisión, consideramos que estos pueden ser igualmente producidos mediante la adopción de banda ancha. Adicionalmente, corresponde preguntarse si el impacto combinado de televisión distribuida en señal digital en las nuevas frecuencias atribuidas y servicios entregados a partir de la banda ancha móvil no genera un efecto mas importante para aquellas poblaciones que están excluidas al día de hoy de acceso a la banda ancha. EFECTOS ESPERADOS COMO CONSECUENCIA DE LA ASIGNACIÓN DE ESPECTRO A LA RADIODIFUSIÓN 126

177 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA 8. CONCLUSIONES Este estudio ha comparado el impacto socio-económico de escenarios alternativos de utilización del espectro de 700 MHz en América Latina. El trabajo se enfocó en dos aspectos: la contribución económica al ecosistema de las tecnologías de la información y comunicación (TIC) y el beneficio económico y social. En el primer análisis, se estudió el beneficio comparado a ser generado para los proveedores de la industria de telecomunicaciones móviles (equipamiento de red, industria de la construcción, sistemas de información, etc.) y de televisión (industria de la construcción, equipamiento, producción de programas, etc.). Al mismo tiempo, se consideraron los beneficios potenciales al tesoro público como resultado de la subasta de espectro hecho disponible a las empresas privadas. También se hizo una estimación de los ahorros en el despliegue y operación de red que serían posibles en caso de poder satisfacer la demanda de transmisión de datos móviles con la utilización de la banda de 700 MHz, derivada de una mayor propagación de la señal y por lo tanto, de una menor necesidad de estaciones radio base para alcanzar una cobertura similar. Asimismo, se estimó cuál sería la cobertura adicional que sería económicamente viable y que no sería posible alcanzar con la utilización de bandas superiores. Finalmente, se evaluó el excedente del productor a ser generado para la industria de las telecomunicaciones móviles como resultado de la reasignación de espectro, suponiéndose que este excedente será trasladado en parte a los precios minoristas, beneficiando al usuario final. En el segundo análisis, se evaluó el impacto económico y social de ambos escenarios de asignación en términos de su contribución directa al PIB (producto interno bruto) nacional, como resultado de la oferta de bienes y servicios, y de la contribución indirecta, como resultado de los efectos de derrame y externalidades generadas en otros sectores de la economía. Al mismo tiempo, se calculó la creación de empleo directo e indirecto, la contribución a la recaudación impositiva y la creación de un excedente del consumidor. Los resultados del análisis de contribución al ecosistema muestran una diferencia importante en la generación de valor en términos de adquisición de bienes y servicios según el escenario de utilización del Dividendo Digital. En el capítulo 6 se demostró que si la banda de 700 MHz fuese asignada a la banda ancha móvil, esta contribuiría entre US $8.296 millones y US $ millones en los cinco países estudiados en detalle. El rango está determinado por la proporción de espectro a ser subastado en caso de su reasignación a la banda ancha móvil. Si la subasta típica en los cinco países estudiados en detalle fuese de 60 MHz, el monto estimado de valor a recaudar sumaría US $5.042 millones. Si el espectro a ser subastado llegase a 90 MHz, el valor estimado sería de US $7.561 millones. El resto de valor a ser generado incluye el correspondiente a inversiones en la adquisición de infraestructura, servicios operativos y comerciales. Más allá del impacto en el ecosistema, el costo-beneficio de asignación del espectro a la telefonía móvil se manifiesta en otras dos áreas. En primer lugar, la utilización de ese espectro permite reducir la inversión requerida para satisfacer la demanda creciente de tráfico de datos, especialmente en lugares donde la cobertura con bandas superiores no se justifica económicamente. Los despliegues en bandas superiores (900 MHz, MHz, MHz, MHz, MHz, aunque habrá casos en donde será posible utilizar la banda de 850 MHz), que tienen menor calidad de propagación, lo que requerirá un mayor número de sitios que si se utilizara la banda de 700 MHz. Asimismo, un menor número de sitios está asociado con un gasto inferior de operación y mantenimiento. Aunado a esto, la utilización de la banda de 700 MHz por los servicios móviles permite conseguir mayor cobertura del territorio, ya que su alcance es significativamente mayor (10 kilómetros de radio, comparado con rangos inferiores a kilómetros en otras frecuencias ). Así, el valor fundamental de la reatribucion del espectro de 700 MHz conlleva la posibilidad de despliegue masivo de banda ancha móvil, ya que las inversiones adicionales requeridas para ofrecer este servicio a toda la población en todo el territorio en las otras bandas de frecuencia no se justifica económicamente. De esta manera, la cobertura de banda ancha móvil, que al día de hoy alcanza a 75% de la población en Argentina y Brasil, estimándose este número en 52% para Colombia, podría extenderse significativamente ayudando a cerrar la brecha digital. Secundariamente, la utilización de la banda de 700 MHz por los servicios móviles permitiría cubrir aproximadamente 20, Fuente: FCC. The broadband availability gap, OBI Technical Paper No.1. Abril CONCLUSIONES 127

178 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA millones de personas en América Latina que hoy radican en áreas aisladas que no tienen cobertura de telefonía móvil, lo 210 que equivale a 4.8% de la población. Todo ello sería logrado 211 con un ahorro de más de US $3.701 millones, de los cuales US $2.280 millones (equivalentes a US $3.690 millones nominales a lo largo de 8 años) corresponden a menor inversión y US $1.420 millones a menores costos de operación. Para el resto de la región, el ahorro en inversión y operación sería de US $ Éste es el valor del Dividendo Digital desde la perspectiva del despliegue de infraestructura. Alternativamente, si el espectro quedara asignado a la radiodifusión para la transmisión de señales de televisión, ésta contribuiría US $ millones en los cinco países y US $ 823 adicionales en el resto de los países latinoamericanos. Es importante mencionar que en esta estimación no se ha incluido gasto alguno por la licitación de espectro, dado que es difícil establecer un costo dada la experiencia limitada en asignaciones de este tipo. Como los números lo indican, una de las contribuciones más importantes en el ecosistema de la radiodifusión es la adquisición de servicios de programación para generar contenidos a ser transmitidos en las nuevas señales locales que podrían ser lanzados en caso de que el espectro fuese asignado a la televisión. Evaluando el impacto económico y social, los resultados favorecen igualmente al escenario de atribución de espectro a los servicios móviles. En primer lugar, la asignación de la banda de 700 MHz a las telecomunicaciones móviles contribuye directa e indirectamente al PIB 7,0 veces más que la radiodifusión. En el caso de la contribución directa, consideramos la oferta de productos y servicios adicionales a los existentes generada por el acceso a la banda de 700 MHz. Ésta incluye abonados nuevos a la telefonía móvil en zonas rurales y conexiones a computadoras portátiles (derivadas de una reducción de precios de 10% con base en una elasticidad de demanda de 0,6). En el caso de la televisión, los ingresos adicionales se refieren a la venta de publicidad por señal de televisión digital adicional y la suscripción pagada por abonados a señales premium. Sin embargo, es importante mencionar que dado que la utilización del espectro por la radiodifusión sería hecha por canales de televisión abierta, la contribución en términos de venta de subscripción en ciertos países sería nula. En el caso de la contribución indirecta de la banda ancha móvil, se estimaron las externalidades (o efectos de derrame en otros sectores de la economía); de manera conservadora, consideramos que solamente las conexiones de banda ancha móvil adicionales son aquellas que generan un impacto económico. Estas suman US $1.062 millones en América Latina. Así, las telecomunicaciones móviles contribuyen al PIB 7,0 veces más que la televisión. En segundo lugar, la asignación de la banda de 700 MHz a las telecomunicaciones móviles contribuye significativamente más a la creación de empleo. En el caso de empleos directos (incluyendo aquí los empleados de las empresas, así como también aquellos existentes en los proveedores a las mismas) a ser creados por cada industria, la radiodifusión tiende a crear un número mayor de puestos de trabajo: empleos comparados con para las telecomunicaciones móviles. Esto se debe a que en el caso de la televisión, la variable determinante es el número de señales completamente nuevas a ser lanzadas al mercado ubicadas por sobre el canal 51, lo que multiplica de manera constante la cantidad de empleados necesarios por señal (entre 60 y 70, dependiendo del país). En el sector de la banda ancha móvil, la variable determinante en la creación de empleos directos es el número de abonados adicionales incorporados como resultado de una mayor cobertura de los territorios nacionales y de un mayor número de líneas de banda ancha móvil, consecuencia de menores precios y mayor cobertura. En este caso, dadas las importantes economías de escala de la industria móvil y el hecho de que en todos los países considerados la misma está operando a niveles óptimos de despliegue, el incremento marginal de empleados como resultado de las líneas adicionales es muy reducido. La diferencia importante entre ambos escenarios ocurre en virtud del efecto de creación de fuentes de trabajo en otras industrias. Dadas las externalidades ya debidamente verificadas de la banda ancha móvil, la adopción de nuevas líneas de conexión de computadoras portátiles resulta en un efecto multiplicador en la creación de empleo de otros sectores de la economía. Por otra parte, la televisión, más allá del efecto de publicidad adicional en las nuevas señales (factor que está limitado por la dimensión total del mercado publicitario) no genera empleo indirecto significativo. Por otra parte, como se mencionó arriba, todo aumento de empleo en producción y publicidad para la televisión está incluido en los efectos directos. En tercer lugar, la atribución de la banda de 700 MHz a la banda ancha contribuye en los cinco países estudiados en detalle US $ millones más que la radiodifusión a la 210. La cobertura adicional sería aproximadamente de 1% en Argentina, 4% en Brasil, 6,4% en Colombia, casi 7% en México, y 4,5% en Perú. Estos valores se refieren al despliegue de todo tipo de red de telefonía móvil. Se debe considerar que para alcanzar cobertura universal de banda ancha móvil, la cobertura debería ser más elevada dado que estas cifras incluyen tecnología 2G Cifra en valor presente neto descontada al 10% de 2012 a CONCLUSIONES 128

179 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA recaudación impositiva. En el caso de la radiodifusión, se 212 consideran los impuestos directos acumulados en ocho años por ingresos adicionales de suscripción y publicidad generados por señales nuevas. En el caso de las telecomunicaciones móviles, se consideran los impuestos directos por ingresos acumulados generados en ocho años por el aumento de penetración de banda ancha móvil y expansión de cobertura en zonas rurales. Finalmente, la atribución de la banda de 700 MHz a la industria móvil genera beneficios importantes a través del excedente del consumidor. El excedente de consumidor mide la diferencia entre la voluntad de pago (como métrica de beneficio al consumidor) y el precio practicado de un bien o servicio. En el caso de la radiodifusión, si bien la introducción de señales adicionales puede crear beneficios vinculados al 213 servicio público y la mejor información de ciudadanos, éstos son difíciles de cuantificar. Por otra parte, es importante mencionar que, pese a esta dificultad, la atribución de espectro a la radiodifusión genera un efecto de segundo orden que se traduce en mayor espacio publicitario (creado por los señales adicionales) con el consiguiente posible excedente del productor y consumidor. Por el lado de las telecomunicaciones móviles, el traslado de mayores economías de inversión de capital a los precios de banda ancha móvil para computadoras conectadas y smartphones corporativos crea un excedente derivado de una reducción de precios en banda ancha móvil de aproximadamente 10%, beneficiando directamente al consumidor. Midiendo esto en términos de la reducción acumulada de tarifas sobre la base de usuarios a lo largo de ocho años, resulta en un excedente del consumidor de US $3.842 millones para los cinco países estudiados en detalle y un total de US $5.157 millones para toda la región. Este excedente del consumidor contribuye a su vez al crecimiento del PIB en la medida que puede traducirse en mayor consumo. Es posible concluir que una porción del excedente del consumidor puede resultar en un aumento del consumo en rubros como artículos del hogar, educación, esparcimiento y cuidado personal. En resumen, el estudio permite concluir que la asignación del espectro de 700 MHz a las telecomunicaciones móviles en América Latina genera más valor económico y social que si éste permaneciese asignado a la radiodifusión (ver figura 8-1). FIGURA 8-1. Comparación de los beneficios generados según la utilización de la banda de 700 MHz para los cinco países estudiados en detalle (cifras en MM US $ corrientes excepto empleo) Radiodifusión Telefonía móvil Contribución al ecosistema de TIC (espectro, red y otros activos) $ $ x Ingresos adicionales del sector y contribución al crecimiento del PIB $ 2.75 x Generación de empleo directo e indirecto x Impuestos (recaudación marginal adicional en ventas) $ $ x Excedente del consumidor $0 (*) $ (*) Efecto de segundo orden se traduce en mayor espacio publicitario con el consiguiente posible excedente del productor y consumidor Fuente: análisis TAS 212. Impuesto al valor agregado (IVA) o equivalente Por ejemplo, calidad de vida, inclusión social, ciudadanía bien informada, pertenencia a una comunidad. CONCLUSIONES 129

180 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Al extrapolar estos resultados al resto de América Latina, se observa que asignar el espectro de 700 MHz a la banda ancha móvil en América Latina genera más valor económico y social que si éste permaneciese asignado a la radiodifusión (ver figura 8-2). FIGURA 8-2.América Latina: Beneficio económico comparado según la utilización de la banda de 700 MHz (cifras en millones de US $ corrientes excepto empleo) Radiodifusión Telefonía móvil Contribución al ecosistema de TIC (espectro, red y otros activos) $ $ x Ingresos adicionales del sector y contribución al crecimiento del PIB $ $ x Generación de empleo directo e indirecto x Impuestos (recaudación marginal adicional en ventas) $ $ x Excedente del consumidor $0 (*) $ (*) Efecto de segundo orden se traduce en mayor espacio publicitario con el consiguiente posible excedente del productor y consumidor Fuente: análisis TAS Estos resultados son consistentes con los generados por investigaciones realizadas en otras regiones del mundo (ver figura 8-3). FIGURA 8-3. Beneficio relativo si el espectro es asignado a la banda ancha móvil (valor para telecomunicaciones móviles dividido por el valor para la radiodifusión) América Latina Asia Europa Contribución al ecosistema de TIC (espectro, red y otros activos) x N.A. x Ingresos adicionales del sector y contribución al crecimiento del PIB x x x Generación de empleo directo e indirecto x x x Impuestos (recaudación marginal adicional en ventas) x x N.A. Excedente del consumidor $3.8 N.A. C 70 (*) Efecto de segundo orden se traduce en mayor espacio publicitario con el consiguiente posible excedente del productor y consumidor Fuente: análisis TAS CONCLUSIONES 130

181 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Así, los resultados del estudio indican los beneficios a ser generados por la asignación de la banda de 700 MHz a la telefonía móvil en los cinco países estudiados en detalle: Un aumento en cobertura de banda ancha a través de la disponibilidad de banda ancha móvil, elemento del cual hay consenso es fundamental para el crecimiento económico de América Latina; Un despliegue y operación de nuevas redes mas óptimo resultando en una reducción de inversión de US $3.701 millones comparada con el despliegue en bandas superiores, alcanzando una mejor cobertura 214 ; Contribución al ecosistema de TIC (adquisición de espectro y de equipamiento y servicios) que excede en más de US $8.129 millones a la contribución generada por la 215 radiodifusión ; Contribución directa (ingresos adicionales de la industria) e indirecta (externalidades positivas) al PIB que excede en más de US $2.474 millones a la contribución generada por 216 la radiodifusión ; Creación de más de empleos directos e indirectos 217 adicionales a los generados por la radiodifusión ; Contribución impositiva adicional superior a US $2.142 millones; Excedente del consumidor adicional superior a US $3.842 millones. Extrapolando los resultados de los cinco países estudiados en detalle al resto de América Latina, los beneficios son, como es de esperar, mayores: Un incremento significativo en la cobertura de banda ancha móvil con una tecnología mas eficiente resultando en un ahorro en despliegue y operación de red de US $5.440 millones comparado con redes en bandas superiores; Contribución al ecosistema de las Tecnologías de Información (adquisición de espectro y de equipamiento y servicios) que excede en más de US $ millones a la contribución generada por la radiodifusión; Contribución directa (ingresos adicionales de la industria) e indirecta (externalidades positivas) al PIB que excede en más de US $3.069 millones a la contribución generada por la radiodifusión; Creación de más de empleos directos e indirectos adicionales a los generados por la radiodifusión; Contribución impositiva adicional superior a US $2.602 millones; Excedente del consumidor adicional superior a US $5.157 millones. Alcanzar un aumento del 31,5% de la cobertura de banda ancha móvil de nueva generación, alcanzando un total estimado del 92,7% de la población latinoamericana. Esto permitirá aumentar significativamente la adopción de la banda ancha y alcanzar mayores velocidades, lo cual son objetivos de política pública de la mayoría de los gobiernos de la región. En conclusión, la reasignación de la banda de 700 MHz a los servicios móviles, en la medida que permite a estos ampliar la entrega de banda ancha móvil y acrecentar la cobertura del servicio conlleva efectos económicos y sociales substanciales, al mismo tiempo que responde a necesidades del mercado. Las autoridades públicas latinoamericanas no pueden soslayar este impacto y deben seguir el ejemplo de países como Colombia, Perú, U r u g u a y y M é x i c o q u e a l i g u a l q u e o t r o s p a í s e s desarrollados están tomando decisiones que permitan la materialización de este impacto Valor nominal de US $6.155 millones (inversión: US $3.690 millones; costos operacionales: US $2.465 millones) calculado como Valor Presente Neto a ocho años descontado al 10% Gasto inicial no recurrente 216. Contribución directa anual y contribución indirecta acumulada a ocho años 217. Empleos directos anuales y empleos indirectos/año acumulados a ocho años. CONCLUSIONES 131

182 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA APENDICES Apéndice A. Sobre los autores RAUL L. KATZ Raúl L. Katz es Presidente de Telecom Advisory Services LLC, una firma especializada en consultoría para la industria de telecomunicaciones. Antes de fundar su firma, el Dr. Katz trabajó durante veinte años en Booz Allen Hamilton, de la cual se retiró hacia finales de Durante su carrera en Booz Allen, el Dr. Katz ocupó diferentes responsabilidades, como, por ejemplo, Socio Líder de la Práctica de Telecomunicaciones en las Américas, Director de Operaciones de América del Norte y miembro del equipo de dirección de la firma. El Dr. Katz ha trabajado en la industria de telecomunicaciones durante más de veinticinco años. Durante su carrera, él se ha especializado en estrategia de negocios, marketing industrial y de consumo y administración de empresas de telecomunicaciones en entornos altamente competitivos. El Dr. Katz trabajó intensamente en el asesoramiento de mercados domésticos e internacionales de servicios, equipamientos y software de telecomunicaciones. Durante su carrera, el Dr. Katz lideró proyectos en las áreas de análisis competitivo, estimación de demanda, estrategias de ingreso de mercado y desarrollo de nuevos productos, así también como la reestructuración y reorganización de empresas de telecomunicaciones. En los últimos tres años, el Dr. Katz ha trabajado asesorando a empresas de telecomunicaciones en el desarrollo de escenarios futuros de la industria y estrategia regulatoria. El Dr. Katz gerenció proyectos en los EE.UU., Europa, América Latina, Corea y Japón. El Dr. Katz posee un Doctorado en Ciencias Políticas y Administración de Empresas, y una Maestría en Política y Tecnología de las Comunicaciones del MIT (Massachussets Institute of Technology). Además, es Licenciado en Historia y Master en Ciencias Políticas de la Universidad de París-Sorbona. Finalmente, posee una Licenciatura y una Maestría, con honores, en Ciencias de la Comunicación de la Universidad de París. El es actualmente Profesor Adjunto en la División de Finanzas y Economía del Columbia Business School, Director de Investigación de Estrategia Empresaria en el Columbia Institute for Tele-Information y Profesor Visitante en el Programa de Gestión de Telecomunicaciones en la Universidad de San Andrés en Buenos Aires, Argentina. El Dr. Katz ha publicado artículos en periódicos, tales como Telephony, Telecommunications Policy, The Information Society, y America s Network. Su libro The Information Society: An International Perspective, centrado en las tendencias de desregulación en la industria de las telecomunicaciones a nivel mundial, se publicó en Su segundo libro Creative Destruction: Business Survival Strategies in the Global Internet Economy fue publicado en 2000 y traducido al Japonés. Su tercer libro La Contribución de las tecnologías de la información y las comunicaciones al desarrollo económico: propuestas de América Latina a los retos económicos actuales fue publicado en Su disertación doctoral obtuvo el premio K. Kyoon Hur Memorial Dissertation Award de la Asociación Internacional de Comunicaciones en ERNESTO FLORES-ROUX El Dr. Ernesto Flores-Roux es Investigador en el Centro de Investigación y Docencia Económica (CIDE) (México). Previamente, fue coordinador de asesores de la Subsecretaría de Comunicaciones de la Secretaría de Comunicaciones y Transportes en México. Tiene 17 años de experiencia en el sector de telecomunicaciones, habiendo trabajado para diversos gobiernos, empresas y asociaciones en aspectos de regulación, mercadotecnia, estrategia, definición de precios, fusiones y alianzas. Fue Director de Marketing y Estrategia para Telefónica Móviles México ( ), época en que se ejecutó el cambio de imagen corporativa de la empresa. En 2006 se transfirió a Telefónica del Perú, donde asesoró a la presidencia de la empresa en temas relacionados a las tarifas reguladas. Durante 2007 estuvo sediado en Beijing APENDICES 132

183 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA como Director de Estrategia para Telefónica en Asia, trabajando en la relación con China Netcom. Finalmente, se transfirió a Brasil como asesor del presidente de Telefónica en São Paulo. El Dr. Flores-Roux trabajó de 1993 a 2004 en McKinsey & Co., Inc. Durante esta época, primeramente asesoró a la Secretaría de Comunicaciones y Transportes en México en la creación de la Ley Federal de Telecomunicaciones de 1996 y aspectos de ella derivados, en particular respecto a la interconexión de redes. En 1996 se transfirió a Brasil, donde durante más de dos años trabajó directamente con el Ministerio de Comunicaciones, Telebrás y posteriormente Anatel en el proceso de definición del modelo económico de la liberalización del sector, la creación de la agencia reguladora, la privatización de Telebrás y varios aspectos de regulación. De 2002 a 2004 fue el socio encargado de la oficina de McKinsey en Río de Janeiro. El Dr. Flores-Roux posee un Doctorado en Estadística de la Universidad de Chicago (1993) y es Licenciado en Matemáticas de la Universidad Nacional Autónoma de México (1989). JAVIER G. AVILA El Lic. Javier Ávila es un analista de Telecom Advisory Services LLC, especializado en modelos regulatorios y análisis económico y basado en Nueva York. Antes de incorporarse a Telecom Advisory Services, el trabajo como analista regulatorio con VTR Globalcom S.A, en Santiago, Chile. Javier Ávila posee una Maestría en Economía Aplicada de la Universidad de Chile y es Licenciado en Economía especializada en Economía del Sector Publico de la Escuela Politécnica Litoral de Guayaquil, Ecuador. GIACOMO MEILLE El Lic. Meille es un analista de Telecom Advisory Services, LLC, basado en Nueva York. El posee un BA en Economía y Matemáticas de la Universidad de Chicago. APENDICES 133

184 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Apéndice B. Bibliografía. ESTUDIOS SOBRE TRÁFICO DE DATOS: ABI Research. Cellular M2M Connectivity Services. Oyster Bay, NY. September 21, Cisco. Cisco Visual Networking Index: Global Mobile Data Traffic Forecast Update, February 1, 2011 Convergencia Research. Contribución de la Telefonía Móvil a las Economías De Latinoamérica y El Caribe. Informe a la GSMA América Latina. IDATE. Mobile traffic forecasts : A report by the UMTS Forum Report 44. Wireless Analytics. Handset Data Traffic October, ESTUDIOS SOBRE EL DIVIDENDO DIGITAL: Aegis. Digital TV Spectrum Requirements and the Digital Dividend. Briefing Note Prepared for GSM Association. May 25, Aegis. Digital TV Spectrum Requirements WP4: Status of Digital TV Spectrum in Latin America. Report for GSMA. May 27, Analysis Mason. Valuation of the digital dividend in France. A report for ARCEP. May 27, Analysis Mason, Econ, Hogan & Hartson. Exploiting the digital dividend: a European Approach. A report for the European Commision. August 14, Avanzini, D. y Muñoz, R. El valor social de la banda 700 MHz en América Latina. Santiago: DIRSI, Agosto Bazelon, C. An analysis of accelerated digital television transition. Washington, DC: Analysis Group, May 31, 2005 Boston Consulting Group. Socio-economic impact of allocating 700 MHz band to mobile in Asia Pacific. A report to the GSMA. October Bohlin, E.; Forge, S.; Blackman, C. Economic Impacts of Alternative Uses of the Digital Dividend. Bucks, UK: SCF Associates, Ltd. September Cabello, S. Oportunidades del Dividendo Digital. México, Octubre 10, Guerreiro Consult. Social benefit of licensing Wimax spectrum in Brazil: an Economic analysis. May Hazlett, T.; Muñoz, R. A welfare analysis of spectrum allocation policies. George Mason University Law and Economics Research paper series April 7, Oliver & Ohlbaum Associates Ltd and DotEcon Ltd. The Effects of a Market-Based Approach to Spectrum Management of UHF and the Impact on Digital Terrestrial Broadcasting. February 27, Parnias, A; Torras, D. The digital dividend in Europe: in the eye of the "perfect storm". Delta Partners. April, Roetter, M. Spectrum for Mobile Broadband in the Americas: Policy Issues for Growth and Competition. A report to GSMA Americas. January 2011 Spectrum Value Partners. Getting The Most Out Of the Digital Dividend: Allocating UHF spectrum to maximize the benefits for European society. March 2008 Spectrum Value Partners and Venture Consulting. Getting the most out of the digital dividend in Australia; Allocating UHF spectrum to maximize the economic benefits for Australia. April 2009 ESTUDIOS SOBRE EL IMPACTO ECONOMICO DE LAS TELECOMUNICACIONES: Atkinson, R., Castro, D. & Ezell, S.J. (2009). The digital road to recovery: a stimulus plan to create jobs, boost productivity and revitalize America. The Information Technology and Innovation Foundation, Washington, DC. Atkinson, R. and Schultz, I. (2010). Broadband in America. New York, NY: Columbia Institute for Tele-Information. BMWi (2009a). Breitbandstrategie der Bundesregierung, Stand (2009, February). Retrieved from de/dateien/bba/pdf/breitbandstrategie-der-bundesregierung, property =pdf,bereich=bmwi,sprache=de,rwb=true.pdf. BMWi (2009c). Konjunkturgerechte Wachstumspolitik Jahreswirtschaftsbericht 2009, Redaktion/PDF/Publikationen/jahreswirtschaftsbericht- 2009,property=pdf,bereich=bmwi,sprache=de,rwb=true.pdf. Clarke, G. (2008). Has the Internet Increased Exports for Firms from Low and Middle-Income Countries? Information Economics and Policy 20. APENDICES 134

185 GSMA / AHCIET BENEFICIOS ECONÓMICOS DEL DIVIDENDO DIGITAL PARA AMÉRICA LATINA Crandall, R., Jackson, C., & Singer, H. (2003). The Effect of Ubiquitous Broadband Adoption on Investment, Jobs, and the U.S. Economy. Washington DC: Criterion Economics. Crandall, R., Lehr, W., & Litan, R. (2007). The Effects of Broadband Deployment on Output and Employment: A Crosssectional Analysis of U.S. Data. Issues in Economic Policy, 6. Czernich, N., Falck, O., Kretschmer T., & Woessman, L. (2009, December). Broadband infrastructure and economic growth (CESifo Working Paper No. 2861). Retrieved from Darby, L. F., Fuhr, J. P. Jr., & Pociask S. B. (2010). The Internet Ecosystem: Employment Impacts of National Broadband Policy. Washington, DC: American Consumer Institute for Citizen Research. Dreher, A, Noel G., & Pim M. (2008). Measuring Globalization Gauging its Consequences. (New York: Springer). Flores-Roux, E. Banda ancha: Una herramienta para la movilidad social I Foro Iberoamericano para el Impulso de la Banda Ancha AHCIET São Paulo, 21 de junio de Galperin, H. and Jordan, V. (2010). Speeding up the digital revolution: broadband in Latin America and the Caribbean. Santiago: Chile, Economic Commission for Latin America. Gillett, S., Lehr, W., and Osorio, C., & Sirbu, M. A. (2006). Measuring Broadband's Economic Impact. Technical Report , National Technical Assistance, Training, Research, and Evaluation Project. Greenstein, S. & McDevitt, R. (2009). The Broadband Bonus: Accounting for Broadband Internet's Impact on U.S. GDP (NBER Working Paper 14758). Retrieved from Greenstein, S. & McDevitt, R. (2010). Evidence of a Modest Price Decline in US Broadband Services (NBER Working Papers 16166). Retrieved from Katz, R. L., Zenhäusern, P. & Suter, S. (2008b). An evaluation of socio-economic impact of a fiber network in Switzerland. Polynomics and Telecom Advisory Services, LLC. Retrieved from _Broadband_Study_E.pdfwww.polynomics.ch/dokumente/Pol ynomics_broadband_study_e.pdfwww.polynomics.ch/dokum ente/poynomics_broadband_study_e.pdf. Katz, R., L. & Suter, S. (2009a). Estimating the economic impact of the broadband stimulus plan (Columbia Institute for Tele-Information Working Paper). Retrieved from elinoam.com/raulkatz/dr_raul_katz_-_bb_stimulus_ Working_Paper.pdf. Katz, R. L. (2009b). La Contribución de las tecnologías de la información y las comunicaciones al desarrollo económico: propuestas de América Latina a los retos económicos actuales. Madrid, España: Ariel. Katz, R. L. (2009c). The Economic and Social Impact of Telecommunications Output: A Theoretical Framework and Empirical Evidence for Spain, Intereconomics, 44 (1), Katz, R. L. (2009d). Estimating broadband demand and its economic impact in Latin America. Paper submitted to the ACORN REDECOM Conference 2009, Mexico City, September 5, program.html. Katz, R. L., Vaterlaus, S., Zenhäusern, P. & Suter, S. (2010a). The Impact of Broadband on Jobs and the German Economy. Intereconomics, 45 (1), Katz, R. L. (2010b). La banda ancha: un objetivo irrenunciable para Brasil. Paper presented at 54o Painel Telebrasil. Guarujã, Agosto. Katz, R.L., Flores-Roux, E. & Mariscal, J. (2010c). The impact of taxation on the development of the mobile broadband sector. GSM Association, London. Katz, R.L., Avila, J.G., Meille, G. (2011a). The impact of wireless broadband in rural America. Washington, D.C.: Rural Cellular Association. Katz, R. L. (2011b) " Cuál es el impacto de la banda ancha en México?", Anuario AMIPCI Mexico, D.F.: Asociación Mexicana de Internet. Katz, R.L. (2011c). "The impact of broadband on the economy: research to date and policy issues", Trends in Telecommunication reform Geneva: International Telecommunication Union. Koutroumpis, P. (2009). The Economic Impact of Broadband on Growth: A Simultaneous Approach. Telecommunications Policy, 33, Qiang, C. Z., & Rossotto, C. M. (2009). Economic Impacts of Broadband. In Information and Communications for Development 2009: Extending Reach and Increasing Impact, Washington, DC: World Bank. Roeller, L-E, & Waverman, L. (2001). Telecommunications Infrastructure and Economic Development: A Simultaneous Approach. The American Economic Review, 91 (4), Shideler, D., Badasyan, N., & Taylor, L. (2007, September 28-30). The economic impact of broadband deployment in Kentucky. Telecommunication Policy Research Conference, Washington D.C. Shiu, A. & Lam, P-L. (2008). Causal relationship between telecommunication and economic growth: a study of 105 countries. Paper presented at the 17th Biennial Conference of the International Telecommunications Society (ITS). Montreal, June Thompson, H., & Garbacz, C. (2008). Broadband Impacts on State GDP: Direct and Indirect Impacts. Paper presented at the International Telecommunications Society 17th Biennial Conference, Canada. Varian, H., Litan, R., Elder, A. & Shutter, J. (2002). The net impact study: the projected economic benefits of the internet in the United States, United Kingdom, France and Germany, Available from: Waverman, L., Meschi, M. & Fuss, M. (2005). The Impact of Telecoms on Economic Growth in Developing Countries. Vodafone Policy Paper Series, 2, London, United Kingdom. Waverman, L. (2009, February 29). Economic Impact of Broadband: An Empirical Study. London: LECG. APENDICES 135

186 El estudio completo puede ser descargado de los sitios: GSMA LA AHCIET

187 BENEFÍCIOS SOCIOECONÔMICOS DA BANDA-LARGA MÓVEL Bandas: UHF, L, C Luciana Camargos, Gerente Sênior de Políticas Públicas, GSMA Brasil 30 Abril 2015 GSMA 2015

188 GROWING WORLD MOBILE ECONOMIC IMPACT SPECTRUM FOR MOBILE GSMA 2015

189 WHILE IN LATIN AMERICA SPECTRUM FOR MOBILE GSMA 2015

190 UNIQUE SUBSCRIBERS AND CONNECTIONS SPECTRUM FOR MOBILE GSMA 2015

191 CHALLENGES SPECTRUM FOR MOBILE GSMA 2015

192 DIGITAL DIVIDEND STUDY IN LATAM Done by TAS, commissioned by GSMA and Ahciet Concluded in October 2011 SPECTRUM FOR MOBILE GSMA 2015

193 COMPARISON BETWEEN BS AND IMT BROADCASTING IMT Contribution to the ICT ecosystem (spectrum, network, other assets) $ 3,508 $ 14,800 x 4.2 Additional revenues and contribution to GDP growth $ 513 $ 3,582 x 7.0 Direct and indirect employment creation 5,198 10,738 x 2.1 Taxes (collected on additional sales) $ 818 $ 3,420 x 4.2 Consumer surplus $ 0 (*) $ 5,157 SPECTRUM FOR MOBILE GSMA 2015 (*) Second-order effect which translates into more advertising space with the consequent potential impact on producer and consumer surplus (in US$ millions, except for jobs)

194 THE VALUE CHAIN Spectrum acquisition Acquisition of production goods Acquisition of operational services Acquisition of commercial services Investment for the acquisition of spectrum at auction or contest 60 MHz 90 MHz Sites Transmission equipment Repeaters Civil engineering Maintenance and repair Distribution Energy Programming (acquisition and production) Billing Argentina Brazil Colombia 474 2, ,014 1, Mexico 1,330 1, Peru Rest of Latam 1,259 1,888 1, Total 6,301-9,449 4, SPECTRUM FOR MOBILE GSMA 2015 (in US$ millions)

195 L-BAND IN LATAM AND THE CARIBBEAN Done by BlueNote, commissioned by GSMA Presented in CITEL in October MHz and MHz WRC-15 SPECTRUM FOR MOBILE GSMA 2015

196 WHY L-BAND? SPECTRUM FOR MOBILE GSMA 2015

197 THE BENEFITS SPECTRUM FOR MOBILE GSMA 2015

198 CURRENT SUPPORT SPECTRUM FOR MOBILE GSMA 2015

199 C-BAND IN APAC Done by Frontier Economics, commissioned by GSMA GHz and GHz WRC-15 SPECTRUM FOR MOBILE GSMA 2015

200 THE BENEFITS SPECTRUM FOR MOBILE GSMA 2015

201 COMPARISON IN OUR BASE CASE SCENARIO THE BENEFITS EXCEED COSTS BY APPROXIMATELY INCREASE IN GOVERNMENT INCOME INCREASE IN NEW JOBS 8X $52b 100k+ SPECTRUM FOR MOBILE GSMA 2015

202 LOW-UHF STUDY IN LATAM Done by TAS as a complement to the previsous study, commissioned by GSMA Presented in GSMA Latam Plenary in November 2014 and in CITEL in February MHz WRC-15 SPECTRUM FOR MOBILE GSMA 2015

203 THE DEMAND SPECTRUM FOR MOBILE GSMA 2015

204 THE BENEFITS SPECTRUM FOR MOBILE GSMA 2015

205 THANK YOU GSMA 2015

206

207

208

209 The Mobile Economy 2015 Copyright 2015 GSM Association

210 THE MOBILE ECONOMY 2015 About the GSMA The GSMA represents the interests of mobile operators worldwide, uniting nearly 800 operators with more than 250 companies in the broader mobile ecosystem, including handset and device makers, software companies, equipment providers and Internet companies, as well as organisations in adjacent industry sectors. The GSMA also produces industry-leading events such as Mobile World Congress, Mobile World Congress Shanghai and the Mobile 360 Series conferences. For more information, please visit the GSMA corporate website at Follow the GSMA on This report is authored by GSMA Intelligence, the definitive source of global mobile operator data, analysis and forecasts; and a publisher of authoritative industry reports and research. Our data covers every operator group, network and MVNO in every country worldwide from Afghanistan to Zimbabwe. It is the most accurate and complete set of industry metrics available, comprising tens of millions of individual data points, updated daily. GSMA Intelligence is relied on by leading operators, vendors, regulators, financial institutions and third-party industry players, to support strategic decision-making and longterm investment planning. The data is used as an industry reference point and is frequently cited by the media and by the industry itself. Our team of analysts and experts produce regular thought-leading research reports across a range of industry topics.

211 THE MOBILE ECONOMY 2015 Contents Executive Summary 2 1 Global market overview Mobile continues to scale rapidly Rapid shift to mobile broadband underway Competition, regulation and the impact on margins Revenue trends and outlook Investment in capacity and next-generation networks 21 2 Mobile empowering people and society Mobile is a cornerstone of the global economy Mobile delivering digital inclusion to the still unconnected populations Delivering financial inclusion across the world Mobile addressing social challenges in developing markets 41 3 Delivering the Digital Future Mobile at the heart of the new digital ecosystem Connected living: Mobilising the Internet of Things Digital commerce The personal data opportunity Network Global enablers to spur investment and growth Enabling and encouraging investment Enabling and encouraging innovation Building trust and confidence 76 Global market overview 1

212 THE MOBILE ECONOMY 2015 Executive Summary The mobile industry continues to scale rapidly, with a total of 3.6 billion unique mobile subscribers at the end of Half of the world s population now has a mobile subscription up from just one in five 10 years ago. An additional one billion subscribers are predicted by 2020, taking the global penetration rate to approximately 60%. There were 7.1 billion global SIM connections at the end of 2014, and a further 243 million machine-to-machine (M2M) connections. The world is seeing a rapid technology migration to both higher speed mobile broadband networks and the increased adoption of smartphones and other connected devices. Mobile broadband connections will account for almost 70% of the global base by 2020, up from just under 40% at the end of Smartphone adoption is already reaching critical mass in developed markets, with the devices now accounting for 60% of connections. It is the developing world driven by the increased affordability of devices that will produce most of the future growth, adding a further 2.9 billion smartphone connections by Fuelled by the growing range of new services and applications, data traffic is expected to see an almost ten-fold increase by Slowing subscriber numbers, as well as competitive and regulatory pressures, have led to a slowdown in industry revenue growth in recent years. Revenue growth is forecast to slow further over the coming years, with a compound annual growth rate (CAGR) of 3.1% per annum through to 2020, down from 4% in the period More encouragingly, operators are showing an increasing ability to monetise the explosive growth in data traffic. Operators have invested heavily in their infrastructure over the past three years, with capital expenditure (capex) set to increase further to support mobile broadband network deployments. Capex is forecast to total US$1.4 trillion for the period out to 2020, with 3G coverage set to reach 86% of the population by Additionally, 4G is now being built out more rapidly than was the case with 3G. However, this magnitude of investment will be dependent on operators continuing to diversify their revenues, and developing new and more sustainable business models. The mobile ecosystem is a major driver of economic progress and welfare globally. In 2014, the mobile industry generated 3.8% of global gross domestic product (GDP), a contribution that amounts to over US$3 trillion of economic value across 236 countries. This figure captures the direct, indirect and productivity impacts of the mobile ecosystem, but does not include broader socio-economic effects. In the period to 2020, mobile s contribution will grow at a faster rate than the rest of the global economy, contributing 4.2% to the world s GDP by the end of the period. *GSMA Intelligence estimates that the total number of active SIM connections at end 2013 was 6.3 billion. 2 Executive Summary

213 THE MOBILE ECONOMY 2015 The mobile ecosystem directly employed nearly 13 million people in 2014, rising to over 15 million by The sector also indirectly supported nearly 12 million jobs in the broader economy in 2014 and this figure is predicted to rise over 13 million by The industry also makes a very large contribution to public funding in the form of general taxation. In 2014, approximately US$410 billion was contributed globally, with spectrum auctions generating additional revenues of over US$14 billion. Mobile is at the heart of the new digital ecosystem. It is driving innovation and the development of new services in areas such as digital content, social networking and online commerce. Mobile is delivering a new and vibrant ecosystem, based on mobile broadband networks, advanced smartphones and tablets, and a growing range of other connected devices and objects. Mobile has already redefined consumers experiences in many aspects of their daily life, as well as created a range of new business opportunities and services. New technologies, imaginative use cases and business models are likely to generate even more profound innovations, with mobile increasingly linking the digital and physical worlds. Rapid smartphone adoption allows for new products and services to be created, whether they are based on apps or on the development of low power components that are the building blocks for new connected devices. Consumers are beginning to realise the transformative potential of the Internet of Things (IoT), with an increasing number of services and launches focused on, for example, wearables and the smart home. The benefits of the new mobile ecosystem are not limited to the developed world. Innovative mobile solutions are helping to provide underdeveloped, underserved and poverty stricken regions with the opportunity to overcome socio-economic challenges, particularly in the areas of financial inclusion, health, education and disaster response. For the full potential of mobile to be realised, populations across the world need access to mobile broadband networks, and affordable devices and services. At the end of 2014, the number of people using the mobile internet reached 2.4 billion. This is expected to rise to 3.8 billion by 2020, driven by growth in developing countries. The unconnected population is predominantly rural, with low incomes and high levels of illiteracy creating barriers to mobile internet adoption. Operators, other ecosystem players, as well as governments and regulators all have a role to play in addressing these barriers and improving the reach and affordability of mobile services. With a supportive regulatory framework, the mobile sector will continue to drive socio-economic progress, benefiting individuals, companies and governments alike. While regulatory frameworks will differ from market to market, there are some general principles that apply across the globe. There are a number of steps that policymakers can take to encourage investment. These include reducing constraints on market-driven restructuring as operators seek to gain the necessary scale, while also ensuring there is a solid business case for deploying mobile technologies and services. Governments also have a role to play in encouraging innovation, and policymakers can help the mobile industry build the necessary trust and confidence in the digital economy. If policymakers and regulators encourage investment, competition and innovation, both the mobile sector and the wider digital economy will expand, creating prosperity and new jobs. A precious and finite resource, radio spectrum is fundamental to the delivery of mobile services. The International Telecommunication Union (ITU) estimates between an additional 1340MHz and 1960MHz of spectrum will be required to meet the anticipated demand in The next opportunity to identify additional harmonised spectrum for mobile broadband is at the World Radiocommunication Conference in 2015 (WRC-15) in Geneva. The outcome of WRC-15 will determine whether the telecoms industry can continue to meet the growing demand for affordable, ubiquitous, high-speed mobile broadband services. As mobile disrupts and affects more areas of consumer and business life, the potential for collaboration also grows. Players from across the digital ecosystem, ranging from mobile operators to new entrants and existing players in adjacent industries, will increasingly recognise the need for collaborative innovation, rather than competition, if they are to realise the full potential of mobile. Executive Summary 3

214 GLOBAL MARKET Unique subscribers bn bn % CAGR % % PENETRATION RATE Global SIM connections Mobile operator revenues 10bn 7.3bn Note: Including M2M 5.4% CAGR Data growth driving revenues and operator investments 3.1% CAGR US$1.15tn US$1.4tn Operator capex of up to US$1.4tn for the period Mobile broadband networks and smartphone adoption 3G/4G CONNECTIONS SMARTPHONES 3G 4G % % 2.6bn 5.9bn Data traffic to grow TENFOLD

215 Mobile contributing to economic and social development across the world Delivering digital inclusion to the still unconnected populations Mobile internet penetration 2014: 33% 2020: 49% Delivering financial inclusion to the unbanked populations 255 live services across 89 countries as of December 2014 Delivering innovative new services and apps Number of M2M connections to reach 1bn by 2020 Mobile industry contribution to GDP % GDP US$3tn 4.2% GDP 2020 US$3.9tn Public funding Mobile ecosystem contribution to public funding before regulatory fees US$411bn 2014 Employment Jobs directly supported by mobile ecosystem 13M JOBS 15M JOBS US$465bn 2020 Plus an additional 11.8M indirect jobs supported in 2014

216 THE MOBILE ECONOMY Global market overview 1.1 Mobile continues to scale rapidly At the end of 2014, half of the world s population had at least one mobile subscription, totalling over 3.6 billion unique mobile subscribers. By 2020, around three-fifths of the global population will have a mobile subscription, with close to one billion new subscribers added over the period. The global mobile subscriber base increased by just over 5% in Developed markets are growing more slowly as penetration rates approach levels close to saturation. For example, in Europe and North America, unique subscriber growth was below 1% in At the other end of the spectrum, Sub-Saharan Africa was still the world s most under-penetrated region with subscriber growth at nearly 12%. 6 Global market overview

217 THE MOBILE ECONOMY 2015 Source: GSMA Intelligence Unique subscribers by region (Millions) 2,346 2,569 2,798 3,013 3,210 3,463 3,636 3,838 4,020 4,191 4,336 4,470 4,596 Sub-Saharan Africa North AMERICA Middle East and North Africa LATIN AMERICA EUROPE Commonwealth of Independent States Asia Pacific % CAGR % CAGR Global market overview 7

218 THE MOBILE ECONOMY 2015 Unique subscriber penetration in the developed world is already very high and approaching saturation, standing at 79% at the end of The penetration rate will climb only modestly to around 81% by the end of the decade. In contrast, less than half of the population in developing markets currently has a mobile subscription, with the penetration rate at 44.6% at the end of This leaves significant room for growth, with the penetration rate expected to rise by about 11 percentage points by 2020 to 56%. The major challenge facing mobile operators and other industry stakeholders is to connect the still unconnected populations in these developing regions. The increasing level of maturity in developed markets, combined with the recent strong growth in developing markets, means that there will inevitably be a slowdown in global subscribers. Over the six years to 2014, unique subscribers grew at a CAGR of 7.6%. This figure is forecast to slow to 4.0% over the period out to Source: GSMA Intelligence Unique subscriber penetration by region 78.9% 82.2% 80.8% 70.0% 70.0% 71.1% 49.9% 59.3% 58.6% 52.3% 44.6% 57.2% 50.5% 54.3% 39.0% 48.7% EUROPE Commonwealth of Independent States NORTH AMERICA GLOBAL AVERAGE LATIN AMERICA Asia Pacific Middle East and North Africa Sub- Saharan Africa Global market overview

219 THE MOBILE ECONOMY 2015 Multi-SIM ownership is common across all regions, with a global average of 1.8 SIM cards per unique subscriber. The rate varies significantly by region, with an average of nearly two or even slightly higher in many developing regions, where prepaid plans are the norm and subscribers are most price-sensitive. At the end of 2014, there was almost one SIM card for every person, with global connection penetration standing at 97%. With the average number of SIMs per subscriber expected to be broadly stable over the next six years, the growth rates of connections will closely correlate with underlying subscriber growth. Compared with just under 10% annual growth in the past six years, only a 4.2% annual increase is expected, taking global connection penetration to 116% by the end of Source: GSMA Intelligence Mobile connections by region (Millions, ex-m2m) 5,327 4,633 4,016 5,985 6,381 6,737 7,057 7,517 7,910 8,240 8,527 8,783 9,015 Sub-saharan africa North AMERICA Middle East and North Africa LATIN AMERICA EUROPE Commonwealth of Independent States Asia Pacific % 4.2% CAGR CAGR Global market overview 9

220 THE MOBILE ECONOMY Rapid shift to mobile broadband underway There is an accelerating technology shift to mobile broadband networks across the world. Mobile broadband connections (i.e. 3G and 4G technologies) accounted for just under 40% of total connections at the end of 2014, but by 2020 will increase to almost 70% of the total. This migration is being driven by greater availability and affordability of smartphones, more extensive and deeper network coverage, and in some cases by operator handset subsidies. While 2G remains the dominant network technology globally today, its position has already declined materially. 2G connections accounted for 90% of the total in 2008, but this had fallen to around 60% at the end of In absolute terms, the number of 2G connections peaked in 2013 and fell by 6% during Source: GSMA Intelligence Global connections by technology (Millions, excluding M2M) 10,000 9,000 8,000 7,000 6,000 5,000 4,000 3,000 2,000 1, G 3G 4G The greatest impact of this technology migration is now taking place in the developing world. Mobile broadband already accounts for over three-quarters of connections in the developed world and, by 2020, the figure will reach 92%. In contrast, less than a third of connections are currently on higher speed networks in the developing world. However, this is projected to nearly reach two-thirds of connections by In absolute terms, the number of mobile broadband connections in developing markets will increase by 3.1 billion over the period. 10 Global market overview

221 THE MOBILE ECONOMY Mobile broadband coverage expanding rapidly The ongoing technology migration to higher speed networks is also facilitated by significant operator investments. Recent research from GSMA Intelligence 1 predicts that more than four out of five people will have access to 3G networks by 2020, up from 70% today. The report also highlights that 4G networks are being rolled out at a faster pace than was the case with 3G. While it took 10 years for 3G network coverage to reach half of the global population, it will take 4G networks eight years after launch to reach the same milestone, therefore reaching this level in Source: GSMA Intelligence Global mobile broadband population coverage 73% 78% 81% 84% 86% 86% 86% 58% 66% 56% 60% 63% 50% 50% 43% 44% 34% 35% 22% 19% 26% 11% 0% 0% 2% 5% G 4G 1. https://gsmaintelligence.com/analysis/2014/12/mobile-broadband-reach-expanding-globally/453/ Global market overview 11

222 THE MOBILE ECONOMY G networks becoming dominant in developed world The build out of LTE networks continued apace in 2014, with 335 networks having been deployed in 118 countries. 4G network coverage is expanding rapidly and now reaches 26% of the world s population, although there is a clear bias towards developed markets. In December 2014, 4G coverage reached 90% of the population across developed markets and 15% in the developing world. Deployments across countries in Latin America and Asia Pacific will drive global 4G coverage over the next five years. Source: GSMA Intelligence Increasing access to LTE networks globally Number of Countries Global LTE Network Deployments North America has the world s highest 4G coverage at 97%, as well as the largest proportion of 4G connections (over 40% against a global average of just over 7%). The early allocation of spectrum in the Digital Dividend band (700MHz) and programmes to expand coverage in rural areas, helped to position the US as one of the most advanced 4G markets in the developed world. Europe is now also seeing an increasing migration to 4G, with the majority of EU countries (24 out of 28) having had spectrum auctions and assigned the 800MHz band. With operators rapidly building out network coverage (reaching 63% at the end of 2014), 4G now makes up 10% of the connection base and will account for over half by By the end of the decade, developed countries are expected to reach full coverage (defined as 95% of the population); while LTE networks in developing countries will reach the majority of their population by 2019 and 60% by This will facilitate an increasing migration to LTE connections, with close to one quarter of connections forecast to be 4G by Global market overview

223 THE MOBILE ECONOMY Smartphone adoption extending beyond developed markets The increasing proportion of higher speed connections largely reflects the accelerating rate of smartphone adoption. Adoption rates have already reached 60% of the connection base in the developed world, ranging from 51% in Europe to 70% in North America at the end of Over the next four years, smartphone adoption in the developed world is expected to reach the 70-80% ceiling, the level at which growth tends to slow. Source: GSMA Intelligence Smartphone connections (millions) and adoption 4% 6% 8% 13% 20% 28% 1,900 37% 2,630 44% 3,324 59% 55% 50% 3,948 4,514 5,019 62% 5,475 5,895 65% ADOPTION RATE Sub-Saharan Africa NORTH AMERICA Middle East and North Africa LATIN AMERICA EUROPE Commonwealth of Independent States 1, Asia pacific % 14.5% CAGR CAGR Global market overview 13

224 THE MOBILE ECONOMY 2015 Driven by the increased affordability of devices, the developing world will lead most of the growth in global smartphone adoption, reaching 63% by the end of the decade. The number of smartphones across the developing world will increase by 2.9 billion out to 2020 and nearly all of these will be running on mobile broadband networks as highlighted above. Given their relatively high levels of maturity, Europe and North America will likely see an average annual growth rate in the single digits over the coming six years, while all other regions will grow by double-digits. Source: GSMA Intelligence Proportion of smartphone connections by region Developed region Developing region Smartphones Basic / feature phones Data terminals Affordability has been the key restraining factor in developing markets, but this constraint is rapidly abating. GSMA Intelligence estimates that smartphone average selling prices (ASPs) are now 30% below their 2008 levels in Asia, 25% in Latin America and 20% in Africa. 2 The majority of smartphones in the developing world are still priced above US$100, but the sweet spot for these regions is considered to be in the US$25-50 range. In 2013, Mozilla announced plans to bring a low cost smartphone in the sub-us$50 range to developing countries through collaboration with a number of handset manufacturers and operators. The company also recently announced that it will begin marketing a device priced as low as US$25 in a number of markets including India and Indonesia before the end of https://gsmaintelligence.com/files/analysis-subscription/?file= smartphones-emerging-markets.pdf Global market overview

225 THE MOBILE ECONOMY 2015 Google has also been active in the low-end smartphone market, using its Android One operating system. The company is working with local manufacturers in India to provide a good quality handset at a price point below US$100, and has already launched several devices. Android One has now been launched in Sri Lanka, Nepal and Bangladesh, with other launches in the Asia Pacific region expected over the course of Price declines have also been driven by local handset manufacturers who focus specifically on these markets and develop products that are both affordable and meet their specific needs. This includes differentiating offers with local content, apps and language support. A number of national champions previously unknown on the global stage (such as Micromax in India) have had much success with this strategy. Xiaomi is the most notable internationally, with strong sales growth in China allowing the company to now become the third largest smartphone maker globally behind Samsung and Apple Video a key driver of data traffic growth The growing number of smartphones and other advanced devices (e.g. tablets) are increasing the use of data-intensive applications, such as video streaming, on mobile networks. Cisco estimates that smartphones generate 37 times more data traffic than feature phones, while 4G smartphones generate almost three times as much data traffic as 3G smartphones. The increasing use of mobile broadband-enabled smartphones will generate an explosion of data traffic, with volumes forecast to grow at a CAGR of 57% out to 2019, an almost tenfold increase. Source: Cisco VNI Mobile 2015 Global mobile data traffic (Per month, PB) 16,140 24,314 10,666 1,480 2,523 4,175 6, Global market overview 15

226 THE MOBILE ECONOMY 2015 On-demand video on mobile devices has become increasingly popular. It is the key driver of mobile data growth, with a 66% annual increase through to 2019 compared with 57% for data as a whole. Ericsson s consumer research into active TV and internet users shows that 27% of Germans use a mobile device to watch TV or video on a weekly basis. This is a typical level for Western Europe, but the figure is even higher in Sweden. Sweden also has the highest 4G penetration rate in Europe, where TV viewing via a mobile device is 42%. YouTube stated in October 2014 that mobile devices now generate 50% of its global traffic, up from 41% in Source: Cisco VNI Mobile 2015 Video fuelling strong mobile data growth (PB per month) 24,314 16,140 2,523 4,175 6,765 10, Web/data File sharing Video Audio streaming 16 Global market overview

227 THE MOBILE ECONOMY 2015 On-demand video on mobile is the key driver of mobile data growth YouTube stated in October 2014 that mobile devices now generate 50% of its global traffic Global market overview 17

228 THE MOBILE ECONOMY Monetising data growth Operators in markets across the world are showing signs that they are able to monetise this strong growth in data traffic. This is a key factor at a time when revenues from more traditional services are under pressure and operators have significant investment commitments as they roll out high speed networks. Tiered data plans are an increasingly common tariff trend, especially in developed markets. A growing proportion of contract tariffs now offer unlimited voice minutes and text messages. As a result, the key variable that subscribers choose, and implicitly assign a value to, is the allowance of inclusive data use in their monthly tariff (there are also plans offered by some operators with limited voice and SMS allowances, but even here the data allowance is increasingly becoming the key variable and marketing point). Cisco looked at the impact of such plans in its 2015 VNI Mobile white paper. It found that tiered plans now represent more than half of all offers, up from only 4% three years ago. Unlimited data plans have decreased proportionately over the same period. Despite the drop in unlimited data plans, data usage has continued to grow. In the previous year, average usage per device on a tiered plan grew 17%, from 922MB to 1,081MB per month. This is an encouraging sign for operators as they look to monetise increasing data use at a time when revenues and profitability from traditional voice and messaging services remain under pressure. In Sweden, the first market to launch 4G in December 2009, market leader TeliaSonera now has 40% of its mobile base on a data-centric plan. The company has seen average revenue per user (ARPU) rise from SEK180 when it launched the first such plan (in the first quarter of 2013) to SEK191 by the third quarter of This has supported a doubling of data volumes to nearly 1GB per month for consumers, as over 15% now reach their data limit and close to 80% of those go on to purchase a topup package. However, some markets are also seeing the return of unlimited plans. South Korea, one of the most advanced 4G markets with 100% population coverage and over two-thirds 4G adoption at the end of 2014, has matured to the point that it is seeing a greater number of users upgrade to unlimited plans. In the case of SK Telecom, which introduced an unlimited LTE package in April 2014, data use increased from 2.2GB in the first quarter of 2014 to 3GB as of October. ARPU rose correspondingly from KRW35,300 (US$32.70) in the first quarter to KRW36,400 in the third quarter, an increase of over 3%. The unlimited plans are priced at KRW80,000, so increasing demand for data is likely to continue to provide an ARPU uplift. 18 Global market overview

229 THE MOBILE ECONOMY Competition, regulation and the impact on margins Increasing competition and regulatory intervention have been consistent themes across most global mobile markets over recent years. In addition to the impact of new entrants in certain markets, regulatory measures to increase competition have included the introduction of mobile virtual network operators (MVNOs) and mobile number portability. IP-based services have continued to grow mindshare over recent years, at the expense of mobile operators, with clearly negative implications for traditional voice and messaging revenues. The impact has been most evident in Europe, where WhatsApp has gained particular traction. However, IP-based messaging services are seeing rapid adoption in most markets of the world. A range of regional providers are now looking to gain global scale and offer a growing range of services to their user bases (including voice calls in many cases). These factors have impacted mobile operator profitability over recent years. Between 2008 and 2013, Earnings Before Interest, Taxes, Depreciation and Amortization (EBITDA) margins at the global level fell by 350 basis points. Margin declines have been a particular feature in Europe, where revenue loses and a weak economic backdrop have further combined to reduce operator profitability. There were some signs of a stabilisation in the margin trends in This reflects in part some moves towards market consolidation, especially in Europe, which have helped ease competitive pressures. Operators across many developed markets have also been taking steps to rationalise their cost bases, as well as move away from handset subsidies. Research from GSMA Intelligence highlighted that the move away from handset subsidies was a clear trend across multiple developed regions, with subsidies limited increasingly to high-end 4G devices, and further introducing flexible device upgrade and financing facilities. 4 This trend has been most pronounced in North America, where T-Mobile was one of the first operators to move away from traditional handset subsidies and develop new financing and device upgrade packages. These moves have now been followed by all the main network operators. These plans explicitly separate the monthly payment for services from that of the handset device, giving consumers the option of keeping their existing handset or paying for a new one through an instalment plan. Source: GSMA Intelligence Global profitability beginning to recover from competition and regulation EBITDA Margin 36.8% 36.9% 36.4% 34.9% 33.5% 33.3% 33.5% Q1-Q3 4. https://gsmaintelligence.com/files/analysis-subscription/?file= smartphones.pdf Global market overview 19

230 THE MOBILE ECONOMY Revenue trends and outlook There have been significant variations in revenue growth between regions over recent years, with rates slowing across most parts of the world. Developing markets such as Sub-Saharan Africa and parts of Asia Pacific have seen mid to high single digit revenue increases, reflecting ongoing strong subscriber growth in the regions with the lowest penetration rates. In contrast, overall revenues have been declining in Europe. There are signs in recent quarters of a convergence in growth rates between developed and developing regions, with slowing subscriber growth, competition and regulatory action common themes to many markets. Revenue growth is forecast to slow further over the coming years, with a CAGR of 3.1% per annum through to 2020, down from just over 4% in the period This reflects the ongoing impact of factors such as market maturity, competition and regulation. Partly offsetting these pressures are signs that operators are increasingly monetising data traffic. Combined with an accelerating migration to 4G networks and devices, these are factors that are supporting the current recovery in revenue trends in Europe. After several years of decline, revenues in Europe are now expected to reach a stabilisation on a two to three year view. In contrast, revenue growth in North America appears to be slowing due to increasing competition and market saturation. Source: GSMA Intelligence Total global revenues (US$ bn) 1,029 1,085 1,124 1,148 1,200 1,244 1,284 1,321 1,353 1, % 3.1% CAGR CAGR Global market overview

231 THE MOBILE ECONOMY Investment in capacity and nextgeneration networks After reducing spend during the height of the financial crisis in 2009 and 2010, capital investment began to rise again at the start of the current decade, reflecting the need to increase capacity and deploy mobile broadband networks. Globally, operators have invested heavily in their networks in the past three years. In 2014, the figure was around US$216 billion, an annual increase of more than 9%. Going forward, the rate of growth is likely to moderate as 4G networks are near completion in some regions and the cost of equipment tends to decline as technologies become more mature. Investment levels globally are forecast to grow at a CAGR of 2.5% through 2020, with cumulative future investment over this period totalling over US$1.4 trillion. Developing new revenue streams and moving to more sustainable business models will be key if operators globally are to fund the significant investment levels required to support future data growth. Source: GSMA Intelligence Global mobile operator capex (US$ bn) % 2.5% CAGR CAGR Global market overview 21

232 THE MOBILE ECONOMY people and society Mobile empowering Mobile has had a profound impact on national economies worldwide, particularly in the areas of job creation and economic growth. Increasingly ubiquitous and higher speed mobile networks contribute to many aspects of economic, political and social life in both developed and developing regions. Mobile is the predominant infrastructure in emerging markets and is available to a larger proportion of the population than many other basic services, such as electricity, sanitation and financial. As a result, mobile is already helping to address a number of pressing social, economic and environmental challenges. These challenges are often particularly acute in developing regions, given factors such as high levels of poverty, rapid population growth, and in some areas, the impact of political instability. 2.1 Mobile is a cornerstone of the global economy In 2014, the mobile industry contributed a total of US$3 trillion to the world economy in value added terms, equivalent to around 3.8% of global GDP. This contribution can be broken down into four elements: The direct contribution of mobile operators; The direct contribution of the rest of the mobile ecosystem; The indirect impact on the broader economy; and The increase in productivity brought about by the use of mobile technologies. The direct economic contribution to GDP of mobile network operators and the wider ecosystem is calculated by combining the value added generated by companies operating in the sector across 236 countries. Value added is calculated as the total income generated by the industry to its employees (i.e. wage and other compensation payments), to governments (i.e. tax contributions) and to shareholders (i.e. business profits) 5. The direct contribution from mobile operators in 2014 was US$776 billion in value added terms. The broader mobile ecosystem generated a total value added of over US$300 billion. 5. Value added by the sector can also be approximated as the difference between the value of sales made by the sector and the direct cost of making those sales. 22 Mobile empowering people and society

233 THE MOBILE ECONOMY 2015 Source: GSMA Intelligence Direct GDP contribution of the mobile ecosystem US$ bn, % 2014 GDP % % % % % Infrastructure and SuppORT Services Network operators Handset Manufacturers Distributors and Retailers Content, applications and other services Mobile empowering people and society 23

234 THE MOBILE ECONOMY 2015 As mobile operators and the ecosystem purchase inputs and services from their providers in the supply chain, a multiplier effect on the rest of the economy is created, generating sales and value added in other sectors and industries. This benefit was conservatively estimated at a global value added of approximately US$220 billion in In addition to the direct and indirect contribution to GDP by mobile operators and the wider ecosystem, an estimated 2.2% of global GDP can be attributed to the increased productivity created by the widespread use of mobile technology. Mobile technology has transformed the way in which economic activity is carried out in virtually all the sectors of the global economy, allowing more efficient ways for workers and businesses to communicate and access information. This effect varies significantly by country and sector, and contributed US$1.7 trillion to global GDP in The mobile industry overall made a total contribution of US$3 trillion to the world economy, equivalent to 3.8% of the total GDP. Source: GSMA Intelligence Total (direct and indirect) contribution to GDP (2014 US$ bn) 1,701 3, % 0.3% 2.2% 3.8% 1.0% Mobile Operators Related industries General economy Productivity IMPROVEMENT Total impact Mobile ecosystem 24 Mobile empowering people and society

235 THE MOBILE ECONOMY Employment and public funding contribution in 2014 In 2014, mobile operators and the broader ecosystem directly employed 12.8 million people globally. The largest employment contribution came from the content, applications and services sector, with approximately 4.6 million jobs. However, it should be noted that a number of jobs in this sector were part-time or on a selfemployment basis. Large numbers of jobs were also directly supported by distributors and retailers (3.8 million) and mobile operators (3.1 million). Jobs were also indirectly supported as the industry s economic activity generated demand and jobs in other sectors, in particular, in the direct supply chain of the mobile ecosystem. In 2014, it was estimated that approximately 11.8 million jobs were indirectly supported, bringing the total impact (both direct and indirect) of the mobile industry to just under 25 million jobs. Source: GSMA Intelligence Global mobile ecosystem employment impact (Millions) Infrastructure operators HANDSET MANUFACTURING DISTRIBUTION CONTENT, APPS & SERVICES DIRECT INDIRECT TOTAL Mobile empowering people and society 25

236 THE MOBILE ECONOMY 2015 The mobile industry also makes a very significant contribution to public funding. For most countries, this includes value added, corporation and income tax, and social security from mobile ecosystem employees. It is estimated that the sector contributed more than US$400 billion to public funding in 2014, before considering regulatory and licence fees. Additionally, spectrum auctions generated revenues of over US$14 billion for governments globally. Source: GSMA Intelligence Tax contribution by the mobile ecosystem (2014 US$ bn) MOBILE SERVICES VAT Handset VAT CorporatION tax Employee income & social security TOTAL 26 Mobile empowering people and society

237 THE MOBILE ECONOMY Outlook and trends in the period By 2020, mobile technology is predicted to generate a total economic value of nearly US$4 trillion, increasing the sector s global GDP contribution to 4.2%. This growth will be driven by both demandand supply-side effects. On the demand side, mobile technologies will connect previously unconnected populations to the internet and enable a more efficient use of resources in those economies. Supply-side effects will also make a significant contribution as the number of subscribers grows and new value added services are brought to market, generating revenue and value added growth in the ecosystem. Source: GSMA Intelligence Total mobile contribution to GDP out to 2020 Value Added (US$ bn, bars) and as a % of GDP (top) 3.8% 3, % 3, % 3, % 4.2% 4.2% 4.2% 3,875 3,499 3,637 3, Mobile empowering people and society 27

238 THE MOBILE ECONOMY 2015 The total number of jobs both directly and indirectly generated by the ecosystem will also grow significantly in the period to 2020, reaching 15 and 13 million people respectively. At the same time, the public funding contribution of the mobile ecosystem (excluding spectrum and other regulatory fees) will reach US$465 billion by 2020 if tax rates remain at current levels. This is up from US$410 billion in Source: GSMA Intelligence Employment projections to 2020 (Millions) Indirect employment Ecosystem employment Source: GSMA Intelligence Outlook for global public funding contributions to 2020 (US$ bn) Mobile empowering people and society

239 THE MOBILE ECONOMY 2015 There are 12.8 million jobs supported directly by the mobile ecosystem across the globe In 2014, the mobile ecosystem contributed more than US$400 billion to public funding Mobile empowering people and society 29

240 THE MOBILE ECONOMY Mobile delivering digital inclusion to the still unconnected populations Mobile phones are bringing internet access to populations across the developing world. Many markets are also now seeing an increasing migration to smartphones and mobile broadband networks, with the potential to bring access to a broader range of apps and services. For the full potential of mobile to be realised, populations need access both to mobile broadband networks and affordable devices and services. Despite the progress to date, there remains a significant proportion of the global population who do not have access to the internet. The ITU estimates that global internet users grew from 1.6 billion in 2008 to 2.9 billion by the end of 2014, accounting for approximately 40% of the global population. 6 This leaves 60%, or approximately 4.4 billion people, still unconnected. This lack of internet access has the potential to hinder opportunities for economic and social development in many developing countries, preventing the unconnected populations from truly engaging in the information age. The current global gap in internet access will largely be addressed by mobile networks, which already provide access to billions across the world. At the end of 2014, there were 2.4 billion individuals using mobile devices to access the internet across the globe, of which 1.8 billion were in developing markets. While globally around one in three people have mobile internet access, there is a major difference between developed and developing markets. Around 60% of the population in developed markets have mobile internet access, while in developing markets the figure is only 28%. The unconnected population in these markets is predominantly rural, with characteristics such as low incomes and high levels of illiteracy that create barriers to mobile internet adoption. By 2020, mobile internet penetration rates in developing markets will have reached 45% of the population, although in both Sub-Saharan Africa and the Middle East and North Africa region, the penetration rates will remain below 40% Mobile empowering people and society

241 THE MOBILE ECONOMY 2015 Source: GSMA Intelligence Mobile internet subscriber penetration 33% 49% 32% 49% GLOBAL AVERAGE Asia Pacific 60% 70% 26% 39% DEVELOPED MARKETS Middle East and North Africa 20% 38% Sub-Saharan Africa Mobile empowering people and society 31

242 THE MOBILE ECONOMY 2015 Mobile internet access can create a virtuous cycle in developing countries. In the first instance, mobile connectivity using simpler feature phones can provide communications and basic services to currently disadvantaged populations. Mobile can act as an enabler when other more traditional delivery mechanisms fall short, and is already being used to provide underserved populations with access to information and services. McKinsey has suggested that if internet access achieves an impact on the same scale as mobile telephony has in Africa, it could account for as much as 10% of total GDP by 2025, up from only 1% today. This would be equivalent to over US$300 billion, 7 due to the internet s transformational effects on sectors such as retail, agriculture, education and healthcare. The GSMA launched its Digital Inclusion programme in April 2014 to expand global connectivity and increase mobile internet adoption. The programme will collaborate with mobile operators, governments, internet players and non-government organisations to address four key barriers to mobile internet access. Network infrastructure and policy Increasing network coverage to currently unserved areas. Affordability and taxation The combination of low incomes, the cost of the device, charging fees, and data plan payments creates an affordability barrier to accessing the mobile internet. This issue is compounded by government taxes and fees, such as airtime taxes and handset taxes. Consumer barriers Illiteracy, digital illiteracy and lack of internet awareness are consumer barriers to mobile internet adoption. Local content The availability of content that is both local language and locally relevant can play a vital role in the adoption of mobile internet Mobile empowering people and society

243 THE MOBILE ECONOMY Network infrastructure and policy: Extending coverage to currently underserved areas Network coverage is critical for access to the mobile internet. At the end of 2014, mobile operators reached approximately 85% of the global population with 2G and around 55% with higher speed mobile broadband (3G) networks. Despite the progress to date, there are significant populations across the world who still do not have access to mobile services. A large proportion of the still unconnected population live in rural and in some cases geographically remote areas. These areas have additional challenges such as a lack of electricity infrastructure and low road density, which can provide extra obstacles to extending network coverage. The economic case for mobile operators to expand networks in these areas is often challenging. Industry solutions to extending coverage Infrastructure sharing is a solution that can help reduce the cost of extending network coverage, particularly into remote or geographically challenging areas. Network sharing can also significantly increase capacity in urban areas particularly, for example, where operators are looking to deploy small cell technology. It has the potential to strengthen competition and reduce the carbon footprint of mobile networks, while reducing costs. Several operators in conjunction with the GSMA, announced an agreement in early 2014 to cooperate on network sharing initiatives and to help provide mobile broadband access to unserved rural communities and drive down the cost of mobile services for all sections of the population. 8 Mobile operators in many developing regions are employing a wide array of solutions to tackle the challenge of off-grid connectivity, including the growing use of green options such as solar, wind, water, biomass and fuel cells. Operators have built approximately 43,000 renewable energy sites globally since using a range of technologies. Such alternatively powered cell sites can significantly decrease diesel costs (which is often the primary power source in areas off the main electricity grid) and therefore reduce operating costs and maintenance burdens of rural cell towers. Government solutions to extending coverage Recent research by Frontier Economics examined the role of network competition in promoting rural mobile coverage. 10 Evidence from more than 200 countries over a 15 year period shows that network competition has driven mobile coverage further and faster than has been achieved by single networks. There are also complementary measures that governments can take to promote mobile coverage in rural areas. For example, there may be some rural areas where the economics are too challenging for network competition to deliver coverage, which may require some form of government intervention Mobile empowering people and society 33

244 THE MOBILE ECONOMY 2015 Source: Frontier Economics Complementary government policies to promote rural coverage Using public finances to (co)fund roll-out in uneconomic rural areas RURAL COVERAGE Promoting voluntary infrastructure sharing in rural areas Licensing of low frequency spectrum in combination with coverage obligations Affordability and taxation Unconnected populations in developing regions often have low income levels. According to the World Bank, over two billion people live on less than US$2 a day, which represents the poverty line in developing nations. This highlights the particular challenge of raising levels of digital inclusion amongst lower income populations. The mobile sector in many countries is the target of excess taxation that creates additional barriers to digital inclusion, especially in developing countries. Even though mobile is an essential service, it is often taxed at a substantially higher rate than other sectors. While these taxes are often imposed to meet short-term fiscal targets, they come at the cost of immediate and long-term benefits from increased access to mobile internet, and ultimately greater government revenue. 11 Mobile taxation includes taxes on handset purchases, subscriptions and consumption of services, as well as a number of sector-specific levies on operators. In a recent survey of 19 markets, taxes accounted for more than 30% of sector revenue in over half the markets surveyed, and more than 40% of sector revenue in Jordan, Tunisia, and Brazil. In Bangladesh and Turkey, taxes accounted for almost 60% of sector revenue. Taxes on mobile consumers restrict access and usage by reducing affordability, while taxes on operators limit incentives for investment, for example into expanded network coverage. As a result of these negative impacts, high levels of taxation have caused some communications ministers (e.g. Nigeria and Brazil) and international institutions (e.g. the ITU and the Broadband Commission) to call for reduced mobile taxation. 12 For the sector to achieve its potential, governments should harmonise mobile taxes to the general tax level. 11. Mobile Taxes and Fees: A toolkit of principles and evidence, GSMA/Deloitte, Nigeria: Brazilian: ITU: Broadband commission: https://itunews.itu.int/en/506-broadband-commission-for-digital-development-delivers report.note.aspx 34 Mobile empowering people and society

245 THE MOBILE ECONOMY 2015 Measures to improve affordability There have been considerable efforts over recent years both by mobile operators and other ecosystem players, including device manufacturers, to improve the affordability of mobile services. Operators have also played a role in improving the affordability of mobile services, particularly with regard to mobile data and internet access, and the challenge of making these services available to low income consumers on prepaid tariff plans. 13 For example, a range of flexible tariffs are now available from operators focusing on mobile data. These allow users to access the internet for a period of time or alternatively allow access to a particular web property. Smartphone adoption will be enabled by ongoing reductions in handset pricing. As discussed earlier in this report, a number of smartphones in the US$25-50 price range were launched over the last year, with new models from several handset manufacturers. These launches only mark the start of a price expansion trend towards low-cost levels that will spread to more developing economies. Recent research by the GSMA s Connected Women programme has found that cost remains the greatest barrier overall to owning and using a mobile phone, particularly for women, who often have less financial independence than men. Improving affordability, especially of handsets, will disproportionately benefit women and help to close the gender gap in mobile phone ownership Consumer barriers: Literacy and awareness In 2011, there were 774 million illiterate adults across the world. In developing countries illiteracy is most prominent across rural areas and marginalised groups. This causes a major challenge for these individuals when looking to access internet content which is predominately text based. Combined with an overall lack of awareness about the internet and its potential uses and benefits, these factors create a significant barrier for mobile internet adoption, even where coverage and affordability issues have been addressed. Building awareness of the internet and needed skills in the developing world will require help from multiple stakeholders to deliver awareness building campaigns and literacy training programmes. Measures to address consumer barriers Mobile operators have been at the forefront of the drive to improve awareness of the internet and technical literacy amongst their subscriber bases. This is being achieved through a rage of measures, including direct marketing or by holding regional events and exhibitions to showcase new data products and services. For example, MTN in Uganda launched the MTN Internet Bus to deliver information and communications technology (ICT) education to rural populations. The bus has computer work stations linked to MTN s 3G and LTE networks, and aims to provide training in basic computing skills and e-learning. Governments also have an important role to play in funding and supporting initiatives in areas such as general education and promoting the use of ICT as a medium for learning in schools and colleges. For example, the South Africa government launched its South Africa Connect strategy in This broad programme aims to focus on a range of issues that impact on digital inclusion. The strategy specifically includes plans to introduce an e-readiness campaign in schools as well as more widely focused e-literacy campaigns. 13. https://gsmaintelligence.com/files/analysis-subscription/?file= prepaid-data-tariffs.pdf Mobile empowering people and society 35

246 THE MOBILE ECONOMY Local content A recent GSMA report highlighted that content and services that are relevant, accessible, and available to the users in their own language will be crucial in bringing the full benefits of the mobile internet to users. 14 However, the majority of content is currently in English and is largely focused on data-heavy smartphone apps. Smartphone penetration is still low in the developing world and English is not the primary language for the majority of the population, thereby limiting accessibility and usefulness of the content. In order to reach the widest audience, content needs to be available on as many devices as possible in languages users understand, as well as being relevant to their local needs and interests. Measures to provide more local content There are encouraging signs of progress in developing content that is both more relevant to consumers and also more accessible by being produced in the local languages. There are also growing moves to develop content for feature phones as opposed to smartphones, recognising that feature phones will remain a primary device in many developing markets for a number of years to come. Services like the below are helping address the local content gap in developing countries, with the aim to introduce feature phone users to the benefits of the mobile internet, and instil in them a use pattern that will continue once they transfer over to smartphones. HiviSasa: A free online newspaper in Kenya that engages local users, especially young people, to come online and contribute as citizen news reporters. Every1Mobile: A South African-based company that builds and manages online communities that offer young people opportunities to discuss topics related to health, education, jobs and entertainment with their peers and subject matter experts on mobile. The company has a strong presence in seven countries across the region, and with a growing presence in twelve more. 15 Mxit: A South African social networking app for feature phones with over seven million users 16 offering functionality in areas with a weak 2G signal. binu: An Australian start-up company offering a Java app that can emulate the functions of a smartphone on a feature phone such as updating Facebook and Twitter, reading news and searching the internet https://gsmaintelligence.com/analysis/2014/09/local-world-content-for-the-next-wave-of-growth/445/ Local world content for the next wave of growth Mobile Internet for the Next Two Billion, binu, May Mobile empowering people and society

247 THE MOBILE ECONOMY Delivering financial inclusion across the world Mobile money is a rapidly maturing industry that is bringing financial inclusion to a growing number of previously unbanked and underbanked populations across the developing world. Mobile money services are now available in over 60% of the world s developing markets, providing unbanked customers with choice, security, convenience and affordability that are often missing in cash-based operations. Mobile money providers are working hard to increase the quality, reach and sustainability of their services. Through industry-led initiatives, including partnerships with banks and other third parties, providers are building a broader payments ecosystem, enhancing the customer experience and reaching scale to evolve the sector to a new phase of maturity Rapid growth of mobile money services In the past five years, mobile money services have spread across much of Africa, Asia, Latin America, Europe and the Middle East. As of December 2014, there were 255 live mobile money services in 89 markets compared with 233 live services across 83 markets at the end of Whilst Sub-Saharan Africa still accounts for the majority of live services globally (53%), half of all new launches in 2014 occurred outside the region. In 2014, 22 new services launched compared with 60 in 2013 and 58 in Mobile money was rolled out in six new markets in 2014 Dominican Republic, Myanmar, Panama, Romania, Sudan and Timor- Leste compared to 11 new markets in 2013 and 14 in Mobile empowering people and society 37

248 THE MOBILE ECONOMY 2015 Number of live mobile money services by region ( ; year-end) As of December 2014, there were 255 live mobile money services in 89 markets EAST ASIA AND PACIFIC SUB-SAHARAN AFRICA SOUTH ASIA Middle East and North Africa Latin America EUROPE & CENTRAL ASIA 38 Mobile empowering people and society

249 THE MOBILE ECONOMY 2015 Globally, as the number of players in individual markets increases, competition is intensifying. Worldwide, 56 markets now have at least two live mobile money services, and 38 of these markets have three or more live services. As markets become increasingly competitive, mobile operators are showing a growing interest in the development of interoperable solutions. In 2014, operators interconnected their services in Tanzania, Sri Lanka and Pakistan. In addition, operators in six other markets have already committed to interconnect their services. It is expected that account-to-account interoperability will increase transaction volumes and revenues by making it easier for consumers and businesses to send money domestically across networks. Number of live mobile money services for the unbanked by country (December 2014) THREE OR MORE MOBILE MONEY SERVICES TWO MOBILE MONEY SERVICES ONE MOBILE MONEY SERVICE interoperable markets The number of registered mobile money accounts grew to reach 299 million globally at the end of December Sub-Saharan Africa still has the majority of mobile money accounts and, in East Africa, one in two mobile connections is linked to one. Globally, there are 103 million active mobile money accounts, a 41% increase on 2014 figures. The size and reach of mobile money agent networks are increasing fast, and now out-size traditional financial and remittance service networks. The size of the global mobile money agent network grew by 46% in 2014 to a total of 2.3 million outlets globally. In three-quarters of the markets where mobile money is available, agent outlets outnumber bank branches Data on bank branches come from the IMF Financial Access Survey (FAS) Database. Available at Mobile empowering people and society 39

250 THE MOBILE ECONOMY Building a payments ecosystem The range of payment services offered by mobile money providers is increasing, enhancing the attraction of mobile money propositions and delivering even greater benefits to customers. These new services include international remittances, merchant payments, and bulk payments such as salaries and government to people transfers. Just over 23% of all the value moving through mobile money systems globally in December 2014 was processed in transactions involving an expanding ecosystem of institutional and business users. In December 2014, mobile money users transacted a total of US$16.3 billion through million transactions globally Impact of regulation on the development of the industry The full potential of mobile money deployments remains untapped in some regions, such as Latin America and South Asia, in part due to regulatory barriers. Major markets such as Egypt, Indonesia and Nigeria are also awaiting for reform that would unleash the potential of mobile money. Nonetheless, an increasing number of regulators are recognising the major role that non-bank mobile money providers can play in fostering financial inclusion and economic growth and are establishing enabling regulatory frameworks for mobile money. In 2014, new regulations were issued in Colombia, India, Kenya and Liberia. 19 Today, in 47 out of 89 markets where mobile money is available, regulation allows both banks and non-bank to provide mobile money services in a sustainable way Mobile insurance, credit and savings: A growing range of services The development of other mobile financial services including mobile insurance, savings and credit will allow service providers to deepen financial inclusion by offering services beyond money transfer and payment. Mobile credit services became much more widely available in 2014 fuelled in part by new partnerships between mobile operators and banks. Of the 32 live services across the globe, 12 launched in Mobile credit services are now available in 15 countries globally. Mobile insurance services are now scaling, as the industry appears to have found the right partnership and commercial models. Sixteen new mobile insurance services were launched in 2014, taking the total to 100 live services, and as of June 2014, 17 million policies had been issued. Financial services companies and mobile operators are increasingly making use of the mobile money infrastructure to offer saving facilities: 10 million dedicated mobile savings accounts have been opened worldwide. Some service providers are beginning to pay out interest accrued on the trust or escrow account, thereby incentivising customers to use their mobile wallet Mobile empowering people and society

251 THE MOBILE ECONOMY Mobile addressing social challenges in developing markets Affordable mobile phones and the opportunities they provide for the poor is already becoming one of the most dramatic game changing technologies the world has ever seen. Mobile can deliver cost effective solutions to address a range of social challenges in areas such as access to basic healthcare and education. These are particular challenges in many developing markets where there is limited funding available for government services and with a large proportion of the population living in rural and often inaccessible areas. Countries across the developing world face multiple social and environmental challenges today. Those that have large rural populations in particular face difficulties providing access to basic infrastructure and services such as electricity, education, healthcare and banking. Urbanisation rates vary across many developing regions, though urban growth has translated into an increasing number of people living in informal settlements, increasing poverty and inequality. A diverse range of players, including mobile operators, entrepreneurs, corporates, governments, investors and non-profit organisations have together driven an explosion in mobile-enabled products and services across the developing world, aimed at addressing the challenges outlined above. The following chart shows that the number of new service launches has increased consistently over recent years, with a particular focus on mobile health and mobile money. Source: M4D Impact Tracker, GSMA Intelligence Mobile-enabled products and services in the developing world PRE HEALTH LEARNING MONEY OTHERS Note that others includes Disaster Response, Energy Access, Green Networks, midentity, NFC and Smart Cities Mobile empowering people and society 41

252 THE MOBILE ECONOMY 2015 A diverse range of players, including mobile operators, entrepreneurs, corporates, governments, investors and non-profit organisations have together driven an explosion in mobile-enabled products and services across the developing world 42 Mobile empowering people and society

253 THE MOBILE ECONOMY 2015 mhealth: Facilitating health service delivery Mobile is already facilitating the delivery of basic health services to underserved populations, by successfully addressing challenges such as reducing maternal and infant mortality rates, combatting infectious diseases, creating awareness of HIV and delivering nutritional health and treatment for a variety of health conditions remotely. Mobile-baby is an mhealth service being implemented in Nigeria and Tanzania (known as Safer Deliveries in Tanzania). This service is the result of a partnership between Etisalat, Qualcomm, D-Tree International and Great Connection. The service is aimed at reducing mother and child mortality by helping pregnant women in rural areas reach hospital. Mobile-baby allows medical practitioners to send ultrasound images, video clips and 3D scans directly from ultrasound machines to mobile phones via SMS, MMS and , providing real-time remote medical diagnostics. magriculture: Empowering smallholder farmers One billion people in the developing world are engaged in agriculture, 40% of the total workforce. Mobile services now give farmers the opportunity to stay abreast of pricing movements, gain access to current market information, take advantage of financial services and supply chain solutions thereby helping them to increase their yield, cost crops or livestock more competitively and grow their business. Tigo Kilimo was launched in 2013 by the Tanzanian mobile phone operator Tigo. The service aims to increase productivity for farmers through providing a range of information covering for example weather forecasts, best practices tips and market prices for the major crops grown in the country. Disaster Response Mobile networks can play an important role in disaster response and crisis management, given their resilience and ability to facilitate critical communication between humanitarian agencies, affected populations and the international community. The GSMA is assisting Ministries of Health in Nigeria, Senegal, Guinea, Liberia and Sierra Leone to mobilise the required support of mobile network operators in formulating a timely and effective response to the Ebola outbreak. These coordinated efforts will also involve collaboration with the World Health Organization and ECOWAS. Green Power for Mobile In 2008, the GSMA established the Green Power for Mobile (GPM) programme with an objective to Extend the coverage, reduce the cost and minimise the environmental impact of mobile networks by championing renewable energy. The programme, in partnership with the International Finance Corporation, supports operators to resolve these challenges by developing a body of knowledge and insight, supporting a thriving vendor ecosystem, hosting working group forums, assessing latest technologies and resolving financing challenges. There are about 640,000 off-grid sites globally, primarily powered by diesel generators. GPM has to date catalysed the adoption of green power of almost 43,000 live and planned green sites, details of which can be found on the Green Deployment Tracker. 20 M4D Utilities Globally, it is estimated that 411 million people have access to a mobile network before they have access to energy and 165 million people have access to a mobile network before they have access to an improved water source. The extensive reach of mobile infrastructure, combined with the increasing take-up of mobile financial services and the growing range of M2M services is generating new ways to bring access to reliable energy and clean water. A key trend is the development of low cost GSMenabled smart meters for water and electricity. These in turn allow the development of pay-as-you-go (PAYG) solutions that allow access to water and electricity to be controlled remotely depending on a customer s credit. These solutions have gained particular traction in East Africa, with for example PAYG solutions sold to more than 60,000 customers in Mobile for Employment Millions of young people in developing countries are out of employment, with some unemployment rates reaching up to 27%, for example in North Africa. Some of the main reasons for youth unemployment include lack of appropriate skills and experience, communication barriers, lack of knowledge of jobs available, and inability to travel to work. Many of these challenges can be addressed by mobile technology. Duma Works is a mobile-based networking and recruiting tool in Kenya that connects employers and job seekers, with a particular focus on small and mediumsized enterprises. Job seekers also have the ability to recommend friends and colleagues for other jobs, and receive referrals themselves, building on the power of personal and professional networks. Since its founding in 2012, Duma Works has dealt with over 10,000 jobseekers and generated over 5,000 job matches Mobile empowering people and society 43

254 THE MOBILE ECONOMY Connected Women: Addressing issues of gender inequality Gender inequality in the mobile industry remains an issue across both developing and developed countries. Women tend to be left behind, not only as consumers of mobile services, but also as employees and leaders in the mobile industry. Research in low- and middle-income countries indicates that there are notable and persistent gaps in women s access to mobile and its services which meet their needs. Recent research by the GSMA Connected Women programme suggests women in low- and middle-income markets are 14% less likely to own a mobile phone than men, which translates approximately into 200 million women. While substantial progress toward phone ownership for women is evident on an aggregate level, certain regions, such as South Asia, have made little progress in closing the gender gap in mobile ownership (women of the region are in fact 38% less likely to own a phone than men). Other regions, such as Sub-Saharan Africa, have a 13% gender gap in mobile phone ownership overall, but certain countries exhibit high gender gaps in mobile phone ownership. For example, Niger and the Democratic Republic of Congo had gender gaps in mobile phone ownership of 45% and 33% respectively. A mobile gender gap also exists at the usage level even when women own mobile phones, they tend to use mobile services less frequently or less intensively than men. Closing the usage gender gap will enable women to gain access to the life-enhancing services of mobile, including mobile internet. Additionally, closing the gender gap in terms of mobile ownership and usage in low- and middle-income countries could unlock an estimated US$170 billion market opportunity for the mobile industry in the next five years. Cost was identified as the greatest barrier overall to owning and using a mobile phone, particularly for women, who often have less financial independence than men. Security and harassment also emerged as an important barrier to mobile phone ownership and usage, and is a key concern for women. Other barriers include service delivery issues (network quality and coverage and agent or operator trust) and technical literacy and confidence. Finally, it was found that social norms tend to influence women s access to and use of mobile technology, and often contribute to women experiencing barriers to mobile phone ownership and use more acutely than men. 44 Mobile empowering people and society

255 THE MOBILE ECONOMY 2015 If the mobile ownership and usage gender gaps are not closed, women are at risk of falling further behind as new mobile services are developed, and particularly as mobile is the main means of accessing the internet in many developing countries. According to the UN E-Government Survey in 2014, the causes of this gender divide can stem from disparities between men and women in terms of a lack of education, lack of income, social attitudes towards female usage of technology, women having to balance their roles of mother and worker and lack of Internet content relative to women s needs. 21 Women are critical to realising the potential socio-economic benefits in the areas of health, education and agriculture, and have the most to gain in terms of financial inclusion via mobile services. As a result, closing the gender gap in mobile ownership and use, including for mobile internet, as well as developing mobile services which meet the needs of women, is a key consideration for operators and policymakers to realise the potential contribution of mobile services. The issue of gender inequality in the mobile industry extends beyond female consumers in developing markets. For example, across the world women are less present in many high-growth fields like science, technology and engineering which are important to countries innovation, connectedness and competitiveness in global market. Similarly, women are underrepresented in senior management roles in the technology sector. Women today comprise 40% of the global workforce and account for more than half of university graduates, and yet only 4.8% of Fortune 500 CEOs and five FTSE 100 CEOs are women. 22 This is an important issue and there is a consistent body of research asserting that gender diversity in the workplace is key to foster growth and innovation, as well as to improve performance. For instance, a study by Catalyst, which focused the Fortune 500, found that companies with the highest representation of women board directors achieved a 53% higher return on equity, 42% higher return on sales and 66% higher return on invested capital than those with the lowest representation. 23 There are a number of examples from across the developing world of services focused at addressing these issues. Ooredoo specifically targeted women and the potential for mhealth services when launching its 3G services in Myanmar in In a generally under penetrated market, the company found that family and personal health were major concerns for the local population, in a country with high maternal mortality rates. With the support of both local and international partners, Ooredoo has launched a maternal health app, maymay, to help attract users and address a key need in the community. In Togo and Benin, 61% and 54% of the population lives in rural areas, respectively, half of which are women. Etisalat received a grant by the GSMA Connected Women programme to develop a bespoke brand aimed directly at rural women: Weena. The holistic mobile offering comprises a tariff plan, a loyalty mechanism, value added services, a distribution network and marketing and communication specifically aimed at the rural women segment. The Weena proposition will not only yield social benefits to the rural women of Togo and Benin, but will also help to increase Etisalat s female customer base in these two markets Fortune.com assessment, June The Bottom Line: corporate performance and women s representation on boards, Catalyst, Mobile empowering people and society 45

256 DELIVERING THE DIGital future Connected Living: Mobilising the Internet of Things M2M CONNECTION GROWTH WILL COME FROM A RANGE OF SIGNIFICANT VERTICALS INCLUDING AUTOMOTIVE, HEALTHCARE, UTILITIES AND SMART CITIES GSMA INTELLIGENCE FORECASTS THE NUMBER OF GLOBAL CELLULAR M2M CONNECTIONS TO REACH ALMOST 1 BILLION BY 2020 ASIA PACIFIC WILL ACCOUNT FOR OVER 50% OF CELLULAR M2M CONNECTIONS BY 2020 THE GSMA SUPPORTS THE ACCELERATION OF M2M ADOPTION VIA: GROWTH COST LONGTERM PLANNING AND HIGHER RETURNS FROM BUSINESS MODELS IMPLEMENTING INDUSTRY LEADING SECURITY STANDARDS TO MAINTAIN CONSUMER CONFIDENCE Mobile is at the heart of future commerce MOBILE TO TAKE GROWING SHARE OF DIGITAL COMMERCE US$2,356bn US$1,471bn E-COMMERCE US$626bn US$204bn M-COMMERCE BY END SIM-based NFC launches with over 60 operating as commercial services around the world

257 Mobile identity: Unlocking the potential of the digital economy PROBLEM The password is broken, with multiple login details, lengthy registration processes demanding personal information and high profile security breaches, consumers are growing wary of online safety and privacy 86% OF PEOPLE HAVE LEFT A WEBSITE WHEN ASKED TO REGISTER 40% ADMIT TO USING THE FORGOT PASSWORD FEATURE ONCE A MONTH 78% OF CONSUMERS STATE ITS HARD TO TRUST COMPANIES WITH THIER DATA SOLUTION By using the internet security of the device thats always with customers; the mobile phone, secure and convenient access to digital services can be unlocked SOMETHING I KNOW SOMETHING I HAVE = MORE SECURITY LOGIN REQUEST DESKTOP TABLET OR MOBILE AUTHENTICATION AUTHENTICATION REQUEST MOBILE PHONE PARTICIPATING MOBILE OPERATORS Network 2020: The IP Communications landscape in 2020 Voice & video calls over LTE 360 Commercially launched LTE Networks in 124 countries Rich Communications 41 OPERATORS LAUNCHED IN 32COUNTRIES HD Voice HD VOICE NOW COMMERCIALLY AVAILABLE ON 125 HD Voice Mobile Networks IN 77 COUNTRIES Source: GSA & GSMA, February 2015

258 THE MOBILE ECONOMY Delivering the digital future 3.1 Mobile at the heart of the new digital ecosystem Mobile has already redefined consumer experiences in many aspects of daily life, as well as creating a range of new business opportunities and services. As technology and the broader mobile and digital ecosystems continue to evolve, the impact of innovation and disruption will be felt ever wider. New technologies, imaginative use cases and business models are likely to generate even more profound mobile innovations, increasingly linking the digital and physical worlds. As mobile disrupts and affects more areas of consumer and business life, the potential for collaboration also grows. Digital ecosystem players, ranging from mobile operators to new entrants to existing players in adjacent industries, will increasingly recognise the need for collaborative innovation rather than competition if they are to realise the full potential of mobile. The global app economy is growing rapidly, with revenues from apps and related products and services reaching US$86.3 billion in 2014, 25 representing a 26% increase from Much of this growth is coming from developing markets such as India and China, and out of a total global mobile developer population of 2.3 million in 2013, Asia is home to around a third. 26 In 2014, Google more than doubled the number of apps in its Google Play store, ending the year with more than 1.4 million and overtaking Apple with 1.2 million. Amazon, while a long way behind with 293,000 apps, also enjoyed impressive growth, albeit from a much smaller base. 27 These three platforms and their broader ecosystems currently dominate the global market, accounting for over 90% of smartphone sales worldwide. However, these platforms originated in developed markets, such as the US, Europe and parts of Asia. The developing world poses new challenges to the industry as customers are mostly prepaid, handset subsidies are less prevalent, and the use of mobile data is rising despite smartphone penetration being low. There is also an added importance to customers of brand value and locally relevant content, so importing established content models from developed markets is unlikely to drive customer acquisition to mobile platforms. 25. The European App Economy 2014, VisionMobile, August Delivering the digital future

259 THE MOBILE ECONOMY 2015 In the face of these differing market dynamics, there are a number of budding ecosystems attempting to use the web itself as their platform, for example, Mozilla (Firefox), Samsung (Tizen), Canonical (Ubuntu) and Jolla (Sailfish). These all support HTML5, which has the advantages of strong developer support, open standards, the ubiquity of the internet and access to services. However, operator support is key if these platforms are to achieve scale and meaningfully compete against Android and ios. While some operators are showing support and positioning these as low cost alternatives to Android, uptake of these platforms to date remains relatively slow. Mobile operators themselves are moving beyond traditional core voice, SMS and data propositions to embrace new technologies and business models, and are continually expanding their product portfolio for consumers. Additionally, operators are investing in content to leverage their network assets and commercial presence in support of innovation. Operators have formed a number of hubs and partnerships across the world to promote innovation and the development of new technologies: In May 2014, Orange established a new incubator partnership in Niger, its third in Africa. The project, called CIPMEN, is an incubator for small-to-medium enterprises that will encourage innovation in Niger. It offers support from their launch phase, up to when they are mature and stable enough to make enough profit to sustain themselves. In June 2014, Vodafone announced it is moving its research facility from Silicon Valley to London. The Vodafone xone hub in London will provide a wide range of incubation and investment services for startups, and will be responsible for developing mobile technologies. The focus will be on building and testing prototype devices, as well as developing apps and services tailored for the group s biggest markets in Europe, Africa and India, in order to bring our product development team closer to the customers it serves. 28 The company already has a number of existing xone bases in Europe, including both Dusseldorf and Milan. In early 2014, Telefónica announced it was expanding its Wayra Accelerator Network to Asia via a collaboration with China s VIV incubator. This will be the 14 th Wyra Academy globally, the other being mostly concentrated in Latin America and Western Europe, each of which incubate a yearly crop of tech start-ups in the hopes of spotting the next big thing. 29 According to CB Insights, the final quarter of 2014 saw the largest amount of investor activity (including venture capital, private equity, angel and other investments) in the mobile and telecommunications sector. There was a total US$12.5 billion of funding and 623 deals in the quarter alone. This was up from a previous high of US$5.2 billion of funding in third quarter of 2012, and 583 deals in the final quarter of Additionally, the mobile, telecommunications and internet sectors combined accounted for 36% of funding across all sectors in 2014, up from 28% in CB Insights Delivering the digital future 49

260 THE MOBILE ECONOMY Connected living: Mobilising the Internet of Things (IoT) The M2M market has seen rapid growth globally over recent years as operators and governments recognise the potential of the IoT, and the number of active deployments is beginning to accelerate. The IoT has the potential to play a significant role in realising the potential of the digital future, with positive impacts on both the economy and the broader society. There were 243 million cellular M2M connections globally as of December 2014, and growth is set to accelerate over the next few years. The total number of cellular M2M connections will grow at a CAGR of 26% between 2014 and 2020, bringing the total to just under one billion. However, these forecasts are based on known and tangible deployments, opportunities and barriers for each operator and M2M sector. They may well prove conservative if both industry players and governments successfully enable a number of growth factors, as highlighted in recent research by GSMA Intelligence. 31 This could lead to growth rates similar in the future to those witnessed over the past few years, resulting in an uplift over the current trajectory forecasts of up to 50%. Such growth factors include low power network roll-out, increased standardisation, improved regulation, and greater assurance of end-to-end security. A further requirement is the development of new operator business models in the M2M space, particularly those that move beyond simply providing the mobile connectivity. By the end of 2014, China was the largest cellular M2M market with 73 million cellular connections, or 30% of the global total. The US is second with 42 million connections, followed by Japan with 11 million. Overall, the top 10 markets accounted for 73% of global connections. By 2020, China will have increased its share to 36%, with 355 million cellular M2M connections. 31. https://gsmaintelligence.com/analysis/2014/9/cellular-m2m-forecasts-and-assumptions /446/ 50 Delivering the digital future

261 THE MOBILE ECONOMY 2015 Source: GSMA Intelligence Global growth of cellular M2M (Connections M) 10% 8% % of total connections 2% 2% 1% 3% 3% 4% 5% 6% 7% Delivering the digital future 51

262 THE MOBILE ECONOMY 2015 The IoT is beginning to transform everyday activities There are increasingly tangible benefits for consumers from the growing range of new connected services, with a strong focus on both wearable devices and the potential of the Smart Home. A number of new wearable devices launched over the last year, including the Samsung Gear and the LG Lifeband. Connected devices and sensors could deliver a smarter, more efficient home, with smartphones and tablets interacting with various connected objects and devices, from lighting to basic security systems. Additionally, some of the IoT s key players are connecting everyday items like home security systems and kitchen appliances to networked devices like PCs and smartphones, giving consumers greater control and management of their everyday lives. Samsung recently announced that by 2020, all of its products, from smartphones to refrigerators, would be internet connected. 32 Additionally, Qualcomm announced its intentions to act as an enabler, not only supplying the hardware, but also driving standards and providing an entire, interoperable umbrella platform from base-level connections to sophisticated processing and big data gathering, right up to the applications layer. 33 Many of the connected devices that will deliver IoT services in the smart home will use short-range wireless technologies, rather than cellular connectivity. However, mobile will continue play a crucial role as an enabling technology, acting as an aggregator or hub to connect a range of devices, and offering wide-area connectivity. The fastest growing industry M2M sectors are in the areas of automotive, security, compliance reporting, utilities and health. In the automotive sector, operators in both North America and one or two other markets (including for example Australia) are now deploying 4G-LTE devices, whilst most other regions are still working with 2G/3G. In-vehicle technology is an increasingly important differentiator in North America with a range of new services on offer. There have been a number of recent announcements across the global automotive market Delivering the digital future

263 THE MOBILE ECONOMY 2015 AT&T, keen to make it easier to connect multiple devices to a shared data plan, launched its Mobile Share Value tariff, where existing customers can add a 4G LTE-equipped vehicle, cellular smart watch, or communication and location devices for US$10 per month the same cost as a connected tablet. 34 In June 2014, Vodafone announced plans to take over Italian based telematics firm Cobra Automotive Technologies as it looks to move up the M2M value chain. The operator is offering 145 million for Italybased Cobra, which provides security and telematics solutions to the automotive and insurance industries. 35 In October 2014, China Mobile and Deutsche Telekom signed an agreement to establish a joint venture (JV) in the field of Connected Cars to address the Chinese automotive market, which reached 137 million units in The JV will be established in China, and each partner will hold a 50% stake, with Deutsche Telekom providing telematics expertise, the Connected Car platform technology and customer insights, and China Mobile its growing LTE network. 36 As M2M solutions are becoming more complex, operators are looking to deploy more advanced communication platforms and middleware that enables a smooth integration of devices, networks and applications, simplifying business processes such as billing and subscription management to scale down costs and optimise performance. Due to the demand for M2M interoperability and the growth of operator alliances, operators are moving towards a multiplatform approach, adding more standardised platforms from the key vendors. In order to implement such an approach, the majority of operators have entered into partnerships with specialised global providers such as Jasper, Amdocs and Ericsson, allowing them to offer similar user experiences across geographies. A few operators, namely Vodafone, Telefónica, Orange, Deutsche Telekom, Telecom Italia and Verizon Wireless, have also developed proprietary, in-house M2M operating platform, deploying solutions provided by Ericsson and Jasper on top. This multiplatform approach often comes from the fact that some of their larger enterprise clients are used to working with certain platforms, and do not want to integrate with a proprietary platform. There have also been a number of developments in the area of home automation. AT&T s Digital Life is a security and home automation offering, which is now available across the country. The service is managed through a smartphone app and includes home security monitoring, a smart thermostat and moisture detection. Operators in other regions are launching similar home automation services with, for example, Telefónica in Europe launching a trial service in late Orange has launched its Homelive service in France, which allows users to manage a range of connected devices Delivering the digital future 53

264 THE MOBILE ECONOMY The GSMA s role in realising the potential of M2M The initial focus of the Connected Living programme is to accelerate the delivery of new connected devices and services in the M2M market through industry collaboration, appropriate regulation, optimising networks and develop key enablers to support the growth of M2M in the immediate future. The ultimate aim is to enable the IoT, a world in which consumers and businesses enjoy rich new services, connected by an intelligent and secure mobile network. Working with its partners across the ecosystem, the GSMA is active in a number of areas to drive forward this initiative: Creation of fully interoperable specifications for remote SIM provisioning of M2M devices and and evaluation of convergence between M2M and smartphone/tablet solutions. Evaluation of the viability of existing networks to meet the low power, low data rate and low mobility use cases, as well as an assessment of the viability of alternative complementary networks including integration with macro network & spectrum/policy issues. Deployment of a GSMA sandbox platform to demonstrate how to monetise IoT Big Data. Device security requirements and an accompanying accreditation process for operators to demonstrate that solutions utilising their networks are secure. Engaging with partners across the broader ecosystem, the GSMA is also helping to facilitate the development and drive the adoption of new services in other sectors, with special interest groups for automotive, health, education, utilities and transportation. These groups allow manufacturers and mobile operators to meet and share information, enabling industry cooperation with the goal of resolving barriers to deployments and to accelerate the adoption of services. 54 Delivering the digital future

265 THE MOBILE ECONOMY Digital commerce Mass-market smartphones, fast connections and feature-rich applications are extending the convenience and interactivity of online commerce into the physical bricks and mortar world. Customers are able to look up product and service information, download and store vouchers, search for merchants, explore transport options, run price comparisons, buy tickets, purchase products and order services, all while on the move. At the same time, retailers, transport operators and other service providers are increasingly using mobile technology to improve the experience for their customers, while infrastructure rollouts of contactless point of sale (POS) terminals are no longer holding back service adoption. Digital commerce continues to show strong growth across most regions, with the industry forecast to grow significantly over the next three years, from US$1.7 trillion in 2014 to US$3 trillion in 2018, 37 a CAGR of 16%. Over the last 12 months there have been signs that the increasing availability of services and enabled-devices, supporting infrastructure and consumer readiness have led to a tipping point in the global migration towards mobile payments and commerce. A recent survey of users in the US and Europe found that nearly twice as many consumers started to use mobile payments in 2013 compared with 2012, 38 and that their use for in-store payments is growing rapidly. Mobile commerce is now forecast to account for 21% of total digital commerce by 2018, up from 12% in Mobile operators continue to lead the field in terms of new launches, building on the knowledge of their customer base and existing capabilities that allow them to enable bank-grade secure services. However, retailers and other stakeholders are becoming increasingly active by embracing mobile technologies to merge their online and physical stores into a distinctive consumer proposition to boost their revenues. This includes mobile apps, digital walk-in stores, social shopping (using social media to enhance the shopping experience), and interactive in-store services. Starbucks offers a mobile app that allows consumers to assign their credit card number and then simply scan the phone to pay, which has proved very successful. The company announced in its latest results that 15% of its US transactions, or around seven million mobile payments per week, took place via a mobile device. 39 Square is a financial services, merchant services aggregator and mobile payments company that allows users to take credit card payments on their mobile phones via a card reader and an app, announcing in November 2014 that it will start accepting payments via Apple Pay sometime in Moreover, PayPal has a mobile app that offers a broad range of services, including allowing users to transfer funds to others and make online purchases. The app can also be used for in-store purchases and has seen wide acceptance from merchants, with the company reporting that 1.9 million merchant locations across the US already accepted its payments at the start of Source: emarketer, Goldman Sachs and GSMA Intelligence Starbucks Mobile Dashboard, Q4 FY https://www.paypal-community.com/t5/paypal-forward/paypal-now-at retail-locations-in-the-u-s-including/ba-p/ Delivering the digital future 55

266 THE MOBILE ECONOMY 2015 Source: emarketer, Goldman Sachs, GSMA Intelligence Growth of the global digital commerce market (US$ bn) ,233 1,471 1,700 1,922 2,143 2, E-commerce M-commerce There has been substantial momentum in making mobile payment services available to consumers worldwide in the last year. There were 43 commercial launches of near-field communication (NFC) mobile commerce services during 2014, which brings the total number of commercial and pilot services to more than 300 to date. In addition, there were nine Host Card Emulation (HCE) pilots and commercial launches. NFC is still by far the most established technology for mobile payments, supporting secure element services such as the 300 operator services and ApplePay, as well as HCE services. There have been a number of interesting product and service launches in the past 12 months. Samsung collaborated with China UnionPay to launch an NFC payment service on 3.6 million of its smartphones in China, including the Galaxy Note 4, Galaxy Note 3 and Galaxy S4. Samsung is the first company to cooperate with UnionPay for an NFC payment service. Billing provider Boku has made the move into physical world payments after securing an e-money license that enables it to operate outside of the European Union s Payment Services Directive (PSD) regulations (which limits carrier-billing purchases to digital content). Following this move, Boku has partnered with UK mobile network operators O2, EE and Vodafone to allow customers to charge public transport tickets, takeaway food, cups of coffee, and other click and collect physical world purchases to their mobile phone bill. US telecoms firm AT&T announced that it will use technology from CorFire to power NFC Connect. This is a new platform that will enable it to offer a wide range of NFCenabled services outside of the Softcard mobile commerce venture it co-owns with Verizon and T-Mobile. Banco do Brasil has partnered with mobile network operator Oi and Visa to offer NFC mobile payments in Brazil. Users will be able to make payments with an NFC phone at any of the 1.6 million contactless POS terminals available throughout the country. 56 Delivering the digital future

267 THE MOBILE ECONOMY 2015 Last October, Apple introduced its Apple Pay service in the US, providing a boost to the mobile payments market. It has increased the overall size of the mobile commerce market, created strong consumer interest, and has stimulated competition between the mobile payments and wallet service providers and traditional cash, physical credit cards, and other payment mechanisms. Apple Pay activated one million cards within first 72 hours of launch. In November 2014, three weeks after launch, Whole Foods announced it had processed more than 150,000 Apple Pay transactions, McDonald s said Apple Pay accounted for 50% of its tap-to-pay transactions, and Walgreens said its mobile wallet payments had doubled. 42 The Apple Pay service uses the card schemes tokenisation platforms, enabling banks to digitise their customers account details for a range of mobile wallets through a single and simple entry point. The GSMA is working closely with mobile operators, card schemes and banks to provide a similar single entry point to the card scheme tokenisation platforms, using the SIM to secure payments, which will expand the reach of banks to their non-apple customers. Source: GSMA Intelligence Global commercial NFC-based mobile payment service launches (Cumulative) NFC HCE APPLE PAY 42. Delivering the digital future 57

268 THE MOBILE ECONOMY The GSMA s role in developing digital commerce Working with mobile operators, regulators, banks, retailers, transport operators and other service providers throughout the globe, the GSMA s Digital Commerce programme is active in driving the mass adoption of SIMsecured digital commerce services globally. The GSMA engages regularly with key government and regulatory bodies, providing advice and guidance on how to harness the potential benefits of SIM-based services in transport, retail and other sectors of the economy, and developing industry positions on aspects of policy, highlight the impact of regulation and informing regulators decisionmaking processes. As the number of commercial mobile commerce services around the world rises, the GSMA continues to promote the use of common standards to enable the global interoperability of services and generate economies of scale, liaising with other relevant stakeholders to ensure the consistency of the overall set of specifications involved in mobile commerce deployments. Covering many topics, these specifications set out a common framework of requirements to ensure interoperability and an efficient and consistent development and deployment of mobile commerce services. 3.4 The personal data opportunity As more people access the mobile internet, and as online commerce, social media, gaming and other activities gain popularity; consumers are increasingly voicing privacy concerns and demanding better protection of their data. 58 Delivering the digital future

269 THE MOBILE ECONOMY 2015 In the last year, there have been several high profile internet security breaches, highlighting potential serious flaws in the way personal information and data is currently protected online. ebay announced in March 2014 that its database containing encrypted passwords and other non-financial data had been hacked. 43 In April 2014, researchers announced a flaw in encryption standard used by many websites, the Heartbleed bug. This flaw left approximately 17% (around half a million) of the internet s secure web servers certified by trusted authorities vulnerable to the attack, 44 leaving them open to theft of private keys and passwords. A typical consumer has around 26 different online user names but only five different passwords, 45 leading to serious questions regarding the suitability of the current password model for online safety and privacy. Therefore, there is a growing need for individuals to be able to manage and authenticate their identities online in a more secure and convenient manner. For example, 70% of European consumers surveyed by Orange in 2014 agreed that there are few trusted ways to find out about personal data management and online protection. 46 One solution that is now gaining popularity is mobile authentication. The SIM card and the mobile platform can act as a frontline identity management service provider, meaning that secure and convenient access to digital services can be unlocked using the inherent security of the mobile phone. Mobile operators are well placed to provide identity services given their strong customer relationships, with the provision of other trusted services such as mobile banking further enhancing this relationship. Concerns around mobile identity and personal data 83% 81% 65% 48% 60% 82% of mobile internet users have concerns about sharing their personal information when accessing the internet or apps from a mobile of mobile users think it is important to have the option of giving permission before third parties use their personal information of mobile app users check what information an app wants to access and why before installing it of mobile app users with privacy concerns would limit their use of apps unless they felt sure their personal information was better safeguarded of mobile users want a consistent set of rules applied to any company accessing their location, regardless of how they obtain this information of mobile users want to know when, and what type of personal information is being collected from their mobile devices Source: GSMA Experian, 17 July Mobile Privacy: Consumer research insights and considerations for policymakers, GSMA, February 2014 Delivering the digital future 59

270 THE MOBILE ECONOMY 2015 Mobile identity offers the potential to extend access to a range of different services, including banking, commerce, retail and other identity based digital services. One area of particular focus concerns access to public and government services. In some developing regions, there may be a lack of basic infrastructure to allow individuals to establish and manage their identities, whilst in more developed economies governments increasingly need to develop new and secure ways of engaging with their citizens. There are a growing number of examples of services and deployments aimed at addressing these issues. E-government and public services: Mobile identity is transforming how citizens and governments interact. For example in Finland, the City of Helsinki 48 has launched a new tax receipt app which allows citizens to calculate the total amount of direct or indirect taxes they pay monthly. In Africa, Orange s mobile birth registration solution in Senegal and Uganda Telecom s Mobile Vital Records System platform have shown how mobile operators can use existing technology to solve a previously intractable social problem. 49 Cross border services: Mobile identity solutions are making it possible for citizens and businesses to deal with public administrations in other countries. In Europe, the recently launched electronic identity, authentication and signature (eidas) Regulation 50 is expected to enable and foster seamless electronic transactions across borders. This is making it easier and safer for individuals, businesses and public administrations in different EU countries to identify and authenticate themselves, sign documents and check the authenticity of documents online eidas Regulation (EU) No 910/2014 of 23 July 2014 OJ L257, 28 August Delivering the digital future

271 THE MOBILE ECONOMY The GSMA s role in further developing personal data and mobile identity services The GSMA s Personal Data programme is working with mobile operators who have launched identity services across the globe. The mobile industry needs to deliver common and consistent interfaces to a range of service providers, which at the same time need to offer seamless and convenient solutions to consumers. The use of standards and interoperability are therefore key, in particular the need to create a common, industry-wide set of identityrelated APIs (application programming interfaces). The GSMA is working closely with operators to establish a uniform set of APIs to underpin key mobile identity services. To help realise this goal, the GSMA has launched its Mobile Connect identity management service (supported by Axiata Group Berhad, China Mobile, China Telecom, Elisa, Etisalat, KDDI, Ooredoo, Orange, Tata, Telefónica, Telenor, Telstra and VimpelCom) which aims to facilitate broad interoperability across a range of mobile operators and service providers. 51 Mobile Connect is a fast secure login system that enables people to access their online accounts with just one click or less, and provides different levels of security from lowlevel website access to highly secure bankgrade authentication. People subscribed to a participating operator know that when they click on a website s Mobile Connect button they will can log in to any service or activity without the need to remember complicated username and password combinations Delivering the digital future 61

272 THE MOBILE ECONOMY Network 2020 Consumer appetite for mobile data and richer services is growing rapidly, and as a result new IP-based communications services such as Skype, WhatsApp and Facebook Messenger are becoming increasingly popular. These services will continue to gain traction with the growth of LTE networks and devices, meaning operators will need to consider which type of partnership or over-the-top integration models will allow them to drive revenue and sustain their business models in the longer term. Embracing an all-ip future solution is vital for operators if they are to retain customer relevance and have a network capable of meeting the ever-growing customer demand for data services and increasingly richer communications. One response to these challenges is for operators to develop a rich IP-based communications proposition for their customers. Rich Communication Services (RCS) offer voice and messaging, video-based calling, as well as Voice over LTE (VoLTE), and HD Voice services. RCS is the platform that enables the delivery of communication experiences beyond voice and SMS, providing consumers with instant messaging or chat, live video and file sharing across all devices, on any network. For operators, deploying RCS allows them to offer competitive solutions with alternative applications, which can lead to new revenue streams through the creation of apps and business-to-business services. RCS has been launched by 41 operators in 32 different countries, with a commitment from a further 87 operators to launch by the end of VoLTE offers operators many potential cost and operational benefits. It eliminates the need to have voice and data on separate networks, and can unlock new revenue potential through deployment in parallel with video calls over LTE and RCS multimedia services. As demand for video calls grows, deploying VoLTE will ensure that services are fully interoperable across the operator community. Six countries have launched VoLTE services as of December 2014, and 56 are in the planning stages. HD Voice meanwhile provides customers with high quality, clear and more natural voice calls with reduced background noise on mobile and fixed terminal networks. HD Voice services are rapidly expanding, and 125 mobile operators in 77 countries have commercially launched HD voice services as of February Delivering the digital future

273 THE MOBILE ECONOMY G is still a relatively nascent technology on a global scale, with just under 7% of total connections by the end of Although the rate of migration is accelerating sharply, there is already growing speculation and analysis around the next generation of mobile services, generally referred to as 5G. Discussions centre on whether 5G will be a true generational shift in connectivity technology or the consolidation of existing 2G, 3G, 4G, Wi-Fi and various other technologies to provide vastly greater network coverage and always-on reliability. Considerable advancements towards the hyper-connected society have already been made. Examples include technologies such as network function virtualisation (NFV), software defined networks (SDN) and heterogeneous networks (HetNets). All of these technologies are regularly bundled under the 5G banner, despite the fact that they are already being brought to market by vendors and invested in by operators. These technologies will continue to have a significant impact on the mobile industry over the coming years. However, placing too much focus now on a future, over-arching vision of a new technology generation could adversely affect progress in these areas between now and the anticipated launch of 5G as a commercial service. This latter point is especially true given the early stage of 4G adoption in many countries, particularly across the developing world The GSMA s role in developing next generation networks The Green Button Promise sums up the next generation network proposition. It reflects the previous experience of using a green button on a device to initiate a voice call, and will fulfil customers expectations that services initiated via the calling or messaging buttons will be high quality and fully interoperable, regardless of network. The GSMA s Network 2020 programme is designed to help operators make IP Communications work as a native service behind the Green Button Promise, aiming to address and navigate the complexities of progression to an all-ip communications network. The first phase of the programme focuses on helping networks deploy VoLTE and Conversational Video calls over LTE, and encouraging the RCS Ecosystem to help operators prepare for and launch RCS-based interoperable rich communication solutions for their customers around the world. Additionally, the Network 2020 programme will work with operators to determine the technical and commercial specifications for operator-to-operator Quality of Service (QoS) for IP Services, and encourage them to incorporate the QoS philosophy into their customer solutions. The Network 2020 programme also aims to help catalyse commercial implementations for IP Interconnect solutions between operators and service/content providers. Finally, until such time as the industry requirements and definition of 5G have stabilised, the GSMA will focus on improving the overall sustainability of the mobile telecoms sector, allowing more networks to achieve greater connection numbers by enhancing the business model for expanded coverage and offering connectivity to those in the world that currently have no connectivity at all. Delivering the digital future 63

274 Global enablers to spur investment and growth Spectrum a scarce resource Mobile data traffic is growing so rapidly that new mobile broadband spectrum will be essential to prevent a network slowdown GLOBAL MOBILE DATA EXPECTED TO INCREASE 10x from MHz THE ITU ESTIMATES MHZ NEEDED BY 2020 TO MEET DATA DEMAND Coverage bands (like 700MHz and 800MHz) will ultimately reach capacity putting mobile broadband at risk in emerging markets, rural areas and inside buildings <700MHz 700MHz 900MHz 1800MHz 2100MHz 2600MHz LOW FREQUENCY MOBILE BANDS TRAVEL FURTHER BUT HIGH BANDS HELP FOR CAPACITY ISSUES IN URBAN AREAS CELL SIZE

275 European mobile network operator mergers WHY IS CONSOLIDATION IMPORTANT? WHAT S HAPPENING? 2G 3G 4G CONSOLIDATION CAN BOOST INVESTMENT IN NEXT-GENERATION MOBILE INFRASTRUCTURE MERGER REVIEWS FOCUS ON SHORT-TERM PRICE IMPLICATIONS RATHER THAN LONG-TERM CONSUMER BENEFITS Consumers benefit from increased investment through new technologies that improve quality and reduce unit prices Q Q Predicted price based Effective on changes price per in EBITDA minute WHY IS CONSOLIDATION GOOD FOR THE MARKET? CONSUMER COST QUALITY AND SPEED INVESTMENT AND ROLL-OUT OF NETWORK AND SERVICES MERGERS CAN DRIVE INVESTMENTS REQUIRED TO PROVIDE LONG-TERM SOCIO-ECONOMIC BENEFITS FOR CONSUMERS AND BUSINESS

276 THE MOBILE ECONOMY Global enablers to spur investment and growth The mobile industry is increasingly helping governments across the world achieve their goals. Mobile technologies and services are fuelling economic growth and entrepreneurship, while yielding major social benefits by improving social cohesion, education, financial inclusion and healthcare. With a supportive regulatory framework, the mobile sector will continue to drive socio-economic progress, benefiting individuals, companies and governments alike. Although local factors mean that regulatory frameworks will differ from market to market, there are some general principles that apply across the globe. If policymakers and regulators encourage investment, competition and innovation, both the mobile sector and the wider digital economy will expand, creating prosperity and new jobs. 66 Global enablers to spur investment and growth

277 THE MOBILE ECONOMY Enabling and encouraging investment Entrepreneurs, enterprises and entire countries can benefit greatly from advanced mobile technologies, but only if mobile operators continue to upgrade their existing networks and deploy new networks. Mobile operators need to build a robust business case for new investment that will convince investors to provide the necessary finance. Investor willingness to fund the rollout of advanced mobile broadband networks depends in part on the applicable regulatory framework. Spectrum policy, restrictions on base station siting and network sharing, the tax and incentive regime and the extent to which operators can consolidate their mobile operations are all important factors in determining the business case for investment Spectrum: A scarce resource requiring careful management A precious and finite resource, radio spectrum is fundamental to the delivery of mobile services. Without sufficient spectrum, mobile calls will not connect, web pages will not load, apps will not update and videos will freeze. In many countries, sparse fixed-line networks mean the majority of the population is solely reliant on mobile connectivity to access the vast repository of information and services available over the Internet. Traffic on mobile networks is growing exponentially as more people get online and businesses go digital. Avoiding damaging congestion and realising the enormous socio-economic potential of mobile broadband connectivity will depend on the timely release of suitable spectrum. While management of this scarce resource is primarily the role of national governments, international coordination is crucial. Spectrum allocations need to be harmonised internationally to enable the industry to realise economies of scale and deliver services that work across national borders. Governments should, wherever possible, adopt internationally agreed band plans, thereby reducing the cost of developing and manufacturing a wide range of mobile equipment, from network base stations to handsets and tablets. Although mobile devices can support multiple bands, each additional band increases the device cost, reduces the receiver s sensitivity and drains the battery. Operators require large blocks of contiguous spectrum in order to provide mobile broadband connectivity on a costeffective basis. Moreover, spectrum needs to be licensed in a way that does not curb mobile operators ability to invest in new networks and services. Using spectrum licenses to raise as much cash as possible can be counterproductive in the mid-term. Levying excessive license fees on mobile spectrum will reduce the money available for network deployment, increase consumer prices and limit the potential economic benefits of mobile broadband. Global enablers to spur investment and growth 67

278 THE MOBILE ECONOMY The Digital Dividend is a high priority With mobile traffic rising rapidly and governments objective to bring coverage to most of their citizens, governments around the world urgently need to accelerate the switchover from analogue to digital television and deploy the spectrum being freed up in the process. This so-called Digital Dividend spectrum (700MHz and 800MHz frequency bands) is well suited to cost-effectively expand mobile broadband coverage into suburban and rural areas, while providing connectivity deep inside buildings. The ITU s deadline for countries to switchover from analogue to digital television is the 17 June However, to date, only 50 countries have allocated some of the digital dividend spectrum to mobile services. Source: GSMA Intelligence Digital Dividend spectrum assignments 50 COUNTRIES Digital Dividend spectrum assigned No national Digital Dividend assignments Wholesale network 68 Global enablers to spur investment and growth

279 THE MOBILE ECONOMY 2015 To prevent the process from becoming bogged down by bureaucracy, local and regional policymakers need to play a leadership and proactive role, facilitating dialogue between the key stakeholders, including broadcasters, the mobile industry, set-top box manufacturers and consumers. As with other frequency bands, the allocation of Digital Dividend spectrum should be harmonised internationally as much as possible. At the World Radiocommunication Conference (WRC) in 2012, regulators paved the way for global harmonisation of this band, and over the last two years there has been real progress on regional harmonisation. A majority of countries in Latin America and Asia Pacific have adopted the APT (Asia Pacific Telecommunity) band plan. If governments in Africa, Middle East and Europe adopt a band plan that reuses the lower duplexer of the APT band plan, equipment makers would achieve greater economies of scale, lowering prices for consumers and businesses Meeting future demand for spectrum The data traffic on mobile networks around the world is set to continue growing rapidly over the next decade. The ITU predicts that mobile broadband networks will need between 1340MHz and 1960MHz of spectrum to meet the anticipated demand in In most countries, that equates to an additional MHz beyond today s allocations. Ideally, this new spectrum should comprise a mixture of coverage (i.e., lower frequency) and capacity (i.e., higher frequency) bands harmonised globally, or at least regionally. The next opportunity to identify additional harmonised spectrum for mobile broadband is at the WRC in 2015 (WRC-15) in Geneva, where it is set to be the top agenda item. The outcome of WRC-15 will determine whether the telecoms industry can continue to meet the growing demand for affordable, ubiquitous, high-speed mobile broadband services. In the absence of new spectrum allocations at WRC-15, governments will find it difficult to release enough spectrum to accommodate the ongoing rise in mobile traffic, resulting in a poorer experience for end-users and potentially more expensive mobile services. It takes about eight to 10 years to re-allocate, re-assign and re-license spectrum, so WRC-15 will play a pivotal role in avoiding a capacity crunch in the next decade. Global enablers to spur investment and growth 69

280 THE MOBILE ECONOMY 2015 Through the GSMA, mobile operators are calling for four new bands to be allocated to mobile services and harmonised worldwide: 1. Sub-700MHz UHF ( /8MHz) 2. L-Band ( MHz) GHz 4. C-Band ( GHz) Radio Spectrum: Identified Mobile Bands MHz 1.8GHz 2.3GHz Digital Dividend (700/800MHz) 900MHz 2.1GHz 2.6GHz GHz Candidate Bands for WRC /8 MHz MHz MHz GHz GHz GHz Together, these new bands will provide the mobile industry with the means to improve both coverage and capacity. 70 Global enablers to spur investment and growth

281 THE MOBILE ECONOMY Consolidation versus fragmentation Facing both rapid growth in data traffic and increasing competition from online service providers, mobile operators need sufficient scale to invest in network infrastructure and the development of new services. In overly fragmented markets, individual operators have limited economies of scale and scope, impacting their ability to develop sustainable business cases for additional investment. Many emerging markets, in particular, are highly fragmented. A recent study by Frontier Economics for the GSMA highlighted the fact that in some countries there are five or more players with a market share of at least 5%. In markets with sub-scale operators, policymakers should review the antitrust framework and take steps to reduce constraints on market-driven restructuring. In some cases, policymakers may need to streamline the merger review process and impose less onerous remedies on those mergers that gain approval. When reviewing proposed mergers involving smaller operators, Frontier Economics recommends that competition authorities assess the role of these players across technology cycles, as technology upgrades are likely to be the main driver of consumer benefits in the longer-run. Whereas smaller operators may be willing to compete aggressively in the short-term to build up market share, they are likely to find it difficult to compete in the longer term. Of the emerging market operators that had a market share below 5% five years ago, only 10% have now managed to achieve a market share above 5%, according to the Frontier study. There are several ways in which consolidation gives operators greater incentives and capabilities to invest. Firstly, operators are likely to benefit from a larger customer base leading to economies of scale, which are widely recognised to be significant in mobile markets. Economies of scale improve the business case for expanding coverage to new areas and upgrading the capacity of networks in existing areas. Merged operators can spread any fixed investment costs over a greater number of subscribers and/or higher levels of usage. Secondly, mergers may enable parties to combine assets to open up investment opportunities. For example, by combining the spectrum holdings of the two merging parties, the combined entity may be able to offer better quality or faster services than its rivals or than either party would have been able to do on its own. Global enablers to spur investment and growth 71

282 THE MOBILE ECONOMY Gauging the impact of mergers on prices In some regions, competition authorities have raised concerns about mergers that reduce the number of mobile network operators in a national market from four to three, generally on the basis that such mergers would lead to higher prices at the retail level and less scope for MVNO deals at the wholesale level. Frontier Economics recently conducted a cross-country review of market performance in the EU. Covering more than a decade, this study found no evidence that unit prices in three player mobile markets are systematically higher than those in four player markets. 52 A different study by New Street Research concluded that both prices and overall returns on capital are actually lower in markets that have consolidated than those that are fragmented, implying that consolidation benefits consumers. The study noted that the high fixed costs involved in running mobile networks means that consolidating companies can achieve major cost savings by eliminating overlapping networks. Returns on capital typically improve as fixed costs are reduced, but then decline again as price competition continues. 4.2 Enabling and encouraging innovation As outlined previously in this report, mobile connectivity has become a powerful platform for innovation. However, this potential for innovation should not be taken for granted. It depends on a flexible, consistent and fair regulatory environment, which encourages experimentation and differentiation. If regulators can help to maintain such an environment, entrepreneurs and enterprises will continue to use mobile technologies and services to deliver many more socio-economic benefits. While cultural and market differences mean there is no one-size-fits all regulatory framework, it is important that regulators establish high-level principles, rather than prescribing detailed requirements that could quickly become obsolete in a fast moving digital economy Global enablers to spur investment and growth

283 THE MOBILE ECONOMY The importance of an open internet The internet in particular, and IP networks in general, provide rich and open platforms for innovation. Policymakers need to safeguard the vibrancy of these platforms by ensuring regulatory frameworks provide appropriate freedom and flexibility for the development and delivery of innovative services, alongside general Internet access, rather than imposing detailed, rigid and constraining provisions. At the same time, regulatory frameworks need to recognise the consumer benefits of traffic management, irrespective of the level of congestion. In order to deliver a high quality best efforts service, mobile operators need to be able to manage their limited resources by optimising video traffic, prioritising voice and video services and using other reasonable and proportionate traffic management methods. In particular, an open internet policy framework should enable mobile operators and other companies to develop and deliver innovative services that require enhanced levels of connectivity, such as telemedicine, e-education, virtual private networks for businesses, IP-TV and telepresence. Policymakers and regulators also need to ensure that operators have the regulatory freedom to support the IoT by offering differentiated services tailored to the needs of new applications and customer segments. For example, a connected ambulance relaying images to a hospital will need a different level of connectivity to an environmental monitor in a field or a smart advertising hoarding Transparency and competition are key If regulators are too prescriptive, they will hold back the development of the digital economy, resulting in a lower quality of service for consumers and businesses, while negatively impacting a wide range of players across the digital value chain. Moreover, rigid rules could limit service providers ability to offer internet access to the widest range of consumers via a variety of different offers, curbing the internet s ability to fuel creativity and entrepreneurship. Healthy competition fuels innovation that meets end users needs. Underpinned by a high level of transparency, competition has a track record of delivering the connectivity consumers and businesses require and expect. Where there are low barriers to switching service providers, a high degree of transparency will enable customers to easily find the right service for them. Policymakers should aim to nurture a vibrant market for Internet access, in which service providers are seeking to differentiate themselves with distinctive services, presenting consumers and business users with many different options. Global enablers to spur investment and growth 73

284 THE MOBILE ECONOMY Avoiding digital lock-in Both consumers and businesses want to be able to access their data, information and applications regardless of their mobile device type or platform. However, when individuals try to switch mobile devices or operating systems, they can face numerous barriers. For example, applications they are using on one device might not be available on another or they cannot transfer their personal profile data or content to a new service provider. These barriers can create lock-in effects that ultimately restrict choice and competition. A potential remedy is to develop an open and non-discriminatory regulatory framework that ensures interoperability and portability without hindering the development of the fast moving digital economy and creating an excessive administrative burden, especially for small companies. Furthermore, regulators could consider conditions of use that are transparent and help to ensure that providers of pivotal online services do not abuse their dominant positions Good internet governance With the Internet Governance Forum s mandate up for renewal in 2015, stakeholders around the world continue to debate the best way to govern the internet, which now plays a fundamental role in the modern global economy. If there are any changes to the current governance model, it is crucial that the internet remains open, safe, secure, stable, trustworthy and interoperable. The best way to safeguard these vital attributes is to continue with the decentralised development of the internet, ensuring it is not controlled by any particular business model or regulatory approach. In particular, policymakers should support the existing multi-stakeholder model, encouraging a collaborative, diverse and inclusive approach to internet governance decision-making. 74 Global enablers to spur investment and growth

285 THE MOBILE ECONOMY Driving digital identification and commerce In the digital economy, reliable authentication of individuals is crucial. Mobile technologies and services can help to securely authenticate people in a wide range of contexts, enabling access to buildings, vehicles and public services, as well as supporting digital transactions. Governments should consider creating a digital identity plan that taps the potential of mobile services to provide a straightforward and secure authentication mechanism. With consumers and businesses increasingly using mobile technologies and services to transfer money and make payments, the mobile industry is now on the radar of financial regulators, as well as telecoms regulators. Ideally, both sets of regulators will coordinate the development of a framework that enables mobile operators and banks to develop and support the ecosystems needed to roll out sophisticated digital commerce propositions. In such a young industry, service providers and policymakers need to work together to understand both the potential and the risks associated with mobile commerce services. At this early stage, regulators need to avoid imposing rigid rules. In the case of mobile money transfer services, for example, interoperability should not be mandatory. Instead, regulators should facilitate dialogue between providers, ensuring that interoperability brings value to the customer, makes commercial sense, is set up at the right time, and regulatory risks are minimised. Global enablers to spur investment and growth 75

286 THE MOBILE ECONOMY Building trust and confidence Real-time data is fast becoming the lifeblood of the digital economy. It is used to inform decisions and ensure the efficient allocation of resources, driving substantial productivity gains in almost every sector of the economy from healthcare and education to manufacturing and agriculture. But for individuals, the growing use of data by organisations and businesses raises legitimate concerns about privacy and whether sensitive information could fall into the wrong hands. Such concerns need to be addressed if the global economy is to create the climate of trust that is required to effectively harness the potential of big data. Policymakers can help guide citizens towards digital services that are secure and safe, while working with the private sector to ensure that children, in particular, can use the internet and other online services without endangering their safety Safeguarding personal data and protecting privacy The widespread take-up and use of mobileenabled digital services depends on a robust and effective framework for the protection of personal data and privacy. Policymakers need to work with the private sector to develop data protection rules that deliver consistent safeguards for citizens, while promoting innovation and ensuring a level playing field for service providers. In this regard, the proposed EU General Data Protection Regulation is an important piece of legislation that could yield significant socioeconomic benefits in Europe. Similarly, in the US, the National Strategy for Trusted Identities in Cyberspace, a White House initiative to work collaboratively with the private sector, advocacy groups, public sector agencies, and other organisations, could also lead to significant improvements in the privacy, security, and convenience of online transactions. Ideally, policymakers should seek to develop a data protection framework that is both future-proof and flexible enough to allow the development of new services, while providing robust protection of personal data and privacy. Given the enormous diversity of applications, a policy framework needs to adopt a risk-based approach that recognises the context in which data is captured and used, and the associated risks, and supports measures to address any such risks. Identified risks can be mitigated by adopting privacyby-design principles, impact assessments, and by the use of anonymisation and pseudonymisation and other accountability measures that consider the user experience and that facilitate understanding and choice. 76 Global enablers to spur investment and growth

287 THE MOBILE ECONOMY 2015 In a similar vein, it is not always appropriate to ask for a consumer s explicit consent to use data. Explicit requests for consent should be reserved for those contexts and categories of data that present the highest risks. An over reliance on an explicit consent regime may create a high burden for users and lead to notice and choice fatigue, numbing people to the importance of making privacy decisions when it matters. As a result, an excessive consent regime may lead to less, not more privacy. As the digital economy cuts across national borders, there is a need for greater international co-operation on data protection. It should be straightforward for companies to transfer data between countries and regions. There have been some important steps in this direction. For example, members of the Association of Southeast Asian Nations (ASEAN) have agreed to develop best practices/guidelines on data protection by This initiative needs to be followed up by a real commitment to harmonise legislation at a regional level. Ideally, ASEAN policymakers will establish cross border standards within the region that create minimum standards for consumer privacy, transparency and choice rather than mandating prescriptive rules that could hinder innovation and economic development. Both international bodies and individual governments need to ensure that there is regulatory consistency and clarity around data protection and privacy. Communications and online service providers, for example, need to be governed by the same clear rules - legislation should be technology-neutral and treat functionally equivalent services and data equally. Contradictory and confusing regulations breed uncertainty and inaction, damaging the development of the digital economy. Global enablers to spur investment and growth 77

288 THE MOBILE ECONOMY Supporting and protecting children Across the world, mobile services are improving the lives of children. With a mobile handset a young person can easily contact parents and friends, access music, videos and games, and find information and knowledge that will help them with their studies. Mobile devices can significantly enhance both formal education and informal learning, enabling students to benefit from interactive lessons, textbooks and multimedia materials. Policymakers should look to encourage mlearning, the use of mobile technologies and services by educational institutions and students both inside and outside the classroom. However, both the public and private sectors need to be mindful that mobile devices, like any tool, can be used in ways that cause harm, so children require guidance and a safe, secure environment to benefit from mobile technologies. The mobile industry has taken steps to protect children online, developing self-regulatory initiatives addressing the need for parental controls, education and awareness. The GSMA also takes part in international initiatives related to child online protection, including the ITU s Child Online Protection programme, and proactively engages with governments and regulators looking to address this issue. In November 2014, the GSMA signed an agreement with Child Helpline International (CHI) that establishes a roadmap for promoting child helplines across the globe, including measures to strengthen relationships between national mobile operators and their in-country helplines and foster collaboration to make the Internet safer for children. The GSMA is also working with the United Nations Children s Fund (UNICEF) to promote industry guidelines to safeguard children online in Latin America and the Caribbean. 78 Global enablers to spur investment and growth

289

290 To download The Mobile Economy 2015 please visit the GSMA website at GSMA Head Office Floor 2 The Walbrook Building 25 Walbrook London EC4N 8AF United Kingdom Tel: +44 (0) Fax: +44 (0)

291 The socio-economic benefits of an L-band IMT identification in Latin America Network savings generated by an identification combined with new subscribers due to falling prices are valued at This additional value can comfortably cover the costs of moving existing L-band users to other bands Mobile data traffic is rapidly rising in Latin America LATIN AMERICAN DATA GROWTH 55TB Mobile broadband services could slow down without additional spectrum LICENSED mobile spectrum in MHz US$6.6B CONSUMER SURPLUS ADDED VALUE IS 20x higher THAN MIGRATION COSTS per month in TB per month in MHz REQUIRED mobile spectrum by Falling mobile broadband prices will expand 4G subscriptions 41.7 million NEW 4G SUBSCRIBERS Strong support for an L-band IMT identification IMT SUPPORTERS IN LATIN AMERICA * Brazil supports IMT despite being one of the heaviest users of the L-band for fixed-links ** Canada also supports IMT The growth in 4G penetration will have a major economic impact on Latin America US$19B ADDED TO GDP 13,000 NEW JOBS An identification at WRC-15 will have major socio-economic benefits which comfortably justify moving existing L-band users 1 Average amount of spectrum licensed to mobile services in Latin America towards the end of Minimum amount of spectrum the ITU predicts will be needed in 2020 (based on MHz range) For more information visit: The full report by BlueNote is available from

292 ASSESSING THE SOCIO-ECONOMIC IMPACT OF IDENTIFYING THE L-BAND FOR IMT SERVICES FINAL REPORT Document prepared by BlueNote Management Consulting for the GSMA - September 15 th

293 TABLE OF CONTENT EXECUTIVE SUMMARY... 5 I. Introduction... 7 II. Background... 8 II.a. Spectrum demand projections... 8 II.b. Initiatives and new frequency bands for IMT III. Services currently provided in the LAC region using L-band IV. Impact assessment of identifying the L-band for IMT IV.a. Conceptual background and methodology IV.b. Assumptions used in modeling IV.c. Key results and findings V. Appendix VI. Bibliography

294 LIST OF FIGURES Figure 1: Value gap between consumer surplus and migration costs... 6 Figure 2: Global mobile data traffic forecast, Figure 3: IMT spectrum dashboard Figure 4: L-band spectrum dashboard Figure 5: Network model structure Figure 6: Total additional consumer surplus due to cost savings and retail price cuts Figure 7: L-band current use Figure 8: Total migration costs per country Figure 9: Network efficiency (IMT systems) and design parameters Figure 10: Key inputs and outputs Figure 11: Information sources for modeling Figure 12: Data traffic consumption profile Figure 13: Subscribers and 4G penetration forecast (2026) Figure 14: Coefficient and multipliers Figure 15: Cost savings to network operators Figure 16: Value gap between consumer surplus and migration costs Figure 17: Additional subscribers obtained from cost savings translated into retails price cut Figure 18: Cost savings plus additional consumer surplus over time Figure 19: Indirect impact from L-band s identification for IMT Figure 20: Band plan options for L-band Figure 21: Services allocation in bands mostly used for IMT in Brazil Figure 22: Services allocation in bands mostly used for IMT in Brazil (Cont.) Figure 23: Brazil s mobile market evolution Figure 24: Brazil mobile market structure Figure 25: L-band s services allocation in Brazil Figure 26: Services allocation in bands mostly used for IMT in Mexico Figure 27: Mobile market structure in Mexico Figure 28: Assignation roadmap for IMT bands in Mexico Figure 29: L-band s services allocation in Mexico Figure 30: Actual use of L-band in Mexico Figure 31: Plans for L-band in Mexico Figure 32: Services allocation in bands mostly used for IMT in Colombia Figure 33: Services allocation in bands mostly used for IMT in Colombia (Cont.).. 57 Figure 34: IMT spectrum holdings in Colombia Figure 35: Mobile market evolution of Colombia Figure 36: Main operators of the Colombian mobile market Figure 37: L-band s services allocation in Colombia Figure 38: Timetable for L-band s licenses expiration dates in Colombia Figure 39: Compatibility L-band current services and IMT Figure 40: Services allocation in bands mostly used for IMT in Ecuador Figure 41: Services allocation in bands mostly used for IMT in Ecuador (Cont.)

295 Figure 42: Mobile market structure and spectrum holding in Ecuador Figure 43: L-band s services allocation in Ecuador Figure 44: L-band current usage in Ecuador Figure 45: Services allocation in bands mostly used for IMT in Uruguay Figure 46: IMT spectrum holdings in Uruguay Figure 47: Mobile market evolution of Uruguay Figure 48: L-band s services allocation Figure 49: Frequency allocation in bands mostly used for IMT in Argentina Figure 50: IMT spectrum holdings in Argentina Figure 51: Mobile market evolution and structure in Argentina Figure 52: L-band s services allocation in Argentina Figure 53: Bands identified for IMT in the Dominican Republic Figure 54: Evolution and structure of the mobile market in the Dom. Republic Figure 55: L-band s services allocation in the Dominican Republic

296 EXECUTIVE SUMMARY An important ITU-R Resolution approved at the 2012 World Radiocommunication Conference (WRC- 12) invited members to study additional spectrum requirements for IMT, and to identify potential candidate frequency bands. Members were asked to take into account the protection of existing services and the need for harmonization. The GSMA considers the L-band ( MHz/ MHz) as a potential candidate for IMT, and has commissioned specialist spectrum consulting firm BlueNote Management Consulting (BNMC) to prepare this report. This report estimates the socio-economic impact in Latin America and the Caribbean (LAC) of using the L-band to provide mobile broadband services. The research first investigates the current usage of the band, then briefly reviews its technical suitability, and goes on to provide quantitative modeling of the socio-economic impacts. A sample of seven countries was selected for in-depth analysis, and the results for the whole region were then extrapolated. In line with other GSMA commissioned studies, the underlying methodology focuses on the impact on consumers. The report finds that identifying the L-band for IMT would create cost savings for telecommunications operators, who would enjoy lower capital and operational expenditures when deploying access networks. Given competitive market conditions these savings should be passed on to end users, keeping the sector returns relatively stable. The study estimates the socio-economic impact of identifying the L-band for IMT services in each of the seven sample countries by comparing two scenarios -- having the L-band available for IMT, and not having it available. The difference between these scenarios represents the incremental value, or cost saving, of having the L-band available for IMT. The first scenario includes the migration costs of current users as a capital expenditure for clearing the band. Having estimated the cost savings per country, the analysis assumed that competition between operators would assure those cost savings would ultimately be translated into retail price cuts. The model also assumes dynamic behavior of market demand, with additional 4G users being attracted by the lower prices. The combination of both impacts, that is additional consumers plus cost savings, represents the total consumer surplus. 5

297 As of 2017, the model estimates the combined consumer surplus at US$4.5 billion in the seven sample countries and US$6.6 billion for the whole LAC region 1. In terms of investment payback, the consumer surplus accounts for almost 20 times the total migration costs. Having the L-band available for IMT will not only benefit the telecommunications sector, but it will also generate knock-on effects in other sectors. It will boost aggregate demand, improve economic productivity, and fuel the labor market. As of 2026, the model estimates 41.7 million new 4G users in the LAC region, resulting in US$19.8 billion of incremental GDP 2 and generating more than new jobs in the LAC economy. The quantitative analysis concludes that identifying the L-band for IMT will create material benefits for the telecommunications sector, individual governments and society as a whole. The evidence of the funding being available to create those benefits can be seen in the large gap between the substantial cost savings and the comparatively very small migration costs if the band were made available (the figure below graphically illustrates the gap). Figure 1: Value gap between consumer surplus and migration costs Million US$ at 2017, NPV, Cash flows discounted at social discount rate Consumer Surplus 2.403,8 ~ 20x Migration Costs Migration cost (2017) Consumer surplus (2017) 959,7 0,9 458,1 164,6 394,0 0,9 6,0 81,7 39,6105,2 23,5 1,1 51,2 Argentina Brazil Colombia Dominican Republic Ecuador Mexico Uruguay The cost of migrating the current L-band users to another band is not significant when compared to the consumer surplus generated The outcome of identifying the L-band for IMT is positive in each sample country 1 Estimates based on a 10-years period and using social discount rates 2 Including direct and indirect impacts 6

298 I. Introduction The development of new radio-access technologies for telecommunications, coupled with the wide adoption of mobile services by consumers, has boosted spectrum requirements worldwide. Latin America and the Caribbean, in line with the rest of the world, needs to pave the way for a timely allocation of new spectrum to ensure the speedy development of advanced mobile services in the region. The GSMA considers the low-uhf band ( MHz), L-band ( / MHz), the GHz band and C-band ( GHz), as potential candidates for enlarging the current IMT group to ensure sufficient spectrum is available for the projected traffic growth. In this context, the GSMA has commissioned BlueNote Management Consulting (BNMC) to estimate the socio-economic impact of identifying the L-band for mobile data services in Latin America and the Caribbean (LAC). In order to do this it has first been necessary to understand the potential efficiencies derived from using the L-band for mobile services versus alternative usages. This report is structured as follows: - the second chapter looks at the current situation, presenting the global spectrum demand and supply and ITU, and other stakeholders, initiatives; - the third chapter describes the main findings of the research in the LAC region; - the fourth chapter presents the results of the quantitative modeling, together with the conceptual background and key assumptions; - the fifth and the sixth chapters comprise the appendix and the bibliography. 7

299 II. Background II.a. Spectrum demand projections A few years ago consumers used their mobile devices only for voice and text communications, but in recent times the situation has dramatically changed. The introduction of smartphones, notebooks and tablets, alongside the development of innovative mobile applications, has resulted in an exponential growth of data demand. The strong demand for data is expected to continue to be a growing trend for the foreseeable future. In 2013 global mobile data traffic grew by 81% over 2012, whilst in Latin America the figure was 105%, according to statistics compiled by infrastructure vendor Cisco 3. Although smartphones, tablets and other mobile devices have different usage profiles, video streaming is the most important dataconsumption activity. Videos account for 50% of global mobile data traffic and are expected to account for 66% by Global data traffic is expected to grow at 61% annually from 2013 to 2018, which represents an overall ten-fold growth, climbing to almost 16 exabytes per month by 2018 (see the chart below). Figure 2: Global mobile data traffic forecast, Global mobile data traffic Devices Exabytes per month % 15,9 Smart devices Non-smart devices 7 10,8 79% 74% 68% 61% 54% 46% 1,5 2,6 4,4 21% 26% 32% 39% 46% 54% Global mobile data traffic will reach 15.9EB/month Global mobile data traffic CAGR ( ): 61% Smart devices as key drivers for supporting mobile data growth Source: Cisco VNI Mobile, Cisco,

300 LTE network deployment is accelerating the growth of data consumption as 4G users typically consume six times more 4 data than 3G user s. As well as 4G network deployment, rapid growth in smartphone sales is another key driver explaining the boost in mobile data traffic. The Report ITU-R M provides a future spectrum requirements estimate for terrestrial IMT, with projections on the spectrum requirement per country by The required spectrum is estimated for both lower-user-density markets and higher-user-density markets, based both on data consumption levels and market development. The table below shows the ITU s spectrum estimates for Density Lower-user-density settings Higher-user-density settings Total spectrum requirements for pre- IMT, IMT-2000 Total spectrum requirements for IMT-Advanced Total spectrum requirements 440MHz 900MHz 1340MHz 540MHz 1420MHz 1960MHz The GSMA commissioned Coleago Consulting to develop a model to estimate future IMT spectrum requirements based on existing networks for the year Their results for five Latin American countries are in the table below. Country Low Usage High Usage Argentina 1093MHz 1628MHz Brazil 1129MHz 1676MHz Chile 893MHz 1327MHz Colombia 1057MHz 1578MHz Mexico 977MHz 1545MHz Based on mobile data consumption projections and international estimates of spectrum requirements, an urgent need to identify additional spectrum for the provision of IMT is obvious. Any agenda for the identification procedure should factor in the significant time lags resulting both from often-lengthy assignation processes and network roll out. 4 CISCO VNI

301 II.b. Initiatives and new frequency bands for IMT II.b.i. ITU initiatives Because of spectrum demand growth and the demonstrable positive contribution of mobile broadband services to society and the economy, the ITU is planning ahead. At WRC-12, a resolution was adopted (ITU Resolution 807) to define the agenda for WRC-15. Agenda item 1.1 raises the issue of additional spectrum allocations to the mobile service on a primary basis and identification of additional frequency bands for International Mobile Telecommunications (IMT) and related regulatory provisions, to facilitate the development of terrestrial mobile broadband applications, in accordance with Resolution 233 (WRC-12) 5. Resolution 233 (WRC-12) invites ITU-R members to study additional spectrum requirements and potential candidate frequency bands, taking into account the protection of existing services and the need for harmonization. It also stipulates that appropriate and timely availability of spectrum will be critical to support mobile broadband system growth (i.e. IMT development). In response to this invitation, spectrum administrations, international telecoms associations, service providers and operators from around the world have submitted contributions to ITU groups WP 5D and JTG They have recommended candidate frequencies to be used for IMT deployment, and prepared studies on the compatibility between IMT and services now operating in the proposed bands. As administrations may deploy IMT systems only in specific portions of the band, it is useful to look at which frequency bands are already identified for IMT. These are listed in the following table. Band (MHz) Footnotes of RR identifying the band for IMT AA A, 5.317A A, A A A, 5.432A, 5.432B, 5.433A 5 ITU,

302 Frequency bands identified for IMT 6 The bands which have been identified for possible future mobile broadband use by ITU-R members in response to Agenda Item 1.1 fall into the ranges which are shown below, along with their main attributes: - lower bands for coverage: below 698MHz and above 450MHz - central bands for capacity and coverage: above 960MHz and below 1710MHz - upper bands for capacity: around 2200MHz and 2500MHz - very high frequency bands for performance: above 3600MHz. The coverage attributes of the sub-1ghz bands are crucial in developing countries, as they can provide cost-efficient broadband services in rural and low-population-density areas. The capacity offered by higher-frequency bands addresses the high-traffic demand of new services and denselypopulated urban areas. Finally, the performance attribute of very high frequency bands is ideal for very-high-density hotspots and indoor areas where substantial bandwidth is needed to support high throughput. The L-band falls in the middle and is suitable both for coverage and capacity. However in Latin America coverage is usually provided using sub-1ghz spectrum, because of its widespread geographical areas with low population densities in small rural communities. The L-band has been the subject of often-spirited debates in international forums with existing users of the band or adjacent bands (most of them providers of passive services) demanding protection from potentially-harmful interference. This is what necessitates specific studies on compatibility between the IMT systems and the other systems in the band. II.b.ii. Other initiatives and recommendations In the context of the WP 5D and JTG meetings, some regional organizations, such as the European Conference of Postal and Telecommunications Administrations (CEPT), member states, technology providers and vendors, have sent contributions regarding L-band suitability for IMT services. This section comprises a summary of the recommendation from CEPT European Communications Committee (ECC) Report ITU, 2013 Recommendation ITU-R M about Radio Regulations (RR) edition

303 In 2010, the CEPT instructed its ECC to review possible future uses of L-band to investigate the possibility of refarming 40MHz of spectrum for new services which might provide more social and economic benefits to Europe. In accordance with the ITU s Radio Regulations, much of the upper part of L-band from MHz is allocated to fixed, mobile (except aeronautical mobile), broadcasting and broadcasting satellite services. The CEPT decided to investigate the feasibility of a range of new services in the band including terrestrial broadcasting, mobile broadband, mobile supplemental downlink, satellite digital audio broadcasting, program making and special events (PMSE), broadband public protection and disaster relief (PPDR), and broadband direct air to ground communication. The criteria for evaluating each service were 7 : - Compatibility with the current regulatory framework - Possibility of sharing the band with other applications/uses - Maximisation of social and economic benefits - Roadmap of equipment (on a large scale) and of applications deployment - Potential economies of scale Having completed its analysis, the ECC proposed the sub-band MHz be set aside for Mobile/Fixed Communications Network (MFCN) supplemental downlink (SDL). The ECC report defines the technical conditions for band harmonization based on 8 blocks of 5MHz each, and proposes the elimination of designated sub-band MHz for Satellite Digital Audio Broadcasting 8. In February 2013, Qualcomm, Ericsson and Orange successfully completed the world s first live demonstration of SDL in L-Band 9 in MHz in combination with the 2.1GHz band. Following the ECC report, 3GPP included the SDL feature in LTE and HSPA+ specifications. L-Band for SDL in E-UTRA and UTRA was included in 3GPP Release 12 and the specification was completed in June Although L-Band is widely associated with the 1GHz to 2GHz spectrum range, which is known worldwide as the lower band for satellite services 11, only a portion of this frequency range is actually used for this type of service. In fact, the L-Band has a variety of other current uses including telemetry, 7 CEPT, idem 9 ERICSSON, GPP, Marine Satellite Systems,

304 radar, digital audio broadcasting (terrestrial and satellite), fixed links (point to point and point to multipoint) and radioastronomy. In Region 2 the MHz part of L-band is already allocated to mobile services. In Region 3 there are LTE commercial networks operating in part of the L-Band in line with 3GPP specifications 12, including the two Japanese operators NTT DOCOMO and Softbank. 12 Band 11 and Band 21 in 3GPP Specifications 13

305 III. Services currently provided in the LAC region using L-band To evaluate a new identification of the L-band, it is necessary to consider and thoroughly understand the current usage of L-band in the LAC region. This includes current-user profiles, the feasibility of migrating them to other bands, other IMT allocations and the public agenda, if any, regarding the L- band. BlueNote distributed information requests and held interviews with the local spectrum administrations in the seven LAC countries being sampled for this survey in order to depict current usage and public plans. Following this in-depth country-by-country analysis, results for the entire LAC region were then extrapolated. The survey was split into two areas of investigation, the first of which was to understand current IMT availability and the short-term roadmap for new assignments and the second was to develop an understanding of current uses of the L-band. The following table summarizes the results of the first part of the investigation which was understanding the current availability and roadmap for IMT assignments in the sample countries. Figure 3: IMT spectrum dashboard 2014 Brazil Total current identification for IMT (MHz) 660 IMT bands already assigned 450MHz 850MHz / 900MHz 1800MHz / 1900MHz 2100MHz 2500MHz Current assignations (MHz) 529 ~ 50,000 # BTS Assignation roadmap 700MHz band (80MHz, auction to take place in 4Q 2014) Mexico MHz / 1900MHz AWS 230 ~31, MHz band (90MHz by end 2014) AWS band (30MHz by end 2015) 2.5GHz band (60MHz by end 2015) Colombia MHz / 1900MHz AWS / 2,5GHz 415 ~ 20, MHz (20MHz) and 1.9GHz (5MHz) both in early MHz (total or a portion of the band) 2.5GHz (30MHz) Ecuador MHz / 1900MHz AWS / 700MHz 180 ~ 9,000 Late 2014 or early 2015: 700MHz (portion) and AWS (50MHz) 2.5GHz (190MHz) 1900MHz (60MHz) Uruguay MHz / 1900MHz 1800MHz / 2100MHz AWS / 900MHz 270 ~2, MHz (90MHz by 2016) Argentina MHz / 1900MHz 136 ~ 16, MHz and AWS (auction has already been announced for 4Q 2014) 1900MHz (remaining 20MHz, by 2014) Dom. Rep MHz / 900MHz / 1900MHz AWS / 2,5GHz 450 N / A No defined agenda Note: Figures from Dominican Republic exclude AWS extension ( MHz and MHz) Current assignments estimated on national basis Sources: Local administration survey and web research 14

306 As can be seen above, Brazil is the country that has assigned the most spectrum for the provision of IMT. The bands already assigned are 450MHz, 850MHz, 900MHz, 1800MHz, 1900MHz, 2.1GHz and 2.5GHz with the digital dividend 700MHz band scheduled for auction later in the year. The variations in 4G spectrum assignments is a distinctive aspect of the Latin American market, although Argentina is the only country yet to assign any spectrum for 4G 13. All the other countries have already assigned spectrum in the AWS band. Colombia, Brazil the Dominican Republic and Ecuador have also assigned the 2.5GHz while the latter has also already assigned the digital dividend band. As for the assignation roadmap, most countries have short-term plans for assigning additional IMT spectrum 14. However, although these new assignments will improve the region s IMT spectrum availability, it will not yet match the ITU s spectrum requirements mentioned in the ITU-R Report M Our specific research on the L-band has been structured into three main categories: the current usage profile, the feasibility of migrating current users to another band (including both the operational and the legal feasibility) and the administration position and plans regarding the band. As can be seen in the table below, the most common use for L-band spectrum is point-to-point fixed links. Although Brazil and Ecuador have significant numbers of users, there are few users in the other countries sampled. The links belong to different industries (although telecommunications is the most prevalent) and are widely-spread geographically with a significant number of users in rural areas. In Brazil, where most available licenses have been assigned, the current users aside from telecommunications companies -- include private companies in a variety of other industries, such as utilities and energy. 13 Though it was announced an spectrum auction for late October Some countries have already formally announced their 4G spectrum assignation process (e.g., Brazil, Argentina) whilst in other countries the formal announcement is still pending (e.g., Colombia, Ecuador). 15 According to this report, by 2020, there should be a total of 1,340MHz and 1,960MHz identified to IMT in low-density and high-density markets, respectively. 15

307 Figure 4: L-band spectrum dashboard 2014 Latin America: CurrentL use Band / users dashboard Migration operational feasibility Migration enforcement feasibility L-band Links Governments position regarding L-band Brazil Point-to-point and point-tomultipoint links Large number of licences and users (telcos, oil&gas, utilities) Relatively complex, due to the high number of users Existing licenses last for 10 years (no automatic renewal) ~ 2,800 To be proposed for IMT, in combination with other bands Mexico Colombia Ecuador Uruguay Fixed service microwave pointto-point systems Mainly private companies, few state entities Main use: point-to-point networks (telecommunication support) Few number of users Radio links of fixed services Mainly wholesale carriers Low level of use (no more than 10 users) Users are private and government entities Relatively simple, as the number of users is limited A priori, current users would agree to migrate to other bands Relatively complex, due to the high number of users Simple, due to the reduced number of users Most users have obsolete equipment Assignments last for 15 years Not possible to migrate users from a portion of the band Simple, almost 100% of the licenses will expire by 2020 Assignments last for 5 years, thereafter they should be renewed No expiration date Awarded on a revocable basis, having the government the capacity to modify these licenses ~367 (1427 to 1525 MHz) ~26 point-topoint ~67 point-tomultipoint (DECT) Potentially in agenda To be proposed for IMT, in combination with other bands ~ 650 Potentially in agenda ~ 10 To be defined Argentina Point-to-point links Digital multichannel systems (MXD) are authorized to operate N/A Rule: no expiration date However, 10 years period for telecommunications support spectrum N /A To be defined Dominican Republic Aeronautic radionavigation and point-to-point links 12 users: public services, religious organizations and government offices No near plans to change the allocation of this band nor to migrate its users ~ 100 (12 users) No near plans as for this band Sources: Local administration survey and web research Incumbent users of the L-band will probably be reluctant to move to other bands unless they are compensated as they have already invested in frequency-specific equipment. Faced with this opposition, administrations could either wait for the expiration of current licenses, and then migrate the users, or enforce migration prior to the license expiry dates, although this is only possibly in a handful of regulatory contexts. Some administrations might also decide to force the migration of the current users based on spectrum efficiency fundamentals in Uruguay, Argentina and Ecuador licenses assure spectrum availability to their holders, but they do not assure a specific frequency. None of these adversarial measures is likely however as the impact analysis shows the cost saving of using the L-band for IMT amounts to almost 20 times the migration costs of users in the entire LAC region. This means migration costs could be easily be absorbed by the reallocation beneficiaries. There would be no significant trade-off between current users and new IMT users, since both would have the spectrum they need without incurring in a net loss. As for feasibility of migration out of the band, no major drawbacks have been identified aside from operational issues in the two countries with a large numbers of current users. And in some cases, 16

308 Uruguay being an example, most of the equipment is already obsolete (and current users might have already updated their plans). Regarding the public agenda and plans for the identification of L-band for IMT services, most countries are in favour. Brazil, Colombia, Mexico and Ecuador will presumably support such identification, whilst Argentina and Uruguay have not yet adopted a definitive position and the topic is still under analysis. The Dominican Republic is the only country not assessing the possibility so far. Having completed the research in the seven sample countries the reports concludes there is no material barrier to identifying L-band for IMT. Indeed most of the countries are positive towards the relevant ITU-R reports and recommendations, are already assessing this initiative and are bullish about the possibility of having the L-band available in the not-too-distant future. 17

309 IV. Impact assessment of identifying the L-band for IMT This section comprises the estimation methodology of the economic impact of having the L-band available for mobile services, the key assumptions for modelling this impact and the final results and findings. IV.a. Conceptual background and methodology Having the L-band available for mobile services results in benefits for mobiles operators, for the market and, ultimately, for consumers. If the band is made available, however, then incumbent users must migrate to other frequencies, and so incur migration costs. The cornerstone of the underlying approach is to compare the total benefits of reallocation against its cost, which is basically the migration cost. Our calculation of total benefits involves both the cost savings for mobile operators and the consumer surplus increment for end users. The study first estimates the socio-economic impact of identifying the L-band for IMT in each of the seven sample countries. The model compares two scenarios for each market, a first scenario where the L-band is made available for wireless broadband, and a second scenario where it is not. The difference between these scenarios is the incremental value, or cost saving, of identifying the L-band for IMT. The first scenario includes the migration costs of current users as a capital expenditure for clearing the band. Having estimated the cost savings per country, the analysis assumed competition between operators would assure the savings would ultimately be translated into retail price cuts. The model also assumes dynamic behavior of market demand, with additional 4G users being attracted by the lower prices. The combination of both impacts, that is additional consumers plus cost savings, represents the total consumer surplus. In the following paragraphs it is explained the methodology used to estimate firstly the cost savings, secondly, the consumer surplus, and thirdly, the indirect impacts. Cost saving estimates As has already been noted, the demand for mobile services is increasing rapidly and, to meet this growth, mobile operators need to expand their access network capabilities. Mobile operators can do 18

310 this either by the deploying additional infrastructure in the form of new base station or by utilizing more spectrum. Many will need to do both. The availability of new IMT bands allows mobile operators to reduce their total capital and operational expenditures on radio stations or sites. This is because spectrum and radio access infrastructure are, to some extent, substitutes 16 for each other in the delivery of mobile services. The model calculates the cost savings for mobile operators who have more spectrum by looking at the difference in impact on their total capital and operating expenditures between a scenario having the L-band available versus a scenario of not having it. For these estimates a network model based was developed, based on the following structure: Figure 5: Network model structure # Sites Total Costs of Network Cost savings of L-band Data traffic per subscriber Total data traffic demand x / vs. # Traffic per site (capacity) With L-band Capex + opex per site - # subscribers for data services LTE spectrum efficiency total spectrum availability Without L-band site capex backaul capex capex per site # Sites Total Costs of Network Total LTE data traffic demand is one of the key inputs for the model. Once the total traffic projections were developed, the total traffic per site was estimated. The per-site traffic depends on spectrum efficiency (the more efficient, the more traffic per site) and the spectrum availability (the more spectrum is available, the more traffic per site). The assumptions and considerations taken into account when calculating total traffic per site are described below, in the key assumptions section. The total traffic per site estimate enables a calculation of the number of sites an operator requires to satisfy the traffic demand with a given amount of spectrum. 16 However, this substitution is far from perfect, due to several barriers: timing for setting up new sites, legal authorization constraints for urban sites and specific services demanding high traffic capacity (i.e., streaming). 19

311 For a mobile operator, the benefits of having more spectrum are associated with an increase in capacity and in coverage. However, these benefits depend on the specific frequency: the benefits from low-frequency, sub-1ghz bands include both coverage and a certain amount of capacity, but the benefits of high frequency bands are more associated with capacity due to their limited propagation characteristics. This is certainly the case in Latin America, where coverage is usually provided by sub-1ghz spectrum, providing wireless broadband to low-density communities spread across large areas. So for LAC a conservative scenario was assumed where the L-Band would be used to boost network capacity. Qualcomm has proposed the L-Band be employed to increase downlink capacity through supplemental downlink (SDL). The total number of sites required depends on the amount of spectrum available for use, so having more spectrum made available through reallocation of the L-band to mobile services will result in a lower number of sites being needed until traffic has built up again. The gap in the number of sites needed between having use of the L-band and not having use of it points to the difference in total network costs, and, ultimately, the cost savings of having the L-band. Consumer surplus In a competitive market environment, the cost savings of having the L-band available for mobile services should result in benefits for consumers. The underlying assumption here is that, when competition is a factor, mobile operators do not enjoy extraordinary profits and are forced to pass on the benefits of lower expenditures to their customers in the form of price cuts so as not to lose market share. In line with previous GSMA studies 17, the consumer surplus concept was used for the impact assessment methodology. Economic literature defines consumer surplus 18 as the difference between what a consumer is ready to pay for a service ( willingness to pay ) and what he actually pays for it (the clear market price). The figure below describes this concept. 17 GSMA (2013) et al. 18 Varian Hal R. (1992) 20

312 Figure 6: Total additional consumer surplus due to cost savings and retail price cuts E q : New market price E q : Current market price + = Total additional consumer surplus Price of service Consumer surplus before price cut Additional consumer surplus derived from new consumers Cost savings per unit C u C u E q E q Additional consumer surplus derived from existent consumers Demand Extra subs. Quantity of subscribers As shown in the illustration above, the price a specific consumer is willing to pay (represented by the dotted portion of the demand curve in the figure) exceeds the clearing market price ( E ). The amount of this difference is consumer surplus. Each consumer will enjoy different levels of consumer surplus. In the above illustration, the most voracious consumer enjoys the highest consumer surplus, whilst the marginal consumer enjoys a consumer surplus equal to zero. To obtain the total consumer surplus of the whole market, all consumer surplus of each consumer is aggregated (the grey shadowed triangle). The cost savings derived from having more spectrum available will reduce the cost to operators, and, under competitive conditions, these reductions might be expected to be passed on to consumers via price reductions. Assuming that once the L-band is allocated to IMT services, the operators are able to cut prices, and so market prices fall from E to E (as in the figure above). The total consumer surplus will be higher than the consumer surplus prior to the price cut. The additional consumer surplus comprises two components: the net increase in the consumer surplus of each existing client (the green rectangle in the figure above) plus the consumer surplus of new subscribers (the red triangle in the figure above). 21

313 To estimate the additional demand, which comes after cost savings are translated into price cuts, consideration is given to demand price elasticity (see next section for this assumption) to estimate how many additional consumers will enter the market as the prices drop. To calculate the final retail price cuts assumption are made that the same value of cost savings to the operator is translated into price reductions for the consumer. Finally an assumption is made of a flat supply curve at the margin, or constant productivity at the margin. That means that operators would have to pay the same cost per additional subscriber as they had to pay for their existing subscribers. This may be viewed as a valid simplification as the total number of new subscribers represents a limited portion of the total subscribers (roughly 2%). To arrive at the net socio-economic impact of allocating the L-band for IMT, the net impact to the end user is compared with the migration costs to other frequencies of the current users. In the next section are the detailed assumptions used to arrive at the migration costs estimates. Indirect impacts Having the L-band available for IMT services will not only benefit the telecommunications sector and the overall economy, but it will also generate indirect impacts on other sectors to boost economic efficiency, the GDP and new employment opportunities in the labour market. The additional spectrum from L-band availability means more mobile broadband subscriptions which will have knock-on effects on domestic demand, on spending in other sectors and on the economic productivity in general. The indirect impacts of reallocating the L-band into can be sub-divided into three categories: - GDP impact - Job creation impact - Fiscal impact The allocation of the L-band for mobile services creates economic value as it implies a better use of scarce resources, originating cost savings for the telecoms sector hence boosting the domestic economy and the total GDP. GDP would also experience additional impacts from this reallocation because higher broadband penetration rates improve the efficiency of the economy. There are several ITU studies 19 which 19 ITU

314 demonstrate the positive impact broadband penetration has on productivity, economic efficiency and, ultimately, on GDP. Identification of the L-band for IMT is expected to have a positive impact on each country s job creation capacity. Job creation happens when broadband network construction affects employment by creating direct jobs in facility construction as well as jobs in telecom equipment industries. The increase in household spending related to increased broadband activity also creates employment in the wider economy. It should be noted no specific analysis of the impact on GDP and job creation as a result of identification of the L-band for IMT has been conducted. Instead reliance is made on the findings of existing studies 20 to set the parameters and make estimates. Fiscal impact measures the additional taxes Governments are expected to be able to collect as a consequence of making the L-band available for wireless broadband. The identification of the L-band for IMT will, as previously noted, result in a price reduction which, considering the price elasticity, will end up in a higher number of subscribers and, consequently, in higher tax collections. Scaling the results up for the region Although the scope of the study is the whole LAC region, in-depth analysis has only been conducted within the stated seven countries. The results of the country-by-country analysis are used to scale up to provide a regional viewpoint and present representatives findings for the entire region. To extrapolate the results from the data gathered in the sample countries, a multiple-variable regression analysis was used, including the explanatory variables shown in the next table. Explanatory variable Fundamental 4G subscribers The spectrum value and the impact of having more spectrum are positively linked to the traffic market demand (the more the merrier), and the 4G subscribers are the ones with a higher data consumption profile Markets size, in terms of total population 20 ITU 2012 and Telecom Advisory Services

315 Estimates of consumers surplus and additional impacts included in this study depend on the market size, measured in terms of the country population ARPU The Average Revenue per User has a direct impact on the consumer surplus estimation based on the additional subscribers These three explanatory variables were tested to estimate the socio-economic impacts for the entire LAC region. The impacts to estimate are: - cost savings - additional subscribers - efficiency impact and GDP growth - job creation - fiscal impact Assessment of different combinations of explanatory variables by evaluating their level of significance and explanatory power enabled elimination of the ARPU variable due to its lack of statistical significance. The final version of the scale-up model consists of a multivariable regression based on the total population and 4G subscribers variables 21. IV.b. Assumptions used in modeling In this section the key assumptions underlying the modelling exercise as well as the relevant parameters for the impact assessment are presented. The modelling key assumptions can be structured into three categories: migration costs, technology and financial and market variables. IV.b.i. Other services operating in the band and migration costs As previously noted, in Latin America the L-band is mostly used for fixed services (point-to-point and point-to-multipoint), with Brazil having the largest number of licenses in the band (6,000 licenses and close to 2800 radio links). 21 The ARPU variable shows statistical significance for the GDP estimation only. 24

316 There are a negligible number of radars operating in the sub-band MHz and administrations in the sample countries do not have information about radars operating in this band on record. Current radar users of the band have not been taken into account. Also ignored are migration costs associated with radiolocation services because they are capable of operating below the band floor of 1350MHz. The next figure shows the current use of L-Band by service allocated in the sample countries. Figure 7: L-band current use Radiolocation ( MHz) Fixed Services ( MHz) Broadcasting (Satellite and Terrestrial) Migration Feasibility Brazil No Reports ~6000 licenses ~2800 links N.a. OK Mexico No Reports 367 links N.a. OK Colombia No Reports 26 links point-to-point 67 licenses point-tomultipoint N.a. OK Ecuador No Reports 650 links 1 License (200 subscriber) OK Uruguay No Reports 10 links N.a. OK Argentina No Reports N.a. N.a. OK Dominican Republic No Reports 100 links N.a. OK Sources: Local administration survey In the seven countries sampled there are 3,953 radio links and they operate in the sub-band at MHz. Many of the existing licenses expire by 2017 and the services could be migrated to other frequency bands. For those countries with limited or non-existent information on the expiration dates, for modelling purposes a 30% of the radio links was assumed to turn off by It is technically feasible to migrate current users to other frequency bands allocated for fixed services, such as: - Lower bands below 6GHz: namely 2GHz, 4GHz, 5GHz and 6GHz % of radio links could be turned off by 2017 either because the license expires or because it is possible to remove. 25

317 Millares - Higher bands above 6GHz: namely 7GHz, 11GHz, 15GHz, and 18GHz. For modelling purposes, an assumption of 95% can be migrated to lower bands, while the remaining links would have to be migrated to higher bands. For the latter, new repeater stations are required to guarantee the quality of operation within the same distance between transmitter and receptor. The model includes three main items for cost estimation: - engineering services (including survey, zoning, new equipment installation or retuning of current equipment) estimated for each country radio link equipment: US$64,000 per link (New equipment for transmission and reception stations) 24 - infrastructure for a new repeater station: US$200,000 per site 25 On this basis, and considering the current price level of equipment and services in the market, the next figure shows an estimate of the total migration cost per country. Figure 8: Total migration costs per country 2017, US$ million, point-to-point radio links 39,6 23,5 6,0 1,1 0,9 0,9 164,6 236,6 Brazil Ecuador Mexico Dominican Uruguay Colombia Argentina Total Republic Source: FCC (2010), U.S. Bureau of Labor Statistics, World Bank, IMF, radio link vendors benchmarking, ANATEL spectrum database, ANE Colombia, IFT, INDOTEL, SENATEL, CNC, BNMC estimates IV.b.ii. Network design parameters The core benefit of 4G technologies is the improvement in spectral efficiency, meaning more capacity in the same bandwidth. Spectral efficiency parameters are determinant for network design and sizing, 23 FCC, A Broadband Network Cost Model, and BNMC analysis (benchmarking of engineering companies). The cost was adjusted to each country using PPP (Purchasing Power Parity). 24 FCC A Broadband Network Cost Model and BNMC estimates. 25 Benchmarking COFETEL, OfCom, ACMA, OPTA and BNMC analysis. 26

318 as they allow an accurate estimate of the number of base stations required to support a given traffic demand. In addition to spectral efficiency, there are other design parameters which have a dramatic impact on the number of base stations required. These include the traffic distribution factor, network load factor for an appropriate performance, traffic concentration in peak hours, number of cells per site and the traffic distribution between uplink (UL) and downlink (DL). The next figure shows the spectral efficiency underlying the modelling exercise, in line with the 3GPP TR document and the assumptions mentioned above. Figure 9: Network efficiency (IMT systems) and design parameters bps/hz/cell Downlink Uplink 1,7 1,4 2,2 0,33 0,5 0,74 UMTS/HSPA LTE (Rel 8) LTE Adv Additional Parameters Distribution Factor: 15% of sites carry 50% of traffic (15%/50% = 0.3) Load Factor: It s inconvenient to operate a LTE network with 100% of load (SINR decreases in a high load scenario), the design recommendation is to consider a factor of 50% - 70% in DL and 50% in UL. Relation UL/DL: 70% of total traffic is flowing in DL Cells per site: 3 cells per site is a common factor in mobile networks in Latam Traffic in BH/Daily traffic: 7% Source: 3GPP TR / ITU-T M.2135 / NSN - "Mobile broadband with HSPA and LTE: Capacity and Cost aspects A band plan of 2x35MHz is assumed (Option A, as described in the Annex), due to its higher probability of equipment availability, as it is already commercially launched. Finally, below is the formula to estimate the number of base stations (BTS) required to support the traffic forecast and the design parameters mentioned above. #BTS = Total DL Data Traffic in Busy Hour (Mbps) bps Spectral Effic.in DL ( Hz ) Spectrum bandwidth in DL(Hz) cells Load Factor Distribution Factor Sector site 27

319 Where the total DL Data Traffic is calculated using the next formula Users Profile Users ( GB ) 7%(Busy Hour) 70%(Dl Traffic) Total DL Data Traffic BH = Month 30 ( days ) 3600Sec 8192(Mbits month GB ) IV.b.iii. Capex, opex and market parameters A bottom-up model was developed based on cash-flow projections for a 10-year period (from 2017 until 2026) to estimate the impact of reallocating the L-band to IMT services. The model outcome is fully dependent on the methodology previously detailed and on the assumptions and inputs adopted. The key inputs for the model are those related to the BTS sizing (e.g., subscribers, traffic), the technical inputs, the operational and capital expenditure amounts and the macroeconomic projections. The following chart lists the key inputs feeding the model: Figure 10: Key inputs and outputs Inputs Outputs Subscriptions 4G penetration Subscriber data consumption profile ARPU / CAPEX / OPEX Discount rate Elasticity BTS dimensioning technical parameters... Economic model Cost Saving Consumer surplus Additional subscribers Fiscal Impact GDP Impact Job creation impact The model has six main outcomes: cost saving and consumer surplus, additional subscriber impacts, fiscal impact, GDP indirect impact and job creation impact. The following figure summarizes the main information sources that have been considered for the inputs in the model build-up. 28

320 Figure 11: Information sources for modeling Module Variables Sources BTS sizing Subscriptions Traffic 4G penetration Technical specifications OVUM 3GPP Informa GSMA Intelligence Ericsson traffic exploration Cisco VNI Opex Urban and non-urban BTS cost Urban and non-urban infrastructure BTS sharing Deloitte Informa Ministerio de las tecnologías y la información Capex BTS cost BTS cost annual price erosion Migration cost per country Benchmarking COFETEL, OFCOM, ACMA, OPTA Ministerio de las tecnologías y la información FCC Macroeconomic Inflation and PPP Exchange rate GDP Social discount rate Taxes Price-demand elasticity Multipliers World Bank IMF U.S. Bureau of Labour Statistics United Nations Telecom Advisory Services KPMG Deloitte Bank of America Merrill Lynch CRC ITU The key BTS sizing drivers included in the model are: - Data traffic subscriber profile: two different data traffic consumption profiles have been addressed; smartphones and mobile PCs / routers / tablets (M/R/T). A typical subscriber profile was assumed (expressed in terms of MB/month/subscriber) for the seven sample countries. An Ericsson report 26 was the main information source for traffic projections. Based on Latin America s subscriptions and data traffic total projections, and a 4G to 3G consumption ratio of x (a 4G subscriber uses 6.3 times the data as a 3G one), the average traffic generated by smartphones and M / R / T users is calculated. As the Ericsson forecast goes only to 2019, projections from 2020 to 2026 were forecast using 2019 as the baseline year and growth was predicted using the CAGR (compound annual growth rate) of previous years for each country. - Subscribers: Based on the GSMA Intelligence forecast to For the period going from 2021 to 2026, the growth rate was estimated based on the subscriber CAGR from previous years and using an adoption curve that mimics the 3G adoption rate (+50% in 10 years). 26 Ericsson Based on 2018 data consumption per technology obtained from CISCO VNI

321 - Technical BTS sizing: such as efficiency, sectors, load factor, distribution factor, % downlink and % peak hour (as described in the technology section). The next figure shows the traffic forecast per subscriber and the different behaviour and consumption patterns for smartphone users and M/R/T users, Figure 12: Data traffic consumption profile MB/sub/month x% CAGR ( ) Smartphone Mobile PC / Router / Tablet % % Data traffic reaches 7,000 MB/sub/month for smartphones in 2026 and 32,000 MB/sub/month for mobile pc, router, tablet in 2026 Mobile PC / Router / Tablet still shows bigger place for additional growth per user, which can be seen in the expected growth CAGR Mobile PC / Router / Tablet have a higher per user data consumption, expected to reach a x4.6 ratio by 2026 In 2026, data traffic per user reaches the 7,000 MB/subscriber/month level for smartphone users and 32,000 MB/subscriber/month for M/R/T users. Both because they are at a different stage of development to smartphones and the broader functionality they provide, there is still room for a lot of growth in data per user for M/R/Ts. This can be seen in the expected CAGR (10.5% for M/R/T as opposed to 4.1% for smartphones). The number of subscribers is the second key assumption to estimate the required number of BTS to support expected traffic growth. The following chart shows the expected number of subscribers and 4G penetrations. 30

322 Figure 13: Subscribers and 4G penetration forecast (2026) Thousand subscribers, 4G penetration vs. total subscribers G penetration Argentina Brazil Colombia Dominican Republic Source: GSMA Intelligence, Ericsson - Traffic exploration (Latam) - June 2014, BNMC estimate Ecuador Mexico Uruguay 72% 85% 74% 64% 72% 78% 75% As mobile markets are already mature, it is estimated a total subscribers CAGR of 2.2% for the sample countries 4G penetration rates are estimated based on GSMA Intelligence forecast to 2020, and then replicating 3G s adoption curve on its first years The chart above shows subscriber projections for the last year of the period of analysis (2026). Taking into account the current mobile services penetration rate, based on an Ericsson estimate, the number of subscribers is expected to grow in the selected countries by a 2.2% annual rate. But 4G technology is undergoing its start-up stage in the region, so 4G penetration rates are still low. 4G projections are based on the GSMA Intelligence forecast to 2020, with the 3G historic adoption curve for later years. By 2026 the 4G penetration rates are expected to be in the range of 64% to 85%. Estimated Capex requirements are based on the following components: - Initial Capex: US$90,000 for a 4G BTS, which includes: radio access, backhaul, core network, backbone and installation costs. This amount has been estimated drawing on models from Colombia s MinTic, COFETEL, Ofcom, ACMA, Opta and the FCC. - BTS Capex annual erosion. A 4.0% annual erosion cost to take into account the falling cost of equipment has been factored into deployment of new BTSs in the future. The 4.0% erosion figure is obtained from COFETEL, Ofcom, ACMA and Opta benchmarking models. - Migration cost: the cost of migrating current L-band users was estimated for each of the seven selected countries. The estimated amounts take into consideration the different cost components required for the migration of these users namely backhaul, e-utran, infrastructure, core network and spare parts. 31

323 As operational expenditures have a significant portion of non-tradable elements, the required Opex on a per-country basis, adjusted by PPP 28 is estimated. The Opex is estimated for urban and nonurban BTSs, and the infrastructure operational costs can be shared with other existing BTSs in the urban areas is assumed. The following figure shows other key parameters that have been also considered in creating the estimates, their use in the model and their source. Figure 14: Coefficient and multipliers Coefficient / multiplier Description Source Demand price elasticity GDP multiplier Job creation multiplier Social discount rate Used to estimate the additional subscribers the price reduction brings as a consequence of the cost savings Estimates the impact on GDP the increase in broadband penetration originates Per 1% increase in broadband penetration, GDP grows pp Estimates the impact on jobs the increase in the number of subscribers originates The job creation multiplier is different for each country. Used in computing the value of funds spent on social projects. Different rate for each country. In Latin America the SDR ranges from 2.9% to 5.1% Katz, R. and Berry, T. Driving demand for broadband networks and services (2014) ITU. Impact of Broadband on the economy. Broadband series. April (2012). Telecom Advisory Services. Economic Benefits of Digital Dividend for Latin America (2011). The Social Discount Rate: Estimates for Nine Latin American Countries. Humberto Lopez. The World Bank Latin America and the Caribbean Region Office of the Chief Economist. June (2008) Note that the GDP and jobs creation multipliers from previous estimates are based on the impact of new users of mobile broadband, however an assumption that new 4G subscribers will originate similar impacts was made 29. Elasticity measures how sensitive a variable is to changes in related variables. In this case, demand price elasticity estimates the number of additional 4G subscribers entering the market once prices are cut. Elasticity is calculated using the following formula: Price vs. demand elasticity = subscribers subscribers price price 28 Purchasing Power Parity 29 Though they are not the same, this assumption was made since the traffic demand of a 4G subscriber is 6.3x the traffic demand of a 3G user, and hence having a new 4G subscriber is more close to having a new subscriber than upgrading the traffic capacity of a 3G subscriber only. 32

324 In the model a demand price elasticity of -0.6x (the negative sign implies a reduction if one of the variables results in an increase in the other variable) is assumed. The elasticity estimation is based on a study from Katz and Berry 30. According to a report commissioned by ITU titled Telecommunication Development Sector. Broadband Series. Impact of Broadband on the Economy (2012) there is consistent evidence of the positive economic impact of mobile broadband services. They enhance economic efficiency, boost GDP and activate job creation. The positive economic impact of broadband derives from higher revenues for telecom operators, gains in productivity, employment generation and an increase in consumer surplus. Drawing on this report, the Latin American multiplier is used, which indicates a 0.158x contribution to GDP growth derived from every increase of 10% in broadband penetration. This impact includes direct effects from the telecommunications industry and indirect spill-overs. Together with a positive impact on the economy, broadband penetration aids in job creation. In this case, the multiplier estimates how many additional jobs will be created based on the number of additional subscribers. A different ratio is used for Brazil, Mexico, Colombia and Argentina as specific information was available for those countries. The social discount rate measures the rate at which a society is willing to trade present for future consumption. As such, it is one of the most critical inputs used in cost-benefit analysis of public projects (and more generally public policies) and is used in computing the value of funds spent on social projects. For the purposes of this report, a World Bank report, The Social Discount Rate: Estimates for Nine Latin American Countries was used. The average social discount rate for the nine countries considered in the report is 3.1%. Four of those countries are also the same sample countries and enable use of exact figures but used the average regional discount rate for the other three. These are the rates used Argentina: 2.9% Brazil: 5.1% Colombia: 4.2% Mexico: 3.3% Ecuador, Dominican Republic and Uruguay: 3.1% (average) IV.c. Key results and findings 30 Katz, R. and Berry (2014) 33

325 The quantitative analysis concludes that allocating the L-band to IMT services will originate material benefits for the telecommunications sector, the government and society. The report s economic model estimates the impact of identifying the L-band for IMT according to the following categories: - Economic value o Additional cost savings and consumer surplus to the sector o Additional 4G subscribers - Direct and indirect impact on: o GDP and economy efficiency o Job creation o Fiscal and tax collections The following table lists the key results of the quantitative analysis: Variable Quantitative impact Cost savings to the sector US$6.3 billion (NPV at 2017, cumulative impact for the 10-year period) Additional 4G subscribers 41.7 million new 4G subscribers (by 2026) GDP indirect impact (through more efficiency) US$19.8 billion (NPV at 2017, cumulative impact for the 10-year period) Job creation 13,120 jobs, to be created in the 10-year period Fiscal impact US$3.2 billion (NPV at 2017, cumulative impact for the 10 years period) 34

326 Millares Cost savings to network operators The cost savings to network operators incorporate savings from Capex and Opex resulting from a lower number of BTS to be deployed. The main assumption here is that, if the L-band were not identified for IMT, telecom operators would have less spectrum available and have to compensate with additional BTSs. The impact was estimated for every sample country for a 10-year period, starting in 2017 and ending in 2026, and then extrapolating the findings for the entire LAC region. Final results have been discounted as of 2017 at the social discount rate. The following figure shows the economic impact of the L-band reallocation on the consumer surplus based on cost savings only. Figure 15: Cost savings to network operators US$ million, Net Present Value at 2017, cash flows discounted at social rate Uruguay Dominican Ecuador Colombia Argentina Mexico Brazil Selected Republic countries Other countries Cost Saving Cost savings amounts to USS 6,3 billion, with all selected countries showing positive outcomes As shown above, the cost savings amount to US$6.3 billion for the whole region. This is the net present value (NPV) in 2017 of the cost savings in year-to-year cash flows. The cost savings comprise operational expenditure and capital expenditure reductions, caused by the availability of extra spectrum meaning less BTS to deploy. If a total of 70MHz of the L-band were to be identified for IMT, the total estimated cost savings are far above the total migration costs. None of the sample countries would experiences a loss from the 35

327 identification of the L-band for IMT, and all show positive outcomes, although some are more positive than others There is a strong correlation between total savings and 4G connections and the estimated traffic. Brazil and Mexico account for most of the cost savings, particularly Brazil which accounts for more than 35% of the total. It is also important to highlight the fact migration costs are not significant when compared to consumer surplus derived from the cost savings passed on to end users. The next chart compares migration costs to discounted consumer surplus over a 10-year period. Figure 16: Value gap between consumer surplus and migration costs Million USD at 2017, NPV, Cash flows discounted at social discount rate 2.403,8 Migration cost (2017) Consumer surplus (2017) 959,7 0,9 458,1 164,6 394,0 0,9 6,0 81,7 39,6105,2 23,5 1,1 51,2 Argentina Brazil Colombia Dominican Republic Ecuador Mexico Uruguay The cost of migrating the current L-band users to another band is not significant when compared to the consumer surplus generation The outcome of identifying the L-band for IMT is positive in each sample country Additional net 4G subscribers As the model assumes cost savings are passed on to consumers under competitive conditions, this will result in retail price cuts. Estimated retail price through ARPU levels and assumed cost savings per subscriber imply a similar reduction in the retail price of 4G services. This price reduction motivates the adoption of 4G by more subscribers, but by how many ultimately depends on the price elasticity, which is -0.6 in this case. The following figure shows the impact in terms of price cuts and the resultant additional subscribers for each sample country. 36

328 Figure 17: Additional subscribers obtained from cost savings translated into retails price cut Million of additional 4G subscribers by ,7 16,1 26,9 4,9 0,3 0,5 0,9 1,8 2,5 Uruguay Dominican Ecuador Argentina Colombia Mexico Brazil Selected Republic countries Retail price cut 8.9% 6.6% 5.8% 3.5% 5.9% 5.6% 7.4% Other countries 41,7 Latam Based on demand price elasticity, price reduction leads to increase in subscribers Additionally, the new subscribers will originate US$ 283 million of additional consumer surplus By 2020, an increase in almost 42 million subscribers in the region as a consequence of the cost savings derived from the identification of the L-band for IMT. These new subscribers would originate an additional economic surplus of US$283 million, bringing the total savings to US$6.6 billion. The retail price cut differs from country to country and ranges from 3.5% to 8.9%. Highlighting the time scale of the cost savings and of the additional consumer surplus is also relevant, as the benefits increase year after year with the major impact coming in the latter years. See the chart in next page for the year to year consumer surplus evolution. Figure 18: Cost savings plus additional consumer surplus over time Current US$ million year Consumer surplus increases year-by-year as Opex and recurrent Capex cost savings start to impact on the cumulative BTS deployment which is not required as a consequence of the L-band s identification for IMT First year negative value shows current L-band users migration costs 37

329 Direct and indirect impacts Finally, it is important to highlight the positive impact the identification of the L-band for IMT can have on society because it causes an increase in the economic efficiency, GDP growth, job creation and tax collection. The estimates of the GDP 31 and the job creation impacts based on multipliers to broadband penetration and subscribers respectively, whilst the fiscal impact leading to an increase in VAT 32 payments and income tax collection, is a consequence of the additional revenue generated by new subscribers. Figure 19: Indirect impact from L-band s identification for IMT Cumulative GDP growth Net present value at 2017 US$ billion Yearly job creation Yearly average New jobs to be created per year Cumulative fiscal impact Net present value at 2017 US$ billion 5, ,1 10,0 19, ,5 1,1 3,2 0,6 1,2 3, ,2 0,3 Additional GDP accounts for US$20 Billion, obtained through higher efficiency ratios resulting from more mobile broadband 1,312 jobs to be created on average per year in Latin America as a consequence of identifying the L-band for IMT. Total jobs to be created during the 10-year period ( ) is 13,120 Extra tax collection: US$ 3.2 billion (VAT and income tax) 13,120 new jobs could be created through the identification of the L-band for IMT, as well as an increase of US$13.2 billion in the GDP of the region and US$ 2.2 billion in extra taxes. A relevant assumption in this modelling is that 100% of the cost savings are passed on to end users. But even if assuming just a portion of the total cost savings were transferred to retail prices that would 31 Including both direct and indirect impacts 32 Value added tax 38

330 still have a major impact although, of course, the number of new subscribers and the three indirect impacts will be reduced proportionally. 39

331 V. Appendix V.a. Technical highlights Current services The following figure shows the allocation of L-Band ( MHz) in each region, in accordance with the ITU s Radio Regulations of MHz 1350MHz 1400MHz 1427MHz 1429MHz 1452MHz 1492MHz 1518MHz 1525MHz Region 1 (Europe) Region 2 (Americas) Region 3 (Asia Pacific) FIXED MOBILE RADIOLOCATION FIXED MOBILE except aeronautical mobile FIXED MOBILE except aeronautical mobile BROADCASTING BROADCASTING-SATELLITE FIXED MOBILE except aeronautical mobile FIXED/MOBILE except aeronautical mobile/ MOBILE-SATELLITE (Space-to-Earth) RADIOLOCATION AERONAUTICAL RADIONAVIGATION RADIONAVIGATION-SATELLITE EARTH EXPLORATION SATELLITE (Passive) RADIO ASTRONOMY SPACE RESEARCH (Passive) RADIOLOCATION SPACE OPERATION (Earth-to-Space)/FIXED/MOBILE except aeronautical mobile FIXED MOBILE FIXED/MOBILE/MOBILE-SATELLITE (Space-to- Earth) FIXED MOBILE FIXED MOBILE BROADCASTING BROADCASTING-SATELLITE FIXED MOBILE FIXED/MOBILE/MOBILE-SATELLITE (Space-to- Earth) Note 1: The grey boxes are the adjacent bands to the spectrum range under analysis. Note 2: The red box corresponds to the internationally-protected spectrum used for radioastronomy and passive services. According to RR s footnote 5.340, all emissions are prohibited in this band. Some countries are also conducting passive research under a programme to search for intentional emissions of extra-terrestrial origin (RR footnote 5.341). Note 3: In the USA and Canada, the sub-band MHz is also allocated for aeronautical radio navigation services (RR footnote 5.334). Note 4: The bands MHz, MHz, MHz and GHz are also allocated to space research (passive) and Earth exploration-satellite (passive) services on a secondary basis (RR footnote 5.339). To protect the Earth Exploration-Satellite Service (EESS) from potential interference from active services in adjacent bands, the Radio Regulations recommend mobile services operating in the bands 40

332 MHz and MHz, comply with the emission limits from stations along the 27MHz of the EESS (Passive) band 33 : - -60dbW for mobile service stations, except transportable radio-relay stations dBW for transportable radio-relay stations. In any case, however, mobile services based on IMT systems are likely to comply with these emission limits 34. For ITU Region 2 (America), RR footnote has prioritized the use of the MHz band for aeronautical mobile service (for telemetry) over its being used for mobile services. In the following sections, we outline the most relevant features of the allocated services. Radiolocation Radiolocation is a location service able to perform radiolocating 35, that is finding a specific location through the use of radio waves. Long-range radars and terrestrial radio relay are examples. Radiolocation services include long-range aeronautical radionavigation radar systems. These radar systems are used to monitor aircraft and other aerial objects inside the national airspace, along border areas, and around military bases and airfields. As well as radar systems used for aeronautical radionavigation, the armed forces of some countries operate tactical radar systems in the MHz band 36. Tactical radars are often used to operate in a battlefield environment, alongside other systems. One such system is the Tethered Aerostat Radar (TAR) system, which also operates in the band. The TAR consists of balloon-mounted radars used for monitoring the US southern borders and Caribbean airspace for drug interdiction. But long-range radars could also operate in the band MHz according to the technical features mentioned in ITU-R M Characteristics of and protection criteria for radars operating in the radio-determination service in the frequency band MHz. 33 ITU, Resolution 750 (Rev. WRC-12), Idem. Footnote 3 35 ITU, Radio Regulations Vol I, Specially from USA, Canada and Europe 41

333 There are also different types of radars, with land-based fixed or transportable platforms, operating in the band MHz. The uses of these radar systems include Air Traffic Control and weather observation. These radars use continuous wave (CW) pulses and frequency modulated (chirped) pulses with typical transmitter RF emission bandwidths from 66 khz to 6MHz. Land-based radar systems could also operate at MHz; as already do, for example, radars employed by US Army Field Artillery Units to detect and track incoming projectiles and provide the locations of both the source and the point of impact 37. Broadcasting and Broadcasting-Satellite According to the ITU s Radio Regulations of 2012, broadcasting services are radio communication services, whose transmissions are intended for direct reception by the general public. These services may include sound transmissions, television transmissions and others (CS) 38. In broadcasting-satellite services, the signals are transmitted or retransmitted by space stations, and therefore require international coordination. Satellite Digital Audio Broadcasting (S-DAB) is a satellite-enabled system for the digital broadcasting of a wide variety of audio content to fixed and mobile receivers (including portable, handheld and vehicular). The three main types of S-DAB operation are: - Standalone satellite component (i.e., without complementary terrestrial component); - Satellite component of a hybrid solution with complementary terrestrial broadcasting networks operating in the spectrum used by the satellite. - Standalone satellite component, complementing nationally-licensed terrestrial broadcast networks. Fixed Services A fixed service is a radio communications service between specific fixed points such as the microwave links between two transmitter/receptor stations. The most-used bandwidths for these services in the 1.5GHz band are 250KHz, 500KHz, 1MHz, 1.75MHz and 3.5MHz. 37 NTIA TR Technical Characteristics of radiolocation system operating in the GHz band. (1999) 38 ITU, Radio Regulations Vol I,

334 If L-Band is identified for IMT, current users operating fixed services in this band can be migrated to other bands allocated to fixed services including 1.8GHz, 2GHz, 3.8GHz, 4.2GHz, 6GHz, 7GHz, 11GHz and 15GHz. Mobile Service According to the ITU Radio Regulations definitions, a mobile service is a radio-communication service between mobile and land stations, or between mobile stations such as, for example, mobile broadband based on IMT systems. Mobile broadband is a terrestrial radio communications service connecting base stations with mobile devices to provide various services and applications to end-users, including mobile Internet, entertainment, and video on-demand. In the figure below, we show a non-exhaustive list of technologies available for mobile broadband based on IMT systems, and the spectrum required for their operation. Technology Carrier Size (Spectrum Quantity) HSPA+ FDD and TDD operation Carriers of 5MHz and 10MHz (Dual Carrier) Up to 20MHz in 3GPP Rel 9 and beyond LTE FDD and TDD operation Carriers of 1.5MHz, 3MHz, 5MHz, 10MHz, 15MHz and 20MHz LTE - Advanced FDD and TDD operation Same carriers supported by LTE and up to 100MHz in 3GPP Rel 10 and beyond Radioastronomy A Radioastronomy service is used for studies of astrophysical objects through radio frequencies and as such the ITU invites administrations to cooperate in protecting radioastronomy services from interference. 43

335 Administrations must take into account the extremely high sensitivity which is a feature of radioastronomy stations as well as the need for long periods of observation without harmful interference. Because of these constraints, there are only a small number of radioastronomy stations in each country 39. Article 29 of the ITU Radio Regulations specifies special considerations for radioastronomy services in order to protect them from harmful interference. It is specified, for example, that the locations of radioastronomy stations should be selected with due regard to the possibility of harmful interference, and in the adjacent bands the administrations are urged to take all practicable steps to protect radioastronomy services from harmful interference. Passive Services: Earth Exploration-Satellite Service (EESS) and Space Research Service The ITU Radio Regulations define EESS as a radiocommunication service between earth stations and one or more space stations, including links between space stations 40. EESS provide information about characteristics of the planet and natural phenomena. This is obtained from sensors on Earth Satellites as well as collected from airborne or Earth-based platforms. The information is then distributed to earth stations. Resolution 673 (Rev.WRC-12), concerning the importance of earth observation radiocommunication applications, states that the use of spectrum for earth observations has a material socio and economic value, and urges administrations to protect these systems from harmful interference 41. Channelization plan options Based on the spectrum range under analysis (i.e / MHz), we propose different band plan scenarios, as shown in the next figure. 39 ITU, Radio Regulations Vol I, ITU, Radio Regulations Vol I, ITU, Radio Regulations Vol III - Resolutions,

336 1350.0MHz MHz MHz MHz MHz MHz MHz MHz MHz MHz Figure 20: Band plan options for L-band Option A: 3GPP Band 11 and Band 21 20MHz 15MHz 20MHz 15MHz 2x35MHz Option B 50MHz 40MHz (Unpaired*) 50MHz 2x50MHz + 40MHz Option C 25MHz 25MHz 40MHz (Unpaired*) 2x25MHz + 40MHz Option D 50MHz (Unpaired*) 35MHz 35MHz 2x35MHz + 50MHz Note: *Unpaired spectrum available for Supplemental Downlink (SDL) o TDD Radio Access The worst scenario in terms of the quantity of spectrum made available is the identification of 2 x 35MHz for IMT (Option A: 3GPP band 11 + band 21), while the best scenario is Option B: 100MHz (2 x 50 MHz) paired and 40MHz unpaired. Following are special considerations which need to be taken into account for each option. - Option A: Band Plan based on 3GPP specifications for band 11 and band 21 This option allows the use up to 2 x 35MHz for IMT systems deployment. It does not allow taking advantage of the spectrum available below 1400MHz. The main advantage of this option is related to commercial availability of the equipment and terminals ecosystem Option B: Band Plan includes all spectrum within the spectrum range under analysis. It provides for 100MHz of paired spectrum and 40MHz of unpaired spectrum for use either with Supplemental Downlink (SDL) or TDD technologies. Although this proposal allows using the entire 42 There are two LTE networks in this frequency band in Japan. 45

337 spectrum of frequency band MHz and MHz, there are no technological developments or any related trials or proposals linked with this channelization. - Option C: Based on CEPT studies and Qualcomm s proposal for SDL Option C adopts the Qualcomm proposal for SDL development in frequency band MHz, and additionally considers the results of a trial conducted by Orange, Qualcomm and Ericsson. This option is also aligned with a proposal from Orange focusing on the utilization of MHz to accomplish harmonization and promote economics of scale. - Option D: Based on 3GPP specification for band 11 and band 21 This option adds the band MHz to the channelization defined by 3GPP. In this context, the additional 50 MHz could be used as unpaired spectrum for SDL or TDD. Thus, it is possible to leverage the technology development available for LTE in this frequency band, and use the spectrum available below 1400 MHz. Terminals ecosystem at a glance In 2009, Japan assigned 70 MHz of spectrum within L-Band to mobile operators (i.e., Softbank and NTT DoCoMo) to deploy LTE networks. Both operators launched their commercial networks between 2010 and So there are already two commercial LTE networks operating in the 1.5 GHz band in accord with 3GPP specifications 43, and so a limited amount of equipment for IMT services in L- Band is already available. 43 Band 11 ( MHz/ MHz) and Band 21 ( MHz/ MHz) in 3GPP Specifications since Release 8. 46

338 V.b. Per country analysis This section outlines the findings of the per country research, which entailed the review of public sources as well as direct interviews with local administrations. For each sample country there are a brief description of the IMT spectrum, a glance at the mobile market and specific findings on the L-band. Brazil Frequency allocation and IMT services Current allocation In Brazil the Agencia Nacional de Telecomunicaciones (Anatel) is responsible for the spectrum management and planning. The bands identified for IMT services are: 450MHz, 700MHz, 850MHz, 900MHz, 1800MHz, 1900MHz, 2100MHz and 2500MHz. The figure below shows the services allocation in the already assigned IMT bands 44. Figure 21: Services allocation in bands mostly used for IMT in Brazil MHz 450MHz band 850MHz and 900MHz bands 1800MHz and 1900MHz bands Mobile Fixed Secondary usage Satellite Earth to space Mobile Fixed Satellite Earth to space Space to earths Mobile Fixed Secondary usage Radiolocation satellite Meteorological aids Source: Anatel 44 The figure includes only the already assigned bands. 47

339 Figure 22: Services allocation in bands mostly used for IMT in Brazil (Cont.) MHz 2,1GHz band 2.5 GHz band Mobile Fixed Satellite Space research Space operation Space to space Mobile Fixed Satellite Radiolocation satellite Meteorological aid Space to earth Earth to space Space to earth Space research Secondary usage Earth exploration satellite Radio astronomy Source: Anatel Brazil has a total of 660MHz identified for IMT services, positioning Brazil as one of the country with most spectrum identified for IMT. However, to date Brazil reaches only between 34% and 49% of the total spectrum requirements estimated by ITU 45, for the high and low-density settings, respectively. In the same line, when compared to the spectrum demand forecasted by GSMA 46, Brazil reaches only between 39% and 58% of the high or low total spectrum demand forecast, respectively. IMT assignments The bands of 450MHz, 850MHz, 900MHz, 1800MHz, 1900MHz, 2100MHz y 2500MHz have been assigned to four national mobile operators (Vivo, TIM, Claro and Oi), to two regional mobile operators (Algar and Sercomtel), and one yet to commence service as national mobile operator (Nextel). In Brazil the IMT spectrum licensing is structured in 3 regions, 10 services areas and 34 sectors; therefore, the assignments have regional reach, resulting in different spectrum holdings per region per operator. Most of the current spectrum identified for IMT has been already assigned (i.e. approximately 529 MHz). The licenses timeframe is 20 years maximum; however, most of the IMT spectrum has been assigned for a 15 years period, with a one-time renewal at finish. 45 Report ITU-R M Revised spectrum forecasts using the new spectrum model, prepared for GSMA by Coleago Consulting (2013) 48

340 There are more than 50 thousand base stations, with TIM, Vivo and Claro having approximately 15 thousand base stations each and Oi having approximately 4 thousand base stations. The economic compensation for the IMT licenses comprises an up-front payment to be ultimately determined through an auction process; thereafter, there are no additional economic obligations. However, after the first (and unique) renewal, incumbent operators shall pay a biennial fee of 2% on previous year gross income (net of taxes and contributions) and the TFI 47. The Brazilian mobile telecommunications market accounts for million subscribers (136% mobile penetration). As most markets of the region, the Brazilian market has experienced a dramatic growth in last years; it has more than doubled its number of subscribers between 2007 and Figure 23: Brazil s mobile market evolution Mobile services clients evolution Million Penetration % (subscribers / population) As a consequence of the market saturation, in last years the mobile market has slowed down its growth rate. There are four national operatores (Vivo, Claro, TIM and Oi) and two regional operators (Algar and Sercomtel), while one national operator is yet to commence service (Nextel). The following figure shows the current market structure in terms of subscribers. Figure 24: Brazil mobile market structure 47 Tasa de Fiscalización de Instalación 48 Anatel, December

341 In subscribers 2 Others: 0,6% 19% % 69 25% 73 27% Source: Anatel The Brazilian market is featured by its well-balanced structure and the absence of a dominant player as Claro, TIM and Vivo have similar market shares (Oi is the fourth player in the market with 19% market share). IMT roadmap and new bands under study The Brazilian administration has announced an auction of 80MHz in the 700MHz band (digital dividend) to take place in late September In addition, Anatel is assessing new spectrum bands for identification for IMT: MHz, 1427MHz-1518MHz and the MHz. L-band frequencies ( MHz) Current allocation L-band includes the frequencies from 1350MHz to 1525MHz, with this service allocation: MHz: radiolocation MHz: earth exploration satellite, radio astronomy, space research MHz: fixed, mobile, space operation (earth to space) MHz: fixed and mobile MHz: fixed, mobile, broadcasting and broadcasting-satellite MHz: fixed and mobile MHz: fixed, mobile and mobile-satellite (space to earth) The figure below depicts the allocation ofr the L-band. Figure 25: L-band s services allocation in Brazil 50

342 MHz Radiolocation Aeronautical radionavigation Radionavigation Fixed Mobile Space operation Earth exploration sate. Space research Radioastronomy Broadcasting Satellite Earth to space Space to earth The range from 1427MHz to 1525MHz is allocated to mobile services The mobile services allocation is shared with fixed services (1427MHz to 1525MHz) In compliance with international standards, spectrum in the range MHz is allocated to radiolocation Spectrum from 1400MHz to 1427MHz is reserved for earth exploration satellite, radioastronomy and space research Source: Anatel The allocation to mobile services is shared with the allocation to fixed services (range MHz), space operation ( MHz) and broadcasting ( MHz). In line with international guidelines, the spectrum between 1350MHz and 1400MHz is allocated to radiolocation. The frequencies ranging from 1400MHz to 1427MHz are reserved for earth exploration satellite, radioastronomy and space research. Assignments and actual use The L band is used by a high number of stakeholders for point-to-point and point-to-multipoint links, who belong to different industries including: telecommunications, oil and electricity and utilities. Today, there are more than six thousand licenses assigned in the L-band; the vast majority being old assignments with obsolete equipment deployed throughout the country and focused on rural areas. Potential migration Although feasible, the migration of existing users of the L-band to other bands faces some barriers: the large number of users to migrate and the migration of users whose licenses that have not yet 51

343 MOBILE SATELLITE ( EARTH TO SPACE) METEOROLOGHY PER SATELLITE SPACE INVESTIGATION MOBILE SATELLITE (SPACE TO EARTH) MOBILE SATELLITE (SPACE TO EARTH) FIXED FIXED FIXED FIXED MOBILE MOBILE MOBILE MOBILE expired 49. However, note that while licenses could be assigned for a period of up to 20 years 50 they had instead been awarded for a period of 10 years only, excluding automatic renewal. Public agenda regarding for the L-band Anatel has pointed out that the bands MHz and 1427 MHz1525MHz should be taken into account among the candidates bands for the provision of IMT services (to be proposed at the next World Radiocommunication Conference (WRC-15)). Brazil has already submitted to CITEL a proposal for identification of the L-band for IMT in Region 2. Mexico Frequency allocation and IMT services Current allocation In Mexico the Federal Telecommunications Institute (IFT) is responsible for the spectrum management and planning. The following figure shows the services allocation in the currently assigned IMT bands: Figure 26: Services allocation in bands mostly used for IMT in Mexico 850MHz band 1900MHz band AWS band FIXED MOBILE MOBILE Source: Instituto Federal de Telecomunicaciones 49 Similar to the barriers for the assignation of the 700 MHz band. 50 Lei n.º 9.472/1997 (Lei Geral de Telecomunicações LGT) 52

344 The mobile market in Mexico comprises four operators (Iuscacell, Movistar, Nextel and Telcel) and two MVNO (Maxcom and Megacable). All operators offer mobile broadband through UMTS / HSPA+ networks whilst Movistar and Telcel have already launched LTE services in The country has assigned approximately 230MHz of IMT spectrum. Licenses are structured on a regional basis, with 9 regions for the entire country. The 600MHz of the spectrum already identified for IMT services accounts only for a portion (between 31% and 45%, depending on the market density) of the total spectrum requirements estimated by ITU 51. According to GSMA estimates 52, the 600MHz already identified for IMT represent between 39% and 61% of the total spectrum requirements by As for the radio access infrastructure, incumbent operators have deployed almost thirty thousand (29.249) radio stations throughout the country. The economic compensation for the use and operation of radio spectrum is established on a year basis and takes into account the parameters contained in the Federal Law Rights, articles 244-B, 244- D and 244-E. In last years, the penetration of mobile market has reached a two-digit rate of growth, approaching a saturation stage at the present time. Figure 27: Mobile market structure in Mexico Mobile telecommunication services evolution Mobile telecommunications market share Numbers of suscriptors (milliom) Mobile penetration 86% 89% 80% 82% 74% 70% 63% 54% 45% % 36% % 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 22% 4% 4% 70% Mobile telecommunication penetration has grown at a compound annual growth rate of 13% Over 90% of the market is concentrated in two operators 51 Report ITU-R M Coleago Study (2013) 53

345 The mobile market penetration has grown from 30% in 2003 to 89% in The market is highly concentrated with 90% of the total at hands of two operators. Spectrum roadmap and new bands under study Even though the administration has not yet announced an assignation process, the following bands are being considered for potential assignations: Figure 28: Assignation roadmap for IMT bands in Mexico Band (MHz) Sub band (MHz) IMT assignment date To be defined Before the end of / To be defined Tentatively To be defined Tentatively To be defined Tentatively To be defined To be defined Tentatively To be defined Source: Instituto Federal de Telecomunicaciones In addition, the MHz and the MHz bands will be proposed for identification for IMT services. Current allocation In Mexico, the bands MHz and MHz are considered in the scope of the L-band. The figure below shows the frequency allocation in the L-band. 54

346 RADIOASTRO- NOMY BROADCAST- ING SATELLITE MOBILE SATELLITE (SPACE TO EARTH) RADIOLOCATION SPACE RESEARCH FIXED BROAD- CASTING FIXED MOBILE EARTH EXPLORA - SATEL. (PASSIVE) FIXED FIXED Figure 29: L-band s services allocation in Mexico MHz Most of the band, 98MHz ( MHz) is allocated to fixed services MOBILE FIXED Only 9MHz are allocated to mobile service (on a shared basis with fixed service) Radio spectral range between MHz follows the international trend of radiolocation allocation Source: Instituto Federal de Telecomunicaciones The figure above shows that most of the band (i.e., 98MHz in the range MHz) is allocated to fixed services whilst only 9MHz are allocated to mobile services. The range between MHz follows the international trend of radiolocation allocation. Assignments and actual use The range MHz includes licenses for fixed services. The licenses were directly assigned and have a different expiration timetable. The band is mostly used for microwave point to point links, where the private companies are the main users and there is a handful of state-owned companies. Figure 30: Actual use of L-band in Mexico Frequency band (MHz) Current uses Aeronautical radionavigation systems, aeronautical communications and related uses Point to point fixed service systems Point to point fixed service systems Point to point fixed service systems According to the note of radio regulations this frequency band is limited to digital audio broadcasting satellite service. However it is not expected to use the band for such services in Mexico Point to point fixed service systems Source: Instituto Federal de Telecomunicaciones Potential migration This range is in the Mexico s records that are being precessed by ITU for mobile satellite service. However it is not expected to use the band for such services in Mexico 55

347 The migration of users in the range MHz is not deemed possible as they operate protected services. However, it would be possible migrating fixed service operating in the range MHz As mentioned before, there are few licenses of fixed services in the range MHz. Nevertheless, some of those users have also microwave point to point systems operating in other bands: MHz, 7GHz, 15 GH and 23GHz. Position regarding future use of the L-band According to the answer submitted by the delegation of Mexico to the decision CCP.II / DEC. 162 (XXII-13) 53, although it is not expected a modification of the current allocation of the range MHz, the possibility of identifying the range MHz for IMT services is under analysis. It is worth mentioning that there are no definite plans to migrate current users to another band. The table below summarizes the plans for each portion of the L-band. Figure 31: Plans for L-band in Mexico Frequency band (MHz) Action No changes are expected The possibility of identifying this frequency range for IMT is under analysis The possibility of identifying this frequency range for IMT is under analysis The possibility of identifying this frequency band for IMT is under analysis (according to the discussions that are being held within the ITU-R and 3GPP.) The possibility of identifying this frequency range for IMT is under analysis The possibility of identifying this frequency range for IMT is under analysis Source: Instituto Federal de Telecomunicaciones Colombia 53 Information request on current and planned uses of the bands MHz and y MHZ for satellite and land services to be submitted by OEA/CITEL administrations 56

348 RADIOLOC. MOBILE* FIXED RADIOLO- CATIO RADIOLO- CATION MOBILE MOBILE MOBILE SATELLITE (SPACE TO EARTH) MOBILE MOBILE MOBILE - TRUNKING MOBILE* MOBILE - TRUNKING MOBILE* FIXED MOBILE MOBILE MOBILE FIXED FIXED FIXED FIXED FIXED Frequency allocation and IMT services Current allocation Law 1341 of 2009 defines the structure and basic concepts of the telecommunications industry in this country. This law establishes the framework of the telecommunications sector, the competition rules and consumer protection, the coverage and quality of services, the promotion of domestic and foreign investment in the sector, paving the road for the development of new technologies, the guidelines to assure the efficient use of radio electronic spectrum as well as identifies the public entities responsible for the sector administration. The regulatory entities are described in the Law 1341/2009 and its structure follows a hierarchal framework. At the top of the pyramid there is the Ministry of Information and Communications Technologies (MINTIC), followed by the National Spectrum Agency (ANE). Article 26 of Law 1341/2009 describes the 13 functions of the National Spectrum Agency. The purpose of the ANE is to provide technical support for the management, monitoring and control of the radio spectrum, in coordination with other entities. Colombia has identified for IMT the bands: 850MHz, 1900MHz, AWS, 700MHz, 900MHz and 2.5GHz bands. The following figure shows the services allocations in the IMT bands currently assigned. Figure 32: Services allocation in bands mostly used for IMT in Colombia MHz 850MHz band 1900MHz band * Excludes aeronautical mobile Source: ANE - CNABF, 2013 Figure 33: Services allocation in bands mostly used for IMT in Colombia (Cont.) MHz 57

349 FIXED FIXED MOBILE RADIO SAT. MOBILE SPA. RES. MOBILE SATELLIT. MOBILE FIXED SATELLITE MOBILE MOBILE MOBILE FIXED SAT. FIXED FIXED SAT. FIXED FIXED FIXED FIXED FIXED AWS band 2,5GHz band FIXED FIXED SATELLI. MOBILE RADIO SAT. SAT. EXP. RADIO SAT. EXP. RADIO SPA. RES. SPA. RES. Source: ANE - CNABF, 2013 Colombia is using 415MHz out of the total spectrum identified for IMT in the bands 850MHz ( / / / / MHz), 1900 MHz ( / MHz), AWS ( / MHz) and 2.5GHz ( MHz). IMT assignments The Colombian mobile market includes five mobile network operators (Claro, Movistar, Tigo-UNE, Directv and Avantel 54 ), and five MVNOs (ETB, Éxito, UNE, Uff! Mobile and Virgin Mobile). The number of mobile operators has increased recently as a result of the spectrum auction held in June 2013, in which three entrants won licenses: Avantel (2 x 15MHz in the AWS band) consortium ETB-Tigo (2 x 15 MHz AWS) and DirecTV (2 x 15MHz and 40MHz TDD in 2.5GHz band). In June 2012 UNE has already began marketing LTE services. Additionally, the five companies that won spectrum blocks in the G auction (Avantel, Claro, Tigo ETB-Consortium, DirecTV and Movistar) will launch own 4G services during Colombia has achieved a significant increase in the total assigned spectrum, climbing from 190MHz in 2010 to 415MHz in Figure 34: IMT spectrum holdings in Colombia 54 Also an iden operator 58

350 850MHz 1900MHz 4G Transitory 415MHz 415Mhz MHz MHz 220MHz The IMT assigned spectrum has more than doubled in last five years: 225MHz has been assigned Source: Agencia Nacional de Espectro The auction from June 2013 allowed increasing the assigned spectrum from 225MHz to 415MHz. Even though its progress, the spectrum identified for IMT in Colombia (i.e., 590MHz) still accounts only for a portion (between 30% and 44%) of the total spectrum requirements as estimated by ITU by 2020 (Report ITU-R M ) as well as a portion (between 37%-56%) of the GSMA total spectrum demand estimates 55. Figure 35: Mobile market evolution of Colombia Mobile and fixed telecommunication services penetration, percentage Fixed penetration Mobile penetration CAGR % 75% 89% 88% 98% 102% 105% 98% 17% 48% 25% 17% 17% 17% 18% 17% 15% 15% 15% 14% 13% -3 % The mobile penetration climbed from 17% in 2004 to 98% in 2013 Fixed penetration in the first quarter of 2013 was 13%, falling from 2003 at a 3% annual rate Source: EIU, World Celular Information Service, Informa T&M, Informe Primer Trimestre MINTIC 55 Coleago (2013) 59

351 The mobile market penetration experienced a dramatic growth between 2004 and 2013, climbing from 25% to 98%. Between 2004 and 2013 the number of mobile subscribers grew at 17% annual rate, increasing from 9.1 million in 2004 to 46.4 million in Note that between 2012 and 2013 the number of subscribers fell around 5%, reducing the customer base from 49.1 to 46.4 million users. However this drop of 2.7 million subscribers was due to a decrease of Claro s customers, as a result of a purge of its database. Figure 36: Main operators of the Colombian mobile market Million of subscribers, percentage Claro Movistar Total Suscribers Tigo UNE 0,1% Uff Móvil 0.5% 0.7% 0.7% % 12.4% 13.5% 14.8% % 10.0% 25.0% 22.0% 24.0% 24.6% 23.9% 9.0% % 8.0% % 26.0% 9.0% % 66.0% 66.0% 66.0% 68.0% 66.0% 62.4% 61.9% 59.6% 12.0% % 63.0% 56.0% ,1 21,8 29,7 33,9 39,6 40,5 44,3 46,2 49,1 46,4 CAGR % 0% 19% 23% 16% 21 % Source: The market structure remained stable throughout the years Claro in the main operator with 59.6% market share, though its share has been decreasing since 2009 Informa T&M, MINTIC As shown in the figure above, the market shares have remained stable throughout the years. During the first quarter of 2013, Claro held 59.6% of mobile voice market having 27.7 million subscribers; Movistar concentrated 24.8% of the total base with 11.5 million users; Tigo accounts for 14.8% of the market with 6.9 million users; Uff Móvil has 0.7%, with 0.3 million customers; and finally, UNE has an estimated of 0.05 million of subscribers. As for economic compensation, there are two different types of fees related to the spectrum licenses: - up-front payment at the assignation - periodic payment, calculated on the operator gross income IMT Roadmap and new bands under study 60

352 SPACE RESEAR. MOBILE MOBILE SATELLIT. FIXED RADIOASTRO- NOMY FIXED MOBILE SATELLITE EXPLORATION FIXED FIXED In last years, the local administration has been assessing new bands to be identified for IMT services, taking into account the next WRC coming in 2015; the ANE 56 has carried out a study on six selected bands, in order to analyse the suitability and opportunity of its potential future identification for IMT. The frequency bands under analysis are: MHz MHz MHz MHz MHz MHz The next step is recommending the strategy to implement the effective, timely and efficient release of those frequency bands, with the core target that Colombia meeting the spectrum requirements by L-band frequencies ( MHz) Current allocation For Colombia this study considers the ranges MHz and MHz. The figure below shows the frequency allocation for different services in the band under analysis. Figure 37: L-band s services allocation in Colombia MHz Most of the band, 98MHz ( MHz,) is allocated to fixed services Only 33 MHz ( MHz) are shared with the mobile service Radio spectral range between MHz follows the international trend of radiolocation allocation Source: ANE - CNABF, Agencia Nacional del Espectro, the local technical entity responsible for the spectrum planning and management 61

353 Most of the band under analysis, 98MHz in the range of MHz, is allocated to fixed services and only 33MHz ( MHz) are shared with mobile service. The band MHz follows the international trend of radiolocation allocation. Assignments Today seven companies are operating in the range MHz. The band is being mostly used for microwave transmission systems point to point and point to multipoint links for support services to the mobile telecommunication operators. There is also another service operated by a local television company that has two stations transmitting at the same frequency. The figure below shows the number of licenses according to their expiration dates timetable during the period Figure 38: Timetable for L-band s licenses expiration dates in Colombia # licenses 60 Released licences (% accumulate) Expired licences % 100% 40 69% 71% 78% % 6 7% 4 2 4% 2% mar-14 sep-15 nov-15 ene-16 sep-17 dic-18 dic-19 dic-20 dic-21 Sin vto. Source: Agencia Nacional de Espectro According to the previous figure, most licenses will expire by By 2020 more than 90% of the licenses will automatically expire, and by the end of 2021 the total band will be clear 57, paving the way for the reallocation plans. Actual use and potential migration 57 With an uncertainty of 4.3% since there are assignments with unknown due dates. 62

354 According to a survey completed by the ANE in 2014, only one operator, the largest user of the L- band, would be still interested in continuing operating within this band. However, the other point to point links users would agree to migrate to other bands: 5GHz, 7GHz and 8GHz. From year 2021 onwards, there would be no migration barriers since 100% of the current licenses would be expired. The following figure briefly shows the findings of the ANE s study regarding the compatibility between current services and IMT Figure 39: Compatibility L-band current services and IMT MHz Allocation on a primary basis Allocation on a secondary basis Possible compatibility with IMT Radioastronomy Radiolocation Exploration Satellite Space research Space operation Fixed Aeronautical Radionavigation Radionavigation Satellite Mobile Mobile Satellite IMT Compatibility Source: Agencia Nacional de Espectro Additionally, it is worth mentioning that the range MHz is fully compatible with the industry initiatives aimed at implementing Supplemental Downlink in MHz. The identification for IMT services would provide about 91MHz of bandwidth, allowing for the specification of very wide carriers. This would improve networks performance, coming close to the estimated needs of data traffic. Position regarding future use of the L-band It would feasible to identify the L-band for the provision of IMT services, and the administration would presumably propose this band for its identification to IMT. 63

355 FIXED MOBILE 698 MHz 806 MHz 894 MHz 1710 MHz 2025 MHz As a matter of fact, the ANE has already conducted a preliminary assessment on the compatibility with IMT services provision in this band and the involved migration costs. Ecuador Frequency allocation and IMT services Current allocation The telecommunications sector in Ecuador is ruled by the Special Telecommunications Law enacted in August 1992 together with its subsequent reforms. The Ministry of Telecommunications and of Society Information (Mintel), the National Telecommunications Council (CONATEL) and the National Telecommunications Secretary (SENATEL) are the main public agencies responsible for the sector regulation and for the spectrum management and planning. The National Telecommunications Council (CONATEL), in exercise of its authority set out in the Special Telecommunications Reform Law, has approved 58 the reforms and updates of the National Frequencies Plan (NFP) of September The figure below shows the services allocations in the currently assigned IMT bands. Figure 40: Services allocation in bands mostly used for IMT in Ecuador 850MHz band 1900MHz band FIXED MOBILE MOBILE FIXED Source: CNAF - CONATEL 58 Resolution No.TEL CONATEL-2012 of July

356 FIXED MOBILE FIXED MOBILE 698 MHz 806 MHz 1710 MHz 2025 MHz 2100 MHz 2200 MHz Figure 41: Services allocation in bands mostly used for IMT in Ecuador (Cont.) 700MHz band AWS band MOBILE FIXED Source: CNAF - CONATEL IMT assignments Mobile telecommunication operators providing services in Ecuador are CONECEL SA (CLARO) OTECEL (MOVISTAR) and CNT. Ecuador has 180 MHz of IMT spectrum currently assigned. The following figure shows the spectrum holdings per operator and the market structure. Figure 42: Mobile market structure and spectrum holding in Ecuador Spectrum holdings per operator (MHz) Mobile telecommunication market share 700MHz 850MHz AWS 1900MHz % 2% 69% Claro Movistar CNT EP CLARO y MOVISTAR operate only in the 850MHz and 1900MHz bands CLARO is the operator with the highest penetration, with nearly 70% market share Source: SENATEL CLARO and MOVISTAR operate only in the bands 850MHz and 1900MHz, holding each one a total of 35MHz. CLARO is the operator with the highest penetration with almost 70%, followed by MOVISTAR, and finally, by the state owned company, CNT. 65

357 Mobile telecommunication bands should be assigned to the private operators by auction, procedure known as Competitive Process. However, the radio spectrum for LTE services was assigned directly to CNT, a public owned operator (in December 2012, the government assigned 70MHz of spectrum - 30MHz in 700MHz band and 40MHz in AWS to CNT). In addition, the spectrum cap per operator is 65MHz; nevertheless this cap does not apply to state-owned operators. Regarding the economic compensation for the licenses, the conditions included in last spectrum renewal (2008) were: - a fixed value for the renewal - a monthly fee called Tarifa A, calculated according to the number of radio stations and bandwidth as well as a payment per active customer - a variable payment on the operator gross revenues (2.93%) Regarding the number deployed base stations, Claro has almost 4,800 whilst Movistar around 3,000, including in both cases GSM 850, GSM 1900 and UMT 850 technologies. In the case of CNT, it has roughly 1,050 base stations with CDMA, UMTS 1900, LTE and LTE AWS 700 technologies. In February 2013, Ecuador became the fourth country in Latin America (after Colombia, Chile and Mexico) to adopt the APT band plan for the implementation of LTE technology in the 700 MHz band, known as the Digital Dividend. However, this spectrum is still occupied by broadcasting systems and will be released once the TV switch over is completed (analogue switch off commenced in 2010 and should be completed by 2017, with a potential period extension to 2021). IMT roadmap and new bands under study Ecuador has a national broadband plan aimed at having at least 75% of the total population connected by In order to meet the objectives outlined in this plan, the Ministry of Telecommunications would have to ally with mobile operators to cover remote areas and hence more IMT spectrum assignments are expected. L-band frequencies ( MHz) Current allocation The range of frequencies and MHz were considered for this study. The figure below shows the current frequency allocation for different services in the band under analysis. 66

358 Figure 43: L-band s services allocation in Ecuador Most of the band, 98MHz ( MHz) is allocated a fixed services There is no mobile service allocation in the band Radio spectral range between MHz follows the international trend of radiolocation allocation Source: SENATEL The figure above shows that most of the band, 98 MHz ( MHz) is allocated to fixed services, with no allocation to mobile service. The range between MHz follows the international trend of radiolocation allocation. Assignments and actual use The band is being mostly used for national radio links of land fixe service, as detailed in the figure below. Figure 44: L-band current usage in Ecuador Frequency band (MHz) Current use None Radio electronic links of land fixed service Radio electronic links of land fixed service Radio electronic links of land fixed service Radio electronic links of land fixed service Radio electronic links of land fixed service and mobile satellite Source: SENATEL As shown in the figure above, the frequency range between MHz is not used, and the rest of the band is being used by radio electronic links from fixed services. Note that these licences have 67

359 a 5-year timeframe. The L-band links are used mainly by carriers and private and state-owned companies. There are approximately 650 links in the L-band. In the satellite mobile service there is a satellite broadcaster operator with around 200 subscribers. Potential migration The following figure shows the plans for the L-band as informed by Senatel. Frequency band (MHz) Plans The possibility of identifying this frequency range for IMT services is being analyzed The possibility of identifying this frequency range for IMT services is being analyzed The possibility of identifying this frequency range for IMT services is being analyzed The possibility of identifying this frequency range for IMT services is being analyzed The possibility of identifying this frequency range for IMT services is being analyzed None Source: SENATEL In brief SENATEL is assessing the identification of almost the entire L-band for IMT, just excluding the portion MHz. Position regarding future use of the L-band According to the answer submitted by the delegation of Ecuador to the decision CCP.II / DEC. 162 (XXII-13) 59, the possibility of identifying the band for IMT is under analysis, and the administration will presumable proposed such identification. 59 Information request on current and planned uses of the bands MHz and y MHZ for satellite and land services to be submitted by OEA/CITEL administrations 68

360 Uruguay Frequency allocation chart and IMT services Current allocation In Uruguay, the Unidad Reguladora de Servicios de Comunicaciones (URSEC) is the responsible entity for the management and planning of radio spectrum. The next figure shows the services allocation in the already assigned IMT bands. Figure 45: Services allocation in bands mostly used for IMT in Uruguay 850MHz band 1900MHz band 2.1GHz band Radio amateurs Broadcasting Fixed Mobile Radiolocation Mobile (except aeronautical) Fixed Mobile Mobile satellite Fixed Mobile Earth exploration satellite Space operation Space research Mobile satellite Source: URSEC The 850MHz, 1900MHz and AWS bands are used for IMT services. Spectrum in those bands has been awarded in different processes, having the first one occurred in 2002, when, as per Resolution No. 490, the provision of mobile telecommunications services was licensed for a 20 years period. Subsequent assignations have taken place in 2004 and 2013, in both cases also for a 20 years period. So far 310MHz are identified for IMT, accounting only for 16%-23% of the total spectrum requirements as estimated by ITU (Report ITU-R M ). IMT assignments The following figure shows the spectrum holdings in Uruguay. Figure 46: IMT spectrum holdings in Uruguay 69

361 Digital dividend 850 MHz 900 MHz 1800 MHz 1900 MHz AWS 2,1 GHz 2.5GHz Total Last assignation took place in March 2013 Nonassigned * Extended AWS Source: URSEC, Dinatel * Antel holds 43% of the assigned spectrum, followed by Movistar (31%) and Claro (26%) Digital dividend and 2.5 GHz bands still to be assigned As shown, 270MHz have been assigned for the provision of IMT services. Whilst all the assignments to private operators have been awarded through auctions, Antel, the state-owned operator, has been directly awarded 115MHz. Taking into account that the licences have a period of 20 years, the expiration of those licenses will take place in 2022, 2024 and 2033 respectively, having no precedents for renewals. There are around 700 base stations deployed by each operator in the country, totalling approximately 2,000 nationwide. Co-location ratio is low (below 20% of the installed base stations). Finally, the economic compensation is an up-front payment at commencing the license with no additional fees or annual payments. The mobile market includes 5.3 million subscribers and the penetration reaches 155%. The following figure shows the evolution of the mobile market. Figure 47: Mobile market evolution of Uruguay Mobile services clients evolution Million Penetration % (subscribers / population) 3,0 3,5 4,1 4,4 4,8 5,0 5, Source: URSEC, World Bank 70

362 The mobile market has three operators Antel (state-owned company), Movistar and Claro, with one of them (Antel) concentrating more than 50% of the total market, followed by Movistar with 35% market share and Claro with 16% market share. Finally, as for new assignments, the Uruguayan government has already announced its decision to bid additional spectrum in the digital dividend band (700MHz). However, so far there is no detailed information about the plan. IMT roadmap and new bands under study Currently no new bands are being assessed by URSEC for the provision of IMT services. Current allocation The analysis of the L-band in Uruguay involves the MHz and MHz ranges. The first range (i.e MHz) has been allocated to radiolocation services. On the other hand, the frequencies MHz have been allocated to space operation, fixed and mobile (except aeronautical); the frequencies MHz has been allocated to fixed and mobile services, note that frequencies MHz are also allocated to broadcasting and broadcasting-satellite services and that the range going from MHz is also allocated to mobile satellite services. The following figure shows the services allocation in the L-band. Figure 48: L-band s services allocation A portion of the L-band ( MHz) is allocated to mobile services Aeronautical radionavigation Broadcasting Fixed Mobile Radiolocation Source: URSEC Earth exploration satellite Radioastronomy Space operation Space research Mobile satellite Mobile (except aeronautical) Broadcasting satellite 71

363 The figure shows that mobile services are already included in the current services allocation of the L- band ( MHz, MHz, MHz and MHz). The band MHz is allocated to the radiolocation service while the band MHz are reserved, as internationally recommended, to the following services: earth exploration-satellite, radioastronomy and space research, and therefore cannot be identified for the provision of IMT services. Assignments and actual use The utilization of the L-band is low, just used by private and government stakeholders for low capacity point-to-point links (lower than 2 Mbps or 4 Mbps) in the range MHz. Although no in-depth analysis of the actual usage of the band has been conducted it is estimated that the number of incumbents is less than ten. The assignments have occured during the 90s, under a FIFO (first in - first out) scheme and with no expiration date. Nevertheless, these assignments have been awarded on revocable basis, having the government the alternative to terminate these licenses at any time. Finally, as for the economic compensation, it consists of a monthly payment based on the total spectrum holding. The structure and calculation method for the fees date from 1993, having its prices been updated but not its structure. Potential migration No relevant barriers to migrate have been identified. Nevertheless, once considered the technological obsolescence of the installed equipment, it can be expected that the incumbents would voluntarily migrate to other technological solutions in a non-distant future. Position regarding future use of the L-band Neither operational nor economic barriers were identified to use the L-band for the provision of IMT services. 72

364 Argentina Frequency allocation and IMT services Current allocation The National Communications Commission (CNC per its acronym in Spanish) is responsible for the management and planning of the radio spectrum in Argentina. The following figures show the services allocation of the IMT bands already assigned. Figure 49: Frequency allocation in bands mostly used for IMT in Argentina 850 MHz band 1900 MHz band Land Mobile Fixed Mobile Fixed Conditional allocation Source: CNC Argentina has a total of 380MHz identified for IMT; based on the ITU s report on spectrum requirements (Report ITU-R M ), it represents only 19%-28% of the total spectrum the country should achieve by In the same page, according to GSMA 60, the current spectrum identified for IMT account for 23%-35% of the required spectrum by IMT assignments The licenses are structured on a regional basis. The regions considered for the spectrum assignment are: Region I (North of the counry), Region II (Extended Greater Buenos Aires) and Region III (South of the country). The following figure shows the spectrum holdings per operator per region. 60 Coleago study (2013) 73

365 Figure 50: IMT spectrum holdings in Argentina MHz 850MHZ 1900MHz Source: CNC Region Non assigned Total I II 30-12,50 7,50-50 III I II III The spectrum in the 850MHz and 1900MHz have mostly been assigned in the three regions The three incumbent private operators have similar spectrum holdings in the three regions The entire spectrum in the 850MHz and 1900MHz bands has been assigned, totalling 170MHz per region (50MHz in the 850 MHz and 120 MHz in the 1900 MHz band). The last spectrum auction in Argentina was held in the year In 2012, the government called for a new auction in order to award the remaining spectrum from Movistar acquisition of Movicom, as he had surpassed the 50 MHz per region spectrum cap. However, the auction was cancelled and the spectrum assigned to ARSAT. Today, Argentina is one of the countries with less spectrum assigned for IMT services. However, in mid-2014 the government announced a new auction to be held by the end of 2014, in order to promote the deployment of LTE technologies and increase the current available radio spectrum. The auction process has already started and the following frequencies are included: MHz (700MHz band) and MHz paired with MHz (AWS band), having the spectrum cap per region been set at 60 MHz for the advanced wireless services (LTE). Figure 51: Mobile market evolution and structure in Argentina Mobile telecommunication services evolution Mobile telecommunications market share (2013) Numbers of subscribers (million) Mobile penetration 111% 99% 122% 129% 137% 141% 151% 160% 140% 120% 100% 33% 34% % 60% 40% 20% 33% % The mobile penetration has climbed from 99% un 2007 to 151% in 2014 The Argentine market shows an evenly split market with three payers Source: Merrill Lynch Global Wireless Matrix 2Q

366 The mobile market has been growing at an 8.4% compound annual growth rate, having in 2008 surpassed the 100% penetration rate. Three players evenly share the market, having approximately 20 million subscribers each. IMT Roadmap and new bands under study Argentina shows a significant IMT spectrum shortfall, which is expected to be counteracted with the forthcoming 4G auction. In this context and considering that Argentina is one of the few countries with no spectrum assigned to 4G services, it is expected a focus on the 4G bands rather than already assessing new IMT bands. For the time being, the regulator has not formally provided definitive information with regards to the assessment of new bands for the provision of IMT services. Current allocation In Argentina, the L- ranges from the 1350MHz to 1525MHz. The CNC has allocated MHz, in accordance with international guidelines, to radiolocation services. The following figure shows the radio spectrum allocation for the L-band in Argentina. Figure 52: L-band s services allocation in Argentina MHz The L-band has no mobile services allocation Radiolocation Earth exploration satellite Radioastronomy Space research The MHz is almost fully allocated to fixed services Source: CNC Fixed Coditional allocation 75

367 Note that the L-band has no allocation to mobile services whilst the range MHz is almost fully allocated to fixed services. Assignments and actual use The National Frequency Allocation Chart 61 show that ranges MHz and MHz are used by digital multichannel systems 62 (MXD). The authorization to operate digital multichannel systems has no expiration date and remains in effect while complying with the provisions of the regulation. On the other hand, when assigned to providers of telecommunications services, these licenses have a 10 years period. Potential migrations There is no formal evidence of relevant barriers to migrate. Position regarding future use of the L-band The regulator has not yet formally submitted an answer. Dominican Republic Frequency allocation and IMT services Current allocation The Dominican Republic has 710MHz identified for IMT services on a primary basis. When compared to other LAC countries, Dominican Republican has a relatively high amount of spectrum identified for IMT. However, it is still below the ITU estimates for 2020; accounting for 36%- 53% of those estimates. The following figure shows the bands identified for IMT in the Dominican Republic. 61 As set on Disposición GI Nº 734/04 62 As per resolutions Nº 2860 CNT/92 and Nº 228 CNT/94 76

368 Figure 53: Bands identified for IMT in the Dominican Republic Band (MHz) Mobile spectrum 2002 (MHz) Mobile spectrum 2011 (MHz) Source: INDOTEL The 700MHz band will be effective once the switch over to Digital Terrestrial Television (DTT) is completed. IMT assignments The total spectrum currently assigned to mobile services operators in Dominican Republic are 450MHz: 50MHz in the 850MHz band, 40MHz in the 900MHz band, 40MHz in the 1700/2100MHz band, 120MHz in the 1900MHz band and 900MHz in the 2500MHz band. At the moment, there is no certain date for new assignments. However, there are still 50MHz in AWS band available for future assignments. As for economic compensation for the use of radio spectrum, as established by the General Telecommunications Law No , the use of spectrum will be taxed at an annual fee and its calculation methodology is set out in the General Regulations of Radio Spectrum Use, approved by Resolution No The mobile services market in Dominican Republic has grown at an 8.9% compound growth rate since 2003, reaching a penetration of almost 90% in

369 Figure 54: Evolution and structure of the mobile market in the Dom. Republic Mobile telecommunication services evolution Mobile telecommunications market share (2012) Numbers of suscriptors (milliom) 12 0 Mobile penetration 87% 89% 86% 87% 88% 74% 57% 49% 39% 8,6 8,9 8,8 8,9 9,2 7,2 5,5 4,6 23% 28% 3,6 2,1 2, % 80% 70% 60% 50% 40% 30% 20% 9% 35% 5% 51% Mobile subscribers have grown at a 8,9% compound annual growth rate in the period 2003 to 2013, climbing from 23% penetration in 2003 to 88% in 2013 The Dominican market has a dominant player, Claro, that accounts for more than half of the total subscribers Source: Indotel statistics, World Bank, Pyramid Research The Dominican Republic market shows a dominant player, Claro, accounting for more than 50% of the market. Orange has 35% of market share and Viva and Tricom share the remaining 14%. IMT roadmap and new bands under study The Dominican Republic has not defined yet its final position for the next World Radio communication Conference (WRC) to be held 2015, partly because the INDOTEL is still participating in the cycle of meetings of the Permanent Consultative Committee II (CCP.II) of CITEL, where the topic is being assessed. Current allocation For Dominican Republic it was considered the ranges MHz and MHz as scope of this study. The figure below shows the services allocation in the band under analysis. 78

370 Fixed Fixed SPACE OPERATION RADIOLOCATION SATELLITE FIXED RADIOLOCATION Radiolocation MOBILE RADIOLOCATION FIXED AERONAUTICAL RADIONAVIGATION MOBILE FIXED MOBILE Figure 55: L-band s services allocation in the Dominican Republic FIXED Most of the band, 98MHz, is allocated to fixed and mobile services on a primary basis (shared with other services) MOBILE The range MHz follows the international trend of radiolocation but shared with fixed services (on a secondary basis) MOBILE SATTELLITE Source: INDOTEL Most of the band under analysis, 98MHz in the range MHz, is allocated to fixed and mobile services on a primary basis (shared with other services).the band MHz follows the international trend of radiolocation and aeronautical radionavigation, but shared with fixed services (on a secondary basis) As for economic compensation for the use of spectrum in L-band, the terms and conditions are established in the General Telecommunications Law. The L-band is assigned following the same methodology as for the other services. Actual use and potential migrations A significant part of the L-band has been assigned and is being used; there are more than 130 licenses in the MHz range. There are 12 users including public agencies, religious organizations and other stakeholders, accounting for approximately 100 links. The most relevant use of this band is for aeronautic radionavigation and point-to-point links. Position regarding future use of the L-band There is no plan to modify the allocation to the L-band nor to migrate the current users out of this band. 79

371 VI. Bibliography - GSMA, Alternative use value of 700 MHz in Region 1, a report prepared by PLUM, (2013) - GSMA, Economic benefits from making the GHz band available for mobile broadband services in Western Europe, report prepared by AETHA, (2013) - GSMA, Revised spectrum forecasts using the new spectrum model, a report prepared by Coleago, (2013) - GSMA, Economic assessment of C-band re-allocation, a report prepared by Frontier Economics, (2013) - Varian Hal R., Microeconomic Analysis, third edition, (1992) - 3GPP, Active work programme Feature list of release 10, (2014) - 4G Americas, LTE Deployment Status, (2014) - CEPT, Report 188: Future Harmonised Use of MHz in CEPT, (2013) - Ericsson, Orange, Ericsson and Qualcomm have successfully completed the world's first live demonstration of supplemental downlink technology on L-band frequencies, (2013) - GSMA, Compatibility/sharing studies between IMT and radar in and MHz, (2014) - ITU, Radio Regulations Vol I, (2012) - ITU, Radio Regulations Vol III Resolutions, (2012) - ITU, Resolution 750 (Rev. WRC-12), (2012) - ITU, Recommendation ITU-R M , (2013) - ITU, Report ITU-R M : Future spectrum requirements estimate for terrestrial IMT, (2013) - Ericsson, Traffic exploration (Latam), (June 2014) - CISCO, VNI Mobile Forecast Highlights, ( ) - Marine Satellite Systems, Satellite Frequency Bands, (2010) - QUALCOMM, Supplemental Downlink, (2013) - Instituto Federal de telecomunicaciones (IFT), El espectro radioeléctrico en México. Estudio y acciones, (2013) - Instituto Federal de telecomunicaciones (IFT), National Frequency Allocation Chart, (2013) - The Organisation for Economic Co-operation and Development (OECD), Estudio sobre políticas y regulación de telecomunicaciones en México, (2012) 80

372 - P2!R_3541 presented by Mexico s delegation in reply to the decision CCP.II 162 (XXII-13). Request of information regarding current and future use of the MHz y MHZ bands, (2014) - Ministerio de tecnologías de la información y las comunicaciones (MINTIC), National Frequency Allocation Chart - Agencia Nacional del Espectro (ANE), Estudio técnico y económico de las bandas de frecuencias identificadas para IMT según el reglamento de radio de la UIT edición del 2012 y que aún no han sido usadas para implementación de este tipo de sistemas en Colombia, así como de las nuevas bandas de frecuencia que se encuentren en análisis por parte de los grupos de estudio de la UIT dentro del marco del punto 1.1 de la agenda para Conferencia Mundial de Radiocomunicaciones del 2015 (CMR-15), (2014) - Bank of America Merril Lynch, Global Wireless matrix 2Q Consejo Nacional de Telecomunicaciones (CONATEL), National Frequency Allocation Chart, (2012) - P2!R_3549 presented by Ecuador s delegation in reply to the decision CCP.II 162 (XXII-13). Request of information regarding current and future use of the MHz y MHZ bands, (2014) - Comisión de Regulación de Telecomunicaciones (CRT), Estudios de Elasticidades en servicios de telecomunicaciones, República de Colombia (2005) - The World Bank Latin American and the Caribbean Region, The Social Discount Rate: Estimates for Nine Latin American Countries (2008) - KPMG International, Latin America Indirect Tax Country Guide (2012) - ARHCIET y Deloitte, Tributación y Telecomunicaciones en Latinoamérica (2013) - Telecom Advisory Services, Dividendo Digital (banda de 700MHz) en América Latina, Raúl Katz y Ernesto Flores-Roux (2011) - Katz, R. and Berry, T., Driving demand for broadband networks and services (2014) - GSMA Intelligence, Reportes de información por país (2014) - 4G Américas, Análisis de las Recomendaciones de la UIT sobre el espectro en la Región América Latina-Agosto 2013 (2013) - ITU, Impact of Broadband on the economy - Broadband series (April 2012) - United Nations, Department of Economic and Social Affairs, Population Division (2013) - Nokia Siemens Networks, Mobile broadband with HSPA and LTE - capacity and cost aspects. - 3GPP, documento TR (UTRA Baseline) - Regulators websites: SUPERTEL, SENATEL, ANATEL, URSEC, CNC, CRC, COFETEL (IFT), ACMA, OPTA, INODETEL, FCC 81

373 - ANATEL, Annual Report, (2013) - GSMA, License Renewal in Latin América, (2014) 82

374 MOBILE ECONOMY LATIN AMERICA Mobile addressing the economic and social challenges across the region Unique subscribers and SIM connections UNIQUE MOBILE SUBSCRIBERS M 52% PENETRATION RATE 390M 59% PENETRATION RATE Delivering digital inclusion to the still unconnected populations Mobile internet subscriber penetration: 33% 2013, 48% 2020 Delivering financial inclusion to the still unbanked populations 36 live services and 19 planned as of June 2014 Delivering innovative new services Number of M2M connections to grow at a CAGR of 25% per annum out to 2020 CONNECTIONS* M 890M 112% PENETRATION RATE Technology shift fuelling data traffic growth across Latin America *Excluding M2M 134% PENETRATION RATE Mobile ecosystem contribution to GDP 2013 US$242BUS$275B % GDP in % GDP in G/4G connections to increase from a third of total in 2013 to almost 80% BY 2020 Slowing subscriber growth, regulation and competition impacting revenue outlook Operator revenues grew at a CAGR of over 7% By 2020 there will be 605m SMARTPHONE CONNECTIONS up from 154 million at the end of 2013 Data traffic in Latin America to grow at a CAGR of 66% out to 2018 Revenues forecast to slow to a CAGR of just over 2% Public funding ecosystem US$41B 2013 US$53B 2020 Mobile ecosystem contribution to public funding in Latin America before regulatory fees Employment Jobs directly supported by mobile 1M M Plus an additional 1.2M indirect jobs supported in

375 e-waste in Latin America Electronic waste growth in Latin America KT = 1Kiloton = 1,000 Tons Source: UNU 3,511KT WEEE in Latin America, selected countries 4,226KT 4,968KT KG is the average amount of e-waste produced by every person on the planet Source: StEP Initiative A ton of gold ore contains 5g of gold whereas a ton of mobile phones contains 400g OF GOLD What s in a smartphone 45% plastic 10% ceramic 20% copper 20% gold, aluminum and other metals 5% non-metals Source: TIM Recarregue o Planeta US$21M a year in gold, silver, and other metals can be extracted from WEEE Source: e-waste Academy (EWAM) 2015 El Salvador Argentina Panama Mexico Brazil Ecuador Uruguay Chile ,000 1,200 1,400 1,600 1,800 KILOTONS Results of the recycling process BATTERIES Composition: cobalt / nickel copper Batteries, magnets, stainless steel, metal oxides, ink pigment, and salts. COMPONENTS Composition: gold / palladium copper Jewelry, musical instruments, electronic equipment PLASTIC Composition: plastic Traffic cones, chairs, tyres, pet food containers and other everyday objects. In 2015, Latin America will produce From 2012 to 2015 Latin America WEEE will increase Latin America mobile operators willingly develop: 3-5 years 8.6% 17.5% is the average lifespan of a smartphone Source: MIT/StEP Initiative OF THE GLOBAL WEEE Reverse logistic schemes Awareness campaigns Recycling programs Standards adoption

376 The Mobile Economy Latin America 2014 Copyright 2014 GSM Association

377 The GSMA represents the interests of mobile operators worldwide. Spanning more than 220 countries, the GSMA unites nearly 800 of the world s mobile operators with more than 230 companies in the broader mobile ecosystem, including handset makers, software companies, equipment providers and Internet companies, as well as organisations in industry sectors such as financial services, healthcare, media, transport and utilities. The GSMA also produces industry-leading events such as the Mobile World Congress and Mobile Asia Expo. For more information, please visit the GSMA corporate website at or Mobile World Live, the online portal for the mobile communications industry, at GSMA Latin America is the branch of the GSMA in the region. For more information in English, Spanish and Portuguese, please visit This report is authored by GSMA Intelligence, the definitive source of global mobile operator data, analysis and forecasts; and a publisher of authoritative industry reports and research. Our data covers every operator group, network and MVNO in every country worldwide from Afghanistan to Zimbabwe. It is the most accurate and complete set of industry metrics available, comprising tens of millions of individual data points, updated daily. GSMA Intelligence is relied on by leading operators, vendors, regulators, financial institutions and third-party industry players, to support strategic decision-making and longterm investment planning. The data is used as an industry reference point and is frequently cited by the media and by the industry itself. Our team of analysts and experts produce regular thought-leading research reports across a range of industry topics. /

378 THE MOBILE ECONOMY 2014 LATIN AMERICA Contents Executive summary 2 1 The mobile industry in Latin America Strong growth in subscribers and connections A very diverse region Shift to 3G and 4G now accelerating The impact of competition on pricing Revenue growth will continue to slow Ongoing operator investments to support growth 27 2 Delivering digital inclusion in Latin America Network infrastructure and policy Improving the affordability of mobile internet access Consumer barriers: improving digital literacy and awareness Improving the availability of local content 35 3 The social and economic impact of mobile in Latin America 3.1 Mobile makes a major contribution to economic growth in the region Mobile facilitating entrepreneurship and commerce in Latin America M2M services are growing rapidly across the region Delivering access to basic services across Latin America ewaste Delivering financial inclusion across Latin America Mobile policy and the role of the industry Spectrum: A key enabler to promote investment and foster 63 innovation 4.2 Pressures on quality of service: The need to address its causes Removing barriers for infrastructure deployment remains critical in 71 Latin America 4.4 Reducing the taxation burden on mobile services in Latin America The mobile industry and the internet in Latin America Social responsibility: industry initiatives to protect the Latin American mobile user 78 1

379 THE MOBILE ECONOMY 2014 LATIN AMERICA Executive Summary Whilst Latin America is in the midst of an ongoing slowdown in both unique subscriber and revenue growth rates, the region is now seeing an accelerating migration to higher speed networks and smartphone adoption. This is driving strong data traffic growth and incremental revenues for operators, which in turn will help fund the major investments required to further build out both 3G and 4G networks. Mobile is already making an important contribution to economic growth, as well as addressing a range of social challenges, such as closing the digital divide and bringing financial inclusion to previously underserved populations. However, the rapid adoption of more advanced devices and higher speed network coverage means that mobile has the potential to play a much more significant role in the future. Realising the full transformative potential of mobile will require more collaboration between all mobile ecosystem players, as well as a more supportive and transparent regulatory regime that recognises the potential for a true partnership between the industry players and other stakeholders, including governments and regulators, across the region. The Latin America mobile market is now the fourth-largest globally, with almost 326 million unique subscribers and 718 million connections 1 as of September of A little over half of the population in the region have now subscribed to a mobile service, a figure that is expected to reach almost 60% by 2020, broadly in line with the global average. Latin America is now seeing an acceleration in the growth of data traffic, fuelled by the rapid technological shift to higher speed connections that is underway across the region. Whilst over twothirds of the connections were 2G at the end of 2013, by 2020 almost 80% will be running on higher speed 3G and increasingly 4G networks. The increasing proportion of higher speed connections largely reflects the accelerating rate of smartphone adoption, with Latin America seeing one of the highest growth rates over recent years, with 200 million handsets as of September 2014 (almost 30% of the total connection base). The region is forecast to have a total of 605 million smartphone connections in 2020, equivalent to almost 70% of the total connections base. Latin America will then have the second-highest installed base of smartphones in the world. Data volumes and revenues are growing strongly, fuelled by the accelerating adoption of smartphones and mobile broadband services. Forecasts from Cisco suggest that data traffic in Latin America will grow at a compound annual growth rate (CAGR) of 66% out to 2018, well ahead of the growth forecast for more developed markets. 1. Includes cellular M2M connections 2 Executive Summary

380 THE MOBILE ECONOMY 2014 LATIN AMERICA However, operators across the region also face a number of challenges. Whilst operator revenues grew at a healthy CAGR of over 7% over the period , going forward this is expected to slow to a little over 2% out to The slowdown reflects a number of factors, including decelerating subscriber growth in the major markets. Regulatory intervention and competition are also important factors, with the region set to see a wave of mobile virtual network operator (MVNO) launches, especially in the major markets of Brazil and Mexico. Profitability is also under pressure for many operators in Latin America as a result of competitive and regulatory factors, with margins in aggregate now the lowest of any developing region. This makes it more challenging for operators to fund investments at a time when there is ongoing pressure to improve networks to alleviate capacity constraints (and address quality of service issues) and launch higher-speed networks. Operator capital expenditure totalled over US$96 billion in the six years to Investment levels are likely to increase substantially over the coming years to accommodate data traffic growth, particularly as LTE deployments gather pace, with almost US$193 billion set to be invested over the seven-year period from 2014 to However, delivering these levels of investment will require sustainable business models for the operators. The mobile ecosystem is a key source of economic growth and job creation in the region. In 2013, the mobile industry contributed 4.1% to the total gross domestic product (GDP) of Latin America. This figure captures the direct contribution of the mobile ecosystem, as well as indirect contributions in the form of increased business productivity and economic activity generated in the broader economy. The economic contribution of Latin America s mobile industry is expected to continue to increase at a faster rate than the rest of the economy, reaching 4.5% of GDP by In 2013, the mobile ecosystem directly provided employment to nearly one million people, with a further one million indirect jobs supported in the rest of the economy. The number of jobs directly supported within the industry will continue to experience material growth in the future, reaching close to 1.3 million people by The industry also makes a very large contribution to the funding of the public sector, totalling US$41 billion in There are also further payments in the form of licence and regulatory fees and spectrum auctions. Mobile networks have already played an important role in bridging the digital divide and bringing internet access to previously unconnected populations across Latin America. As of September 2014, there were 216 million individuals using mobile devices to access the internet, equivalent to an overall penetration rate of around 35%. By 2020, this figure is forecast to be just under 50% of the population, indicating that an additional 105 million people will gain mobile internet access over the period for a total of 321 million subscribers. Despite the progress to date, many people across the region still lack internet access. Mobile operators and the broader industry are making an important contribution to boosting digital inclusion. For example, innovative new tarrif plans and lower priced handsets are increasing the affordability of mobile services. Governments across the region have recognised the importance of widening access to the internet and broadband connectivity in particular. There is though still more to be done by governments and regulators in addressing these goals. Regulations such as those around quality of service or electromagnetic field (EMF) regulations can hinder network deployments. In addition, ensuring the timely release of Digital Dividend spectrum is an important factor in enabling cost effective network coverage. The taxation of mobile services is also relatively high in a number of markets in the region and can add further to the affordability barrier. The increasing uptake of smartphones and other advanced devices, as well as the improving higher speed network coverage, are allowing the development of a range of new innovative new products and services. The new services can benefit individuals, businesses and governments. There were 16 million cellular machine-to-machine (M2M) connections in Latin America as of September 2014, with growth expected at a CAGR of 25% for the period However, there is potentially a significant upside to these forecasts if a number of growth factors are successfully enabled by industry players and governments in the short to medium term. Executive Summary 3

381 THE MOBILE ECONOMY 2014 LATIN AMERICA Mobile financial services have seen less adoption in Latin America than other developing regions, though with around 60% of the population remaining unbanked, there is a clear opportunity to bring financial inclusion to the currently underserved. Considerable progress has been made in elaborating the necessary legal and regulatory frameworks with, for example, changes in several markets allowing non-banks to offer e-money services. This is translating into increasing interest and investment by the private sector, and a growing number of deployments, suggesting that mobile money services in Latin America are now at a turning point. There were 36 live deployments and a further 19 planned as of June 2014, with new business models and partnerships emerging. Mobile can play a role in addressing some of the region s social, economic and public services challenges, including high rates of urbanisation and an increasing working age population. There is a growing number of new mobile services and applications, many of them developed by small and medium sized enterprises. There are areas of opportunity for these companies to develop local content and applications that are tailored to local needs. There is no doubt that the market and regulatory landscape is changing in Latin America. New telecom laws have been passed in Mexico, as well as the Internet Civil Rights Framework Law in Brazil, while new bills are currently under discussion in Argentina, Ecuador and Perú. All of these developments are likely to significantly transform mobile market conditions. Consolidation is also now a matter for discussion in Brazil, Mexico and Chile. Data-hungry consumers are putting pressure on networks and impacting their quality of service, while regulators are sometimes addressing these demands with a political rather than technical approach. Regulatory intervention has increased across Latin America, creating questions about the future of the industry. The convergence of mobile with the IP world is also creating challenges for the sustainability of operators business models. With reduced margins, increased competition from adjacent industries and higher regulatory pressure, sustaining growth will very much depend on having the right regulatory environment. Having a transparent, predictable, consultative and wellaligned regulatory regime will be key for promoting the right incentives that would unleash future investments in mobile telecom networks and services. A clear roadmap of spectrum allocation, as well as the removal of municipal obstacles for antenna deployment, will be key to satisfy the increasing demand of end users for mobile broadband. Despite these challenges, there are opportunities ahead for mobile operators to expand their businesses and to further increase their social and economic contribution. Latin American operators have demonstrated their commitment to the markets they operate in by sustaining investment and lately by implementing new joint consumer-centric initiatives under the GSMA s We Care campaign, which is focused on addressing social needs. 4 Executive Summary

382 THE MOBILE ECONOMY 2014 LATIN AMERICA The region is forecast to have a total of 605 million smartphone connections in 2020, equivalent to almost 70% of the total connections base. Latin America will then have the second-highest installed base of smartphones in the world. Executive Summary 5

383 MOBILE ECONOMY LATIN AMERICA Unique subscribers and SIM connections UNIQUE MOBILE SUBSCRIBERS 390M M 52% PENETRATION RATE 2020 CONNECTIONS* M 890M 112% PENETRATION RATE 59% PENETRATION RATE *Excluding M2M 134% PENETRATION RATE Technology shift fuelling data traffic growth across Latin America 3G/4G connections to increase from a third of total in 2013 to almost 80% BY 2020 By 2020 there will be 605m SMARTPHONE CONNECTIONS up from 154 million at the end of 2013 Data traffic in Latin America to grow at a CAGR of 66% out to 2018 Slowing subscriber growth, regulation and competition impacting revenue outlook Operator revenues grew at a CAGR of over 7% Revenues forecast to slow to a CAGR of just over 2%

384 Mobile addressing the economic and social challenges across the region Delivering digital inclusion to the still unconnected populations Mobile internet subscriber penetration: 33% 2013, 48% 2020 Delivering financial inclusion to the still unbanked populations 36 live services and 19 planned as of June 2014 Delivering innovative new services Number of M2M connections to grow at a CAGR of 25% per annum out to 2020 Mobile ecosystem contribution to GDP 2013 US$242BUS$275B % GDP in % GDP in 2020 Public funding ecosystem Employment Jobs directly supported by mobile US$41B 2013 US$53B M M 2020 Mobile ecosystem contribution to public funding in Latin America before regulatory fees Plus an additional 1.2M indirect jobs supported in 2013

385 THE MOBILE ECONOMY 2014 LATIN AMERICA 1 The mobile industry in Latin America 1.1 Strong growth in subscribers and connections The Latin American mobile market has delivered strong growth in recent years in terms of both unique subscribers and connections (i.e. SIM cards). It is now the fourth-largest globally, with almost 326 million unique subscribers and 702 million connections 2 by September The subscriber base grew at an average rate of 5.5% over the five years to 2013, and is forecast to grow at a little under 3% per year between 2013 and Growth in connections over the period to 2013 stood at around 8.8% per annum, and is forecast to slow to an average of just over 3.7% in the seven years to Source: GSMA Intelligence Unique subscribers growth rates by region 16.4% 11.2% 5.4% 4.2% 2.4% 1.2% 0.9% 5.5% 8.9% 2.8% 3.2% 2.5% 0.9% 7.0% APAC CIS EUROPE LATAM MENA N AMERICA SSA 2008/ /20 2. Note that this figure excludes the number of M2M connections, which are disclosed separately 8 The mobile industry in Latin America

386 THE MOBILE ECONOMY 2014 LATIN AMERICA The growth rate of both subscribers and connections in Latin America has been slowing for a number of years. However, growth dropped more sharply in 2012 and 2013, after a brief recovery post the global financial crisis. Latin America is now in the mid-range of regional penetration rates and above the global average, in terms of both unique subscribers and connections. The unique subscriber penetration rate was 52% as of September 2014, and 112% for SIM connections. Source: GSMA Intelligence Penetration rates by region September % 125.6% 78.6% 97.4% 70.0% 68.3% 112.3% 105.8% 52.1% 50.3% 89.6% 44.2% 38.3% 70.7% EUROPE N AMERICA CIS LATAM MENA APAC SSA Subscribers Connections The mobile industry in Latin America 9

387 THE MOBILE ECONOMY 2014 LATIN AMERICA The high penetration rate of connections has been driven in large part by users acquisition of additional SIMs to take advantage of competitive price offers, resulting in an average of close to two active SIMs per user across the region. This ratio of active SIM connections to unique subscribers has increased steadily over recent years, from 1.8 SIMs per user at the end of This trend reflects both ongoing price competition in the mass market, as well as the increasing trend towards multi-device ownership in the higher income population segments. Source: GSMA Intelligence Unique subscriber trends Latin America 51.6% % % % % % % % 390 Subscriber penetration (%) UNIQUE SUBSCRIBERS (m) Subscriber growth is projected grow at a steady rate over the period out Subscriber penetration will be close to 60% by 2020, although still well below the 70-80% level at which growth tends to stall in more mature markets. This highlights the potential for sustained growth beyond 2020, as well as the need for a supportive regulatory regime to help connect the still unconnected populations across the region. Rising penetration rates in the future will be driven by higher levels of consumer income and more affordable mobile services. Both ongoing price declines for basic mobile services as well as cheaper devices, including lower-specification smartphones, will be important factors in improving affordability. 10 The mobile industry in Latin America

388 THE MOBILE ECONOMY 2014 LATIN AMERICA 1.2 A very diverse region Latin America is very diverse in terms of economic and social development, and equally in terms of mobile penetration (both unique subscribers and connections). Connection penetration rates range from a low of 73% in Haiti to a high of 157% in Costa Rica. The overall region s penetration rate stood at 112% as of September 2014, well ahead of the global average figure of 96%. Subscriber penetration rates in the larger markets in Latin America range from a low of 37% in Mexico to a high of 77% in Costa Rica. There is no single driver of the variation in penetration rates with, for example, differences in GDP per capita playing only a limited role. Source: GSMA Intelligence Unique subscriber and connection penetration (Selected markets, September 2014) 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% Mexico Haiti Peru Bolivia Colombia Venezuela Paraguay Honduras Brazil Guatemala Nicaragua Ecuador Dom. Republic Jamaica El Salvador Chile Uruguay Panama Argentina Costa Rica Unique subscriber penetration Connection penetration The mobile industry in Latin America 11

389 THE MOBILE ECONOMY 2014 LATIN AMERICA The Latin American mobile market is dominated by Brazil, with 114 million unique subscribers by September 2014, accounting for over a third of the region s total base. Brazil currently ranks as the fifth-largest market globally in terms of subscribers, and is forecast to overtake Japan and become the fourth-largest by the end of The five largest countries in the region have over 230 million subscribers between them, equal to more than 70% of Latin America s total. Source: GSMA Intelligence Five largest markets by unique subscribers September 2014 (M) % The five largest countries in the region have over 230 million subscribers between them, equal to more than 70% of Latin America s total Brazil Mexico Argentina Colombia Venezuela Rest of Latin America 12 The mobile industry in Latin America

390 THE MOBILE ECONOMY 2014 LATIN AMERICA 1.3 Shift to 3G and 4G now accelerating Latin America is now seeing an acceleration in the take-up of mobile broadband and the growth of data traffic. This has been fuelled by a rapid technological shift to higher speed connections that is underway across the region. At the end of 2012, 2G services still accounted for 78% of total connections, but by September 2014 this had fallen to 60%. Correspondingly, 3G connections rose from 22% at the end of 2012 to 39% by the third quarter of This is higher than the global average of 32% and higher than the developing market average of 27%. By contrast, 4G is still in its relative infancy, although adoption is expected to gather pace as network launches take place across the region and ongoing deployments in most of the leading markets are completed. As of September 2014, only just over 1% of connections were 4G, broadly in line with the developing market average but compared to 35% in Northern America, the leading LTE market. However, the number of 4G connections is expected to grow at an average rate of 85% per annum in the seven years to It is also worth noting that 2G will still represent over a fifth of connections by this date. Source: GSMA Intelligence Total connections by technology generation (M) 1, G 3G 4G CAGR -12.2% CAGR +11.1% CAGR +84.9% The mobile industry in Latin America 13

391 THE MOBILE ECONOMY 2014 LATIN AMERICA There is also a wide variation in the pace of migration to 3G connections across Latin America. Amongst the major markets, Colombia has more than four-fifths of 2G connections, while Brazil has the highest proportion of 3G, at around half of the total. 3G penetration has surged in Brazil in recent years, with the number of connections rising by 57% in the year to June 2014, reaching a total of 133 million. This growth has been fuelled by the rapid roll-out of the country s 3G networks, which reportedly provided 92% population coverage by July , and by accelerated take-up of smartphones. As a result of these developments, over the course of 2013, Brazil overtook Venezuela which previously had the highest proportion of 3G connections. Brazil also has the largest number of 4G connections in the region, with 4.6 million as at September At the end of August 2014, LTE services had been launched by 44 networks in 18 countries throughout the region. 4G networks have yet to go live in Argentina, although the spectrum auction was completed in of October Operators will increasingly invest in 4G networks over the coming years as spectrum is increasingly made available to support these deployments The mobile industry in Latin America

392 THE MOBILE ECONOMY 2014 LATIN AMERICA Source: 4G Americas LTE in Latin America* & the Caribbean 44 Networks / 18 Countries as of 29 Aug 2014 Mexico Colombia Movistar Nextel Telcel Avantel Claro Movistar Tigo - UNE 1700/2100 AWS 1700/2100 AWS 1700/2100 AWS 2100 AWS 2500/2690 MHz 1700/2100 AWS 1700/2100 AWS 2500/2690 MHz Bolivia Entel, Tigo 700 MHz Brazil Chile Claro Nextel Oi ON* Sky Telecom* TIM VIVO Claro Entel Movistar 2500/2690 MHz 1800 MHz 2500/2690 MHz 2500/2690 MHz 2500/2690 MHz 2500/2690 MHz 2500/2690 MHz 2500/2690 MHz 2500/2690 MHz 2500/2690 MHz Ecuador CNT Mobile 1700/2100 AWS Paraguay Peru Uruguay Personal VOX Movistar Claro Antel Claro 1900 MHz 1700/2100 AWS 1700/2100 AWS 1700/2100 AWS 1700/2100 AWS Venezuela Digitel 1800 MHz Antigua & Barbuda Digicel* 700 MHz Aruba SETAR 1800 MHz Bahamas BTC 700 MHz Cayman Islands Costa Rica Dominican Republic Puerto Rico Virgin Islands C&W Lime Digicel* ICE/Kolbi Claro Movistar Claro Orange Dominicana Tricom AT&T Claro Open Mobile Sprint T-Mobile AT&T Sprint 700 MHz 700 MHz 2600 MHz 1800 MHz 1800 MHz 1700/2100 AWS 1800 MHz 1900 MHz 1700/2100 AWS 700 MHz 700 MHz 850/1900 MHz 1700/2100 AWS 1700/2100 AWS 1900 MHz * Indicates TDD networks The mobile industry in Latin America 15

393 THE MOBILE ECONOMY 2014 LATIN AMERICA The internet is mainly mobile Mobile networks and services are increasingly becoming the main method of accessing the internet across Latin America, with the number of mobile broadband connections surpassing the number of fixed broadband as early as As a result of the rapid growth experienced in the past two or three years, there are more mobile broadband connections than fixed broadband subscriptions in all of the leading markets in the region, and in some instances considerably more. Source: GSMA Intelligence; ITU Mobile and fixed broadband connections in selected markets (2013, m) Brazil Mexico Argentina Chile Venezuela Peru Colombia Mobile broadband Fixed broadband One of the reasons for this mobile dominance is the lack of fixed broadband infrastructure in many countries and particularly in rural areas. Another important reason is the historical absence of a large base of fixedtelephony connections that can be adapted to support broadband service delivery, as has been the case in many developed markets. Going forward, the dominant role of mobile networks is likely to increase, as it will not be cost-effective to deploy fibre access or other fixed broadband infrastructure into homes and businesses in more rural areas. 16 The mobile industry in Latin America

394 THE MOBILE ECONOMY 2014 LATIN AMERICA Rising smartphone adoption across the region Between 2010 and 2013, Latin America recorded one of the highest growth in smartphone connections of any region, increasing at a CAGR of 77%. The smartphone connection base stood at 200 million at September 2014, representing almost 30% of total connections and putting the region in third position worldwide. Growth rates are expected to slow over the period out to 2020, but with a forecast total of 605 million smartphone connections (approximately 70% of the total connections base), Latin America will have the second-highest installed base of smartphones in the world. The dominant smartphone market in the region, as in many other areas of the region s mobile industry, is Brazil, which had 89.5 million connections at September Brazil is the fifth-largest global smartphone market after China, the United States, India and Indonesia. Source: GSMA Intelligence Smartphone percentage of total connections 68% 31% 3% The mobile industry in Latin America 17

395 THE MOBILE ECONOMY 2014 LATIN AMERICA Source: GSMA Intelligence Smartphone adoption rates (selected markets) 73.2% 72.1% 72.2% 68.0% 69.0% 61.9% 57.4% 45.1% % 32.4% 28.6% 26.9% 17.9% 17.2% Venezuela Argentina Brazil Latin America Colombia Mexico Peru Q Q Operators across the region are reporting increasing smartphone sales. In the second quarter of 2014, TIM Brasil reported that smartphones accounted for 71% of total handset sales, compared with 57% in the same quarter of Personal in Argentina reported that smartphones were over 80% of total sales in the second quarter of 2014, with them accounting for 30% of the company s total connection base. Growth in the next few years will be fuelled by improved affordability of smartphones and associated 3G and 4G services. This reduction in costs will in large measure be achieved through the growing availability of lower-cost smartphones. These will include operator-branded devices, entrylevel phones built on the Windows or Android OS, and those using alternative, reduced-spec OS optimised for lower-cost hardware. A leading innovator in this area is Mozilla, which presented a US$25 smartphone at the 2014 GSMA Mobile World Congress. Telefónica has started to roll out Mozilla Firefox OS-based devices across its Latin American subsidiaries, including the Alcatel One Touch Fire, launched in Chile in February 2014 and priced at 29,990 pesos (around US$52) and the ZTE Open II introduced in Panama in August Along with falling device prices, the cost per megabyte of mobile data will also decrease as 3G and 4G networks deliver more extensive population coverage, and effects of enhanced supply, scale and competition take place. As smartphones are the main devices through which consumers in the region access the internet, there is a close connection between users take-up of mobile broadband services and the acquisition of their first smartphone. There is a mutually reinforcing volume effect as growing smartphone purchases drive data consumption, which then sparks the development of new content, apps and services, which in turn fuels the appetite for smartphones. Growth in smartphone connections is not, however, expected to drive a migration from prepaid to contract plans. The share of prepaid connections is projected to remain high through 2020, at around 75% of the total. This means that smartphones will be taken up primarily as part of low-cost prepaid plans, including so-called sachet data tariffs that allow users to consume data occasionally and on a pay-as-you-go basis. The dominance of prepaid across most of the region further emphasises the important role that lower-cost smartphones will play in driving take-up, given the absence of device subsidies in prepaid plans. 18 The mobile industry in Latin America

396 THE MOBILE ECONOMY 2014 LATIN AMERICA Data traffic is surging Data volumes and revenues are growing strongly across the region, fuelled by the changing device mix with increasing smartphone adoption and mobile broadband services. Data traffic in Latin America is forecast to grow at a compound annual rate of 66% out to 2018, well ahead of the growth forecast for more developed markets such as Europe and North America. Source: Cisco VNI Mobile Forecast, 2014 Latin American mobile data traffic (TB per month) 1,158, ,313 91, , , , Operators across the region are reporting increasing data traffic growth. Vivo in Brazil reported an 86% year-on-year increase in data traffic in the second quarter of 2014, while Telefónica also saw strong growth in a number of its other operations in the region. This included annual growth rates of 83% in Chile and 66% in Peru. The growth in data traffic is generating an increase in the proportion of revenues deriving from data versus voice services. The extent of this trend varies across the region, with operators also following differing strategies to monetise the growth in data volumes. Claro in Chile recorded a 24% increase in data revenues in the second half of 2014, with it now representing over a third of total revenues. Claro s combined operations in Argentina, Uruguay and Paraguay saw stronger data revenue growth of 39%, with it accounting for 46% of total revenues across the three assets. The recent World Cup was a major catalyst for data traffic in Brazil, as well as for network investments that are continuing to benefit the country and help drive the ongoing technology shift and traffic growth. Operators in Brazil invested to increase both 3G and 4G coverage in the main stadiums during the World Cup. Data from Telebrasil highlighted that the operators cooperated to deploy 4,738 antennas at the event stadiums, with investments totalling R$226 million on shared infrastructure. In addition, Telebrasil estimated that the operators invested a further R$1.3 billion in infrastructure across the host cities, in both new antennas and fibre backbone capacity. Over the course of the tournament s 64 matches, there were 4.5 million phone calls made and 48.5 million picture messages sent, corresponding to around 27TB of data. The mobile industry in Latin America 19

397 THE MOBILE ECONOMY 2014 LATIN AMERICA Growing impact of online messaging in Latin America Consumers across the region are increasingly using online messaging service (OMS) providers, reflecting both the increased adoption of smartphones and more ubiquitous mobile broadband networks. The popularity of these IP-based messaging services is already beginning to impact the operating performance of the operators. For example, the decline in Movistar Mexico s SMS revenues in recent quarters is in part a by-product of the popularity of these OMS providers, including WhatsApp. Whilst WhatsApp was one of the first providers to establish itself in the region, a number of other service providers are also now gaining scale. For example, it has been estimated that WhatsApp had been installed on 88% of active smartphones in Mexico by end Similarly, Viber had reportedly signed up 17 million users in Brazil by August 2014, making it the second-most popular online messaging platform in the country after WhatsApp 5. Latin American operators have adopted a number of strategies to counter the threat posed by these OMS providers, as well as to capitalise on the opportunity they present: 1. Incorporating OMS providers into tariff plans: this includes Movistar Mexico s contract and prepaid plans which offer unlimited WhatsApp usage. Other operators have introduced similar measures. For instance, in August 2014, Telcel in Mexico announced it would provide unlimited WhatsApp usage on prepaid top-ups of 100 pesos (around US$7.60) or more, as well as unlimited access to Facebook and Twitter in the case of top-ups of 200 pesos or more. In these cases operators are using the popularity of these services to create an incentive for users to subscribe to particular contract plans, or to stimulate and monetise increased data traffic. 2. Introducing their own social messaging services: another approach is for operators to launch their own social messaging apps to compete with the OMS providers head on. Both Telefónica and TIM (Brazil) have opted for this solution. Telefónica launched the Tuenti service in Mexico in June 2013: a multi-platform website and app combining IM, VoIP and social networking features. At the time of the launch, the operator claimed Tuenti was the most popular social-networking service in Spain, with 15 million registered web users and six million mobile users. Telefónica Argentina launched TU Go in early 2014, another messaging app that had seen some success for O2 in the UK. Similarly, in April 2014, TIM announced it would launch Blah : another subscription-free, carrier- and platformindependent social messaging platform, described as a cross between WhatsApp and Viber. However, as at September 2014, Blah was still in the beta testing phase. There have been some limited attempts in the region to block access to OMS providers, although these have tended to be rejected by local regulators. Regardless of the regulatory considerations, blocking access to certain applications is unlikely to be a sustainable long-term strategy for operators. Operators in some markets are also partnering with social media providers such as Facebook. Millicom in Paraguay announced in December of 2013 a time limited offer giving its subscriber free access to Facebook, whilst Facebook has also made content available in the local Guarani language spoken by the majority of the country s population. 4. Competitive Intelligence Unit (CIU), cited in jsp?id = &idioma=i&tipocontenido=detalle&pagina=content&idcontenido=653302&tipodocumento= The mobile industry in Latin America

398 THE MOBILE ECONOMY 2014 LATIN AMERICA Consumers across the region are increasingly using online messaging service providers, reflecting both the increased adoption of smartphones and more ubiquitous mobile broadband networks. The mobile industry in Latin America 21

399 THE MOBILE ECONOMY 2014 LATIN AMERICA 1.4 The impact of competition on pricing Competition has intensified in recent years, with new players entering the market and new networks being launched. This has contributed to a downward pressure on pricing, with the average effective price per voice minute across the region as a whole falling from US$0.11 in the fourth quarter of 2008 to US$0.06 in the same period in Against this backdrop of falling prices and increased competition, there has been a small but steady decline in average revenue per connection (ARPU) for the region as a whole. Reported ARPU figures are distorted by the high levels of multi-sim ownership. Nevertheless, the total ARPU for Latin America for the second quarter of 2014 stood at US$10.7. This remains well ahead of the developing market average of US$6.8, although it is down from the average of US$11.7 recorded for the final quarter of The ARPU figures for the region overall hide more varied trends at the country level. Brazil has seen some of the steepest declines in ARPU over the period, driven by competition and efforts to improve the accessibility of mobile services. In Mexico, sharp reductions in prices per minute have led to increased usage, but this has been insufficient to prevent an overall fall in ARPU. Recent research by GSMA Intelligence highlighted the shortcoming of traditional ARPU metrics that look at revenue per connection, suggesting instead a focus on average revenue per subscriber (ARPS) as giving a better indication of real levels of consumer spending 6. The ARPS in Latin America has increased over recent years, despite the declines in unit based pricing, highlighting the growing trend towards multiple device ownership. Source: GSMA Intelligence ARPS trends in Latin America (Per month US$, annual average) https://gsmaintelligence.com/analysis/2014/10/evaluating-consumer-spending-the-need-for-a-revised-arpu-metric/448/ 22 The mobile industry in Latin America

400 THE MOBILE ECONOMY 2014 LATIN AMERICA Increased regulatory intervention Greater competition across the region has in part been a consequence of a number of regulatory measures. These include the reservation of frequency blocks for, or preferential treatment of, new entrants and state-owned operators in recent and upcoming spectrum-allocation processes, as well as the obligation to accept MVNOs on incumbent networks in several countries. At the time of writing, auctions of 3G and/or 4G spectrum were expected to be held in Argentina and Venezuela by the end of 2014, and new entrants were entitled to bid in both of these auctions. Spectrum set-asides for new entrants have not always proved to be a successful strategy. In 2012, having cancelled a 3G spectrum tender, the Argentinian government instead assigned the spectrum exclusively to the state-owned satellite operator ArSat. However, the government recently changed the approach and included the spectrum allocated to ArSat in a new auction. Similarly, in Colombia, state-owned fixed operator UNE was awarded 50MHz of LTE spectrum in 2010, three years ahead of any of its competitors. However, the company had limited impact on a market which was already dominated by three strong players. In August 2014, UNE merged with one of the existing operators, Tigo, to form a converged operator intended to possess sufficient scale to compete against its rivals. State-owned CNT in Ecuador also has gained only limited market share, having been awarded 4G spectrum in the 700MHz and AWS bands in 2012, while its much larger competitors Claro and Móvistar are still waiting for the opportunity to acquire spectrum. In Chile, the fixed operator VTR was awarded 30MHz of AWS spectrum in 2009, three years ahead of any of the country s established operators. VTR Móvil struggled in the marketplace and finally decommissioned its mobile network in January 2014, and now provides mobile services on an MVNO basis over the Móvistar network New regulator in Mexico takes action against América Móvil In 2013, the Mexican government established a new telecoms and media regulator, IFT, with greater powers than its predecessor COFETEL to impose remedies for competitive imbalances in the market. In March 2014, IFT ruled that the América Móvil group (including mobile operator Telcel) was a Preponderant Economic Agent (an excessively dominant player as a result of having more than 50% of the market share) in the country s telecoms market, and imposed a series of measures designed to reduce América Móvil s dominance. These measures included an obligation for the operator to provide access to its passive infrastructure on transparent terms, as well as the imposition of asymmetric mobile call termination rates (MTRs). This obliged Telcel to charge lower MTRs than its competitors can charge Telcel, and was followed in August 2014 by secondary legislation abolishing Telcel s MTRs altogether. The full range of measures will make it increasingly difficult for Telcel to compete aggressively on price, particularly in the voice market. It was recently reported in the press that América Móvil was considering selling off part of its customer base and infrastructure to reduce its share of the Mexican telecoms market to under 50% in a bid to get the regulatory restrictions lifted The mobile industry in Latin America 23

401 THE MOBILE ECONOMY 2014 LATIN AMERICA Latin America set for a new wave of MVNOs A number of MVNOs have either entered the Mexican market or announced plans to do so during 2014, encouraged by the more pro-competitive regulatory environment. The first and largest was UK-based Virgin Mobile, which launched services in Mexico in June, operating over Telefónica s network. According to press reports, Virgin aims to sign up one million prepaid subscribers by the end of its first year of operations in Mexico and to win a 3% market share within five years 8. The company is targeting the year-old demographic and has introduced per-second call charging (resulting in lower-cost phone calls). Mexico is the third market Virgin has entered in the region, following Chile in 2012 and Colombia in The company has reported high growth rates adding 100,000 new subscribers per month in Chile and Colombia by the time of the Mexico launch, according to the media 9. In Colombia, Virgin also introduced per-second billing along with unlimited access to the WhatsApp messaging app for contract plans offering 1GB of data per month or more, as part of its drive to target younger users. The company has also stated it wishes to launch 4G services in Colombia, along with value-added services and links to the world of entertainment and music, including through event sponsorship. Latin America as a whole is expected to experience a big wave of MVNO launches in the near future, including no fewer than 22 planned MVNOs that GSMA Intelligence has identified in Brazil, where only three have launched to date 10. While this is an encouraging development, the MVNOs that have already launched (including in Brazil) have found it hard to win market share from their well-established competitors. Indeed, MVNOs face similar issues to the new entrant operators in the region, albeit that their start-up costs are lower given that they do not have to invest in their own network infrastructure. The more successful and enduring MVNOs are those that can also act as agents of service innovation and attract under-penetrated market segments, as Virgin Mobile is trying to do with its data bundles aimed at the youth market and its per-second billing tariff plans The global MVNO landscape, (GSMA Intelligence, June 2014), p The mobile industry in Latin America

402 THE MOBILE ECONOMY 2014 LATIN AMERICA Consolidation is underway Another trend that has the potential to reshape the region s competitive landscape is consolidation among the leading players. The Brazilian market has been a particular focus of speculation, with press reports indicating that several of the leading mobile operators in the country are all studying options to buy, sell or merge assets. The Brazilian Communications Minister was recently quoted as saying the model in Brazil could be different in the future, and instead of having four operators we could have three, but we have to continue our oversight. Governments in the region may be reluctant to approve major consolidation deals if they lead to a reduction in the number of competitors in their countries, which runs counter to their strategy of promoting competition. This has helped encourage speculation as to whether US operators might be interested in acquiring stakes in Latin American carriers. In May 2014, AT&T acquired the pay-tv provider DirecTV. The company has operations across Latin America, as well as possessing 2.5GHz spectrum assets in Colombia and Brazil, and being a prospective bidder for AWS and 2.5GHz spectrum in Venezuela in late However, as a condition of this acquisition, AT&T did have to agree to divest its 8% stake in América Móvil, with which DirecTV competes in the Latin American pay-tv market. AT&T has recently announced the acquisition of Iusacell, the third largest mobile operator in Mexico, for US$2.5 billion. It remains to be seen whether the present wave of merger and acquisition activity will present further opportunities for United States players to invest in the region, which could see a further increase in competitive pressures in the targeted markets. Other operators that have been reported in the press as potentially for sale are Nextel (having sold its Chilean and Peruvian units but still holds Argentina, Brazil and Mexico), and Telecom Argentina (which has been in protracted discussions with investor group Fintech). The mobile industry in Latin America 25

403 THE MOBILE ECONOMY 2014 LATIN AMERICA 1.5 Revenue growth will continue to slow Total mobile revenues in Latin America amounted to US$109 billion in 2013, up 4% year-on-year, making it the third-fastest growing region globally. As with subscriber growth, revenue growth has slowed sharply in recent years, with the average annual growth rate of 7% over the last five years well below the rates in the upper 20% per annum range attained in the previous five-year period. Source: GSMA Intelligence Latin America mobile operator revenues (US$ B) Revenue growth is expected to slow still further in the period out to This slow-down is connected with the trends discussed earlier in this section including increasing competition; market maturity; falling prices; and growing regulatory intervention. Partly as a result of this slow-down in revenue growth, there has been a fall in industry profitability over recent years. It is noteworthy that margins on average in Latin America are well below the level for other developing markets. This makes it more challenging for operators to fund investments at a time when there is ongoing pressure to improve networks to address capacity constraints (and address quality of service issues) as well as to launch higher-speed networks (especially LTE) and, in several countries, to meet new regulatory obligations, such as around geographic coverage. 26 The mobile industry in Latin America

404 THE MOBILE ECONOMY 2014 LATIN AMERICA 1.6 Ongoing operator investments to support growth Operators in the region are continuing to invest heavily in their networks to improve coverage, capacity and quality of service. Total capital expenditure by operators over the last six years has totalled over US$96 billion. Investment levels are likely to increase further over the coming years as LTE deployments gather pace in the region, and as operators continue to improve coverage and capacity for existing 3G networks, with almost US$193 billion set to be invested over the seven-year period from 2014 to Source: GSMA Intelligence Latin America operator capex (US$ B) US$96B US$193B Capital investments grew by an average of 8.7% in the five years to 2013, with a sharp recovery in investment levels after 2010, as the region emerged from the economic downturn. Between 2014 and 2020, the average annual growth in capital expenditure is expected to fall to 4.6%. In this context, it is imperative to foster a favourable investment climate, with more predictability and transparency around regulatory obligations and restrictions, as well as more certainty around access to and cost of necessary spectrum resources in the medium and long term. It is also important to reduce the level of risk surrounding such large-scale investments and to allow operators a reasonable return. It is particularly important that governments and regulators avoid excessive taxation and penalties for failure to meet stringent coverage and quality-of-service targets. It is in the interests of all industry stakeholders including those of the operators to improve performance in these areas. The mobile industry in Latin America 27

405 THE MOBILE ECONOMY 2014 LATIN AMERICA 2 Delivering digital inclusion in Latin America The increasing ubiquity of mobile networks and the uptake of more advanced devices has already had a profound impact on the way people live, and brought communications services to previously unconnected populations. In particular, mobile networks have helped facilitate access to the internet, which through providing new sources of information and access to a range of new services, can deliver broad and economic and social benefits. 28 Delivering digital inclusion in Latin America

406 THE MOBILE ECONOMY 2014 LATIN AMERICA Internet penetration rates across Latin America have grown rapidly in recent years. It is mainly mobile networks and services that are driving this growth, given the relative lack of fixed line infrastructure in the region. International Telecommunication Union (ITU) data indicated that fixed-line penetration in the region on average stands at less than 20% of the population. Even in those markets that have higher than average fixed line penetration, such as Brazil and Argentina, fixed line services tend only to be accessible to higher income groups. For example, research previously commissioned by the GSMA highlighted that 60% of households at the top of the pyramid have adopted fixed broadband in the region, but penetration at the bottom of the pyramid is just 8% 11. At the end of 2013, there were over 200 million individuals using mobile devices to access the internet, with around 60% already doing so over higher speed 3G or in some cases 4G networks. This was equivalent to an overall penetration rate of around 33%, slightly ahead of both the global average and a number of other developing regions across the world. By 2020, this figure is forecast to be just under 50% of the population. Source: GSMA Intelligence Mobile internet subscriber penetration 17% 38% 23% 39% 28% 49% 33% 48% SSA MENA APAC LatAm However, there remain large populations in the region who still do not have internet access, particularly in the poorer and smaller countries of Central America and the Caribbean. This is likely still to be the case by 2020, even with the significant growth in the mobile internet user base described above. There are also variations in terms of internet access across the region. There are a number of countries where under 20% of the population has internet access, either fixed or mobile, including Haiti, Nicaragua and Guatemala Delivering digital inclusion in Latin America 29

407 THE MOBILE ECONOMY 2014 LATIN AMERICA The digital divide refers to the gap between populations that have access to the internet and new technologies, and those that do not. Internet and technology access are two of the key variables in determining a country s position in the human development index. The importance of addressing the digital divide has been recognised by governments across the region, with many countries having both national broadband plans and broader ICT strategies that aim to widen access to broadband connectivity. Mobile will be the key technology to help connect these largely offline populations in the region, particularly in bridging the digital divide. Mobile broadband in particular has the potential to transform access to jobs and education in these countries. A number of studies have shown that broadband access is a key driver of both job creation and economic growth, both in Latin America and other parts of the world. The GSMA launched its Digital Inclusion programme in April of 2014 in order to expand global connectivity and to address the barriers to mobile internet adoption. The programme will collaborate with mobile operators, governments, internet players and non-government organisations to address four key barriers to mobile internet access. Network infrastructure and policy: expanding rural network coverage of mobile internet by promoting regulatory and technical best practice Affordability and taxation: reducing the total cost of ownership of mobile internet Consumer barriers: increasing adoption and usage by addressing literacy and internet awareness barriers Local content: promoting locally relevant content to attract people to use the mobile internet 30 Delivering digital inclusion in Latin America

408 THE MOBILE ECONOMY 2014 LATIN AMERICA 2.1 Network infrastructure and policy Issues around network coverage in Latin America are generally less pressing than in other developing regions. 2G coverage is generally ubiquitous across the region, whilst considerable progress has also been made in building out 3G networks and the region is also now seeing increasing LTE coverage. However, there remain large populations in the region who still are not covered by higher speed networks, particularly in parts of Central America. Many of the still unconnected populations live in rural areas, where a lack of electricity infrastructure can also impact on efforts to extend network coverage Measures to improve network coverage: the industry players One measure to help improve access in remote and inaccessible areas is the use of green solutions to power base stations, such as solar, wind, water, biomass and fuel cells. Such alternatively powered cell sites can reduce the cost of using diesel, the most common power source in areas off the main electricity grid. Telefónica is the leading operator in this respect, having deployed around 100 solar-powered sites in Ecuador; 65 in Peru; and a combination of solar, wind and fuel cell technologies in a number of base stations in Venezuela and Chile. Network sharing is another solution that can help reduce the cost of extending network coverage, particularly into remote or geographically challenging areas. Network sharing can also have a role to play in helping to increase capacity in urban areas, particularly where operators are looking to deploy small cell technology. Historically network sharing has been more common in other parts of the globe. However, there is an increasing trend for operators across the region to adopt a variety of infrastructure models. There are voluntary network sharing deals between the leading operators in most of the major economies of the region, involving passive infrastructure (such as physical sites, towers and power supplies). There are also third-party tower providers in several countries that lease sites and passive equipment to operators. In addition, there are a handful of deals involving sharing of active infrastructure for the deployment of 4G-LTE services. This includes sharing radio access network (RAN) equipment, such as transmitters and control systems. These have been strongly incentivised by regulators, for example in order to meet demanding coverage build-out obligations in Brazil or as part of licence terms in Chile. The largest such deal is between Oi and TIM Brasil in Brazil, involving LTE infrastructure in the 2500MHz band covering 45 cities. It was reported in the press in October 2014 that the arrangement would be extended to a further 88 locations in Delivering digital inclusion in Latin America 31

409 THE MOBILE ECONOMY 2014 LATIN AMERICA Measures to improve network coverage: governments and regulators Whilst operators are starting to address the coverage challenge, there are measures that can also be taken by governments and regulators to help support the goal of improving mobile broadband coverage. These particularly relate to issues around spectrum management, network sharing and reducing local and municipal government red tape. a) Spectrum management and release of the Digital Dividend spectrum bands The timely release of spectrum to accelerate mobile broadband deployments, particularly the Digital Dividend spectrum (in the 700MHz range) and the AWS spectrum (1700/2100MHz), can play an important role in helping to facilitate network deployments. Lower frequency bands in particular have broadest geographical coverage and are therefore more cost-effective in providing service to more remote, rural areas. The spectrum management policies of countries in the region are therefore key in helping realise the potential of mobile broadband, particularly in the countries which to date have seen the lowest levels of mobile broadband uptake. b) Regulators should facilitate further network sharing The GSMA believes that regulators should facilitate further voluntary network sharing arrangements by operators, by, for instance, easing planning controls to encourage shared fibre transmission facilities, including preferential access along roads and other state property. Furthermore, in countries where this is not yet allowed, registered providers (including third party tower companies) should be permitted to construct and acquire passive infrastructure and then sell access to operators. c) Reduction of local and municipal government red tape A big problem across Latin America for the deployment of infrastructure has to do with municipal restrictions and red tape. These restrictions are often against international standards and not aligned with National Broadband Plans or Digital Agendas that seek to promote ICT inclusion. Unfounded health concerns or simply discretional ability to manage permissions create many different negotiating terms for mobile operators and affect their ability to improve network coverage. Peru passed a bill in July 2014 that could be a good example by which national governments could shape municipal administrative requirements so as to be aligned with international standards and prevent barriers for investments in infrastructure See: 32 Delivering digital inclusion in Latin America

410 THE MOBILE ECONOMY 2014 LATIN AMERICA 2.2 Improving the affordability of mobile internet access Affordability can be a significant barrier to internet access. Latin America scores well compared to other developing regions in terms of the proportion of the populations living below the UN s poverty level of US$2 per day (only 13% of the population in ). However, this still leaves a relatively large number of people in absolute terms across the region for whom poverty is still an issue. Even for those well above the poverty line, affordability can still be a challenge for households looking to balance their budgets and to pay for a broad range of goods and services. In Latin America, households in the three lowest income deciles have average monthly household incomes below US$114. Even with the declining prices of mobile data plans and devices, higher speed connectivity therefore remains beyond the reach of many Measures to improve affordability: industry players Falling device prices have played an important role in improving the affordability of mobile services, particularly for more advanced devices such as smartphones and tablets. However, smartphone penetration remains relatively low in many less developed countries in the region, constituting a major obstacle to accessing the internet and, in particular, more advanced services and applications. Smartphone adoption will be helped by ongoing declines in handset pricing. However, for the time being, such devices remain beyond the means of large portions of the population. A recent study by McKinsey showed that the average retail price of a smartphone in 2013 represented 5.8% of GDP per capita in Colombia (US$439), 4.5% in Brazil (US$521) and 2.6% (US$259) in Mexico, putting smartphones well beyond the reach of precisely those poorer sections of the population that have yet to go online 14. The efforts being made by OS developers, local device manufacturers and operators to bring more affordable smartphones on to the market were discussed earlier in the report. It is expected that the greater affordability of smartphones will make an important contribution to boosting penetration to almost 70% of active connections in the region by 2020, from 30% today. Nonetheless, much of the growth in smartphone penetration will come from the replacement of existing basic and feature phones, with mobile internet subscriber penetration set to reach just half the population by that date. This will leave a material proportion of the population still lacking internet access. Operators have already played a role in improving the affordability of mobile services, particularly with regard to mobile data and internet access, and the challenge of making these services available to low income consumers on prepaid tariff plans 15. For example, a range of flexible tariffs are now available from operators focusing on mobile data, including plans that allow users to access the internet for a limited period of time (e.g. one or more periods of 24 hours), set minimal download limits or alternatively allow access to particular web properties (such as Facebook or WhatsApp). Research commissioned by the GSMA highlighted that mobile data prices had fallen significantly across the region. For example, service tariffs have dropped between 7.3% for dongles and 52% for smartphones in the last three years up to At the same time, daily mobile internet plans with low fixed prices are allowing people living in low income households to access the internet on a more regular basis Offline and falling behind: Barriers to Internet penetration (McKinsey & Company, August 2014) 15. https://gsmaintelligence.com/files/analysis-subscription/?file= prepaid-data-tariffs.pdf 16. Mobile Broadband at the Bottom of the Pyramid in Latin America (TAS, July 2013): Delivering digital inclusion in Latin America 33

411 THE MOBILE ECONOMY 2014 LATIN AMERICA Measures to improve affordability: governments and regulators Taxation of mobile services remains relatively high in many emerging markets, including a number across Latin America. A recent study by the GSMA found that across a sample of 19 countries, US$3 out of every US$10 of mobile revenues was paid to the government in the form of taxes, regulatory fees or other charges 17. The report highlighted several highly taxed mobile countries in Latin America, including both Brazil and Colombia. These two countries have a mobile services tax burden of over 30% of revenues. These high levels of taxation add significantly to the cost of services for consumers, and so prove to be additional obstacles to the goal of delivering more affordable mobile services to currently underserved populations. 2.3 Consumer barriers: improving digital literacy and awareness Populations that currently have limited mobile and internet access often also have low levels of literacy. Research published by the ITU in 2012 showed a positive correlation between digital literacy (encompassing both basic literacy in the traditional sense but also the skills needed to use and access technology) and levels of internet access. There are additional and related challenges for some population segments ranging from a lack of basic awareness of the existence of the internet to a lack of knowledge of how to use and access the internet. There are a number of initiatives in the region aimed at improving digital literacy rates, including numerous government and non-government organisation programmes to deliver personal computers, and fixed or mobile connectivity, to schools although, to date, many of these programmes are based around computers rather than mobile phones. With their graphical and tactile interfaces, smartphones (and mobile broadband) can play a vital role in bringing internet access, and a wide range of useful apps and services, to those with limited computer literacy or even minimal basic literacy. Smartphones therefore represent a highly cost-effective means to bridge the digital divide and bring the poorest portions of the region s population online as well as being more affordable and practical for users to run, as they do not require a permanent connection to an electricity supply Delivering digital inclusion in Latin America

412 THE MOBILE ECONOMY 2014 LATIN AMERICA 2.4 Improving the availability of local content Another crucial barrier to the uptake of mobile broadband is the absence of locally relevant content, i.e. content that is accessible, useful and relevant to the livelihoods, wants and needs of people in developing economies. The issues involved in improving local relevance include infrastructure challenges such as network coverage, content hosting availability and device compatibility, as well as more nuanced elements around payment limitations, language, cultural factors, and government or third party support. Consequently, addressing these barriers will necessitate a convergence of efforts from key players across the mobile ecosystem, from operators to handset manufacturers, and content developers to internet players. With respect to language in particular, Latin America is in a stronger position than many other developing regions in that the dominance of Spanish in most countries in Central and Southern America, and Portuguese in Brazil (along with English and French in the Caribbean). This creates a huge consumer and business market for content in local languages. However, there is still a relative paucity of locally created content, focused on the specific needs and interests of the economically and culturally diverse, urban and rural populations of the region, which represent a wealth of untapped opportunity for app developers and content providers. There are a range efforts that are being made to stimulate a local app economy and digital ecosystem, as discussed elsewhere in the report. There are also efforts to develop smartphone-based apps and content linked to public services or business needs. In addition, mobile represents the channel of choice for the delivery of basic information, content and services in areas such as health care to remote communities, education, smart energy, M2M applications, mobile money and agriculture. At present, basic services in these areas are mostly provided via SMS, USSD (Unstructured Supplementary Service Data) and voice, which does not require an internet connection and are delivered over 2G networks. The higher prepaid connection base and the limited penetration of smartphones in some countries is also acting as a constraint to develop this opportunity further. In the shorter term, the challenge in Latin America is as much to drive up the share of the population that is able to access vital services such as these via affordable, basic mobile subscriptions as it is to increase mobile internet usage alone. Greater uptake of and reliance on mobile communications to access vital services then creates a base upon which more feature-rich, locally relevant, real time-dependent, culturally appropriate and socially interactive apps and services that depend on good-quality mobile broadband networks can be developed. This in turn will allow the full social and economic benefits of increased Internet penetration to be delivered. Ultimately, relevant local content gives people added impetus for mobile internet and value added services usage. This increases the digital empowerment of users on all rungs of the socio-economic ladder. For operators in particular, this would lead to an increase in subscriptions, customer loyalty, and revenue through data and value added services. Additionally, the availability of an increasing amount of relevant local content would unfurl a virtuous circle of opportunities for local small and medium sized enterprises. These businesses, which are often labour intensive, would be able to participate more actively in the mobile value chain by developing content, creating apps and stimulating innovation. Delivering digital inclusion in Latin America 35

413 THE MOBILE ECONOMY 2014 LATIN AMERICA 3 The social and economic impact of mobile in Latin America 36 The social and economic impact of mobile in Latin America

414 THE MOBILE ECONOMY 2014 LATIN AMERICA 3.1 Mobile makes a major contribution to economic growth in the region The contribution of the sector to the economy of the region reached 4.1% of GDP in This overall impact includes a direct contribution from mobile network operators of US$56 billion in value added, as well as a contribution from directly related ecosystem industries. The mobile ecosystem generated value added of US$24 billion. The economic activity directly generated by both mobile operators and the ecosystem has a broader knock-on effect on the rest of the economy, inducing further economic activity valued at US$16 billion. Additionally, the use of mobile technologies also means a more efficient use of existing resources. The widespread use of mobile technology in businesses and by workers in the region significantly increases business productivity, generating an additional US$147 billion in value added in 2013, or 2.5% of GDP. The direct economic contribution to GDP of the mobile ecosystem (mobile network operators and other related industries) is estimated by the value created by companies that operate in the sector. The value that is added to the region s economy is equivalent to the payment of wages, tax contributions and business profits by the industry 18. Source: GSMA Intelligence Direct GDP contribution of the mobile ecosystem 2013 (US$ B, % GDP) % % % % % Infrastructure and support services Network Operators Handset manufacturers Distributors and retailers Content, applications and other services 18. Value added by the sector can also be approximated as the difference between the value of sales made by the sector and the direct cost of making those sales. The social and economic impact of mobile in Latin America 37

415 THE MOBILE ECONOMY 2014 LATIN AMERICA Further to the direct economic impact, the revenues generated by the mobile industry in Latin America have a multiplier effect on the rest of the economy as a proportion of the wages, taxes or profits paid out by the industry are subsequently spent across other sectors. This effect means that other economic sectors also benefit from the value added generated by the mobile ecosystem. It is estimated that this effect resulted in the generation of additional economic activity in Latin America with a value of US$16 billion in Finally, and in addition to the direct and indirect contribution to GDP by mobile network operators and the wider ecosystem, an estimated 2.5% can be attributed to the increased productivity brought about by the widespread use of mobile technology in other sectors of the economy. Mobile technology has already facilitated productivity improvements in the region for many workers and businesses, for example by playing an increasingly important role in improving the efficiency of transport and logistics in the region. The increase in productivity is estimated to have generated in 2013 an additional US$147 billion contribution to GDP. Productivity growth has the potential to increase further going forward as more efficient practices are adopted on the back of increasingly advanced services supported by mobile broadband technologies. Overall, in 2013 the mobile industry made a contribution to GDP of US$242 billion or 4.1% of the region s total. Source: GSMA Intelligence Total (direct and indirect) contribution to GDP (2013 US$ B) % % 0.3% 4.1% 0.9% Mobile Operators Related industries General economy Productivity Improvement Total 38 The social and economic impact of mobile in Latin America

416 THE MOBILE ECONOMY 2014 LATIN AMERICA In 2013, the mobile industry provided direct employment to approximately one million people in the region. Most of these jobs were provided by mobile network operators (320,000), distributors and retailers (285,000), and content providers (255,000). There were smaller but also material numbers of jobs in the areas of handset manufacturing and infrastructure provision. Further to the employment that is sustained within the ecosystem, additional jobs were also indirectly supported in the rest of the economy as wages, public funding contributions and profits paid by the industry are re-spent elsewhere in the economy. It is estimated that over one million jobs were indirectly supported in this way, bringing the total impact of the mobile industry in Latin America to more than two million jobs in Source: GSMA Intelligence Direct and indirect job contribtion from the mobile ecosystem 2,194,420 1,227, , ,405 75, , ,116 34,953 Infrastructure Operators Handset Distribution Content Direct Indirect total The social and economic impact of mobile in Latin America 39

417 THE MOBILE ECONOMY 2014 LATIN AMERICA The mobile industry also makes a very important contribution to public funding in the region. In the first place, the mobile ecosystem makes a sizeable contribution through general taxation, which for most countries includes sales and corporation tax, as well as contributions made by people directly employed in the industry through income tax and social security contributions. It is estimated that the sector made a contribution to the public finances of governments of around US$41 billion in Source: GSMA Intelligence Mobile ecosystem contribution to public funding (2013 US$ B) MOBILE SERVICES VAT HandsetS VAT CORPORATION TAX EMPLOYEE INCOME TAX AND SOCIAL SECURITY CONTRIBUTIONS total TAX (EXCL. SPECTRUM AUCTION) Mobile network operators made further contributions to the public finances of government through fees for the licence of spectrum bands required for the deployment of mobile broadband services. In 2013, over US$1 billion was raised in the region through spectrum auctions. Auctions have continued in 2014, with Brazil s 700MHz auction in September raising approximately US$2 billion for the government. Argentina s recent 4G auction results indicate the government will raise US$2.3 billion. 40 The social and economic impact of mobile in Latin America

418 THE MOBILE ECONOMY 2014 LATIN AMERICA Outlook to 2020 Going forward, the mobile industry will continue to increase both in absolute and relative terms its contribution to the Latin American economies. This will be brought about by increasing subscriber numbers and value added services generated by the ecosystem, as well as the growing use of mobile technologies by businesses and workers. This will enable a more efficient use of resources not only within the sector but throughout the whole economy. By 2020, it is estimated that the mobile ecosystem will generate value added of US$275 billion, representing at that point 4.5% of the region s projected GDP in 2020, up from its current contribution of 4.1%. Source: GSMA Intelligence Total mobile contribution to GDP out to 2020 (US$ B) 4.5% 4.1% The predicted increase in GDP contribution over the period will also be accompanied by a strong growth in the number of people directly employed by the industry. The growth in the number of mobile broadband connections and the development of new content, services and products will result in the hiring of new employees by companies across the ecosystem. The number of people directly employed by the industry will reach 1.3 million by 2020, up from just under one million in At the same time, the public funding contribution of the mobile ecosystem (excluding spectrum and other regulatory fees) will reach US$53 billion by 2020, up from US$41 billion in Assuming that tax rates remain at current levels The social and economic impact of mobile in Latin America 41

419 THE MOBILE ECONOMY 2014 LATIN AMERICA 3.2 Mobile facilitating entrepreneurship and commerce in Latin America The increasing adoption of smartphones and the ongoing migration to mobile broadband services in Latin America is creating huge potential for new mobile apps and mobile-driven businesses. According to ComScore, the proportion of web pages accessed via smartphones or tablets in the leading Latin America economies approximately doubled in the year to May 2014, with Mexico (22.3%) and Chile (18.5%) recording the highest shares 20. The most popular apps in the region are freeto-download social-networking, messaging and entertainment services such as Facebook, WhatsApp and Google Play. There is also a growing volume of low-cost mobile games, with one survey estimating the value of the Latin American mobile gaming market at US$400 million in However, the market potential for mobile enterprise apps is largely unfulfilled, with a recent survey by the chipset manufacturer Qualcomm reporting that only 15% to 20% of Brazilian enterprises use mobile apps to address specific business needs, as opposed to generic communication and collaboration apps 22. The reasons for the region s relatively low adoption of apps that offer sufficiently valuable content or business functionality that users are willing to pay for them are complex. One issue highlighted by a 2013 GSMA survey is privacy, with over 80% of mobile internet users in Brazil, Colombia and Mexico having concerns about sharing their personal data when accessing mobile apps. In addition, 48% of users with privacy concerns of this sort stated that they were prepared to limit their use of apps unless they feel their personal information is adequately safeguarded 23. Across the region, private and public actors in the mobile ecosystem have launched a range of new initiatives aimed at fostering the development of a vibrant, locally relevant app economy. One notable example is Telefónica s incubator Wayra, which provides business-development, investment and technical support for online and technology startups in the region s largest countries although not limited to mobile-centric apps and services. 24 Examples of businesses that have been assisted by Wayra include Ocapi (an intelligent onlineadvertising system allowing real-time modification and re-targeting of online ads based on knowledge of the user s interests and present context); and Tryouts (a site setting people learning challenges and coding tasks structured like a gaming quest, and generating job offers based on users skills and achievements as demonstrated on the site). While the apps incubated by Wayra are focused on the needs of businesses, employees and consumers, another operator-led initiative from the Brazilian operators association Telebrasil addresses the social and economic needs that can be met through digital innovation. The project Innovative and Competitive Digital Brazil , launched at a Telebrasil congress in September 2014, includes a 30-point plan under five broad axes. 25 Two of these axes are as follows: Services and Applications : this includes recommendations about infrastructural and institutional change that is required to realise the full potential of ICT to deliver basic services (such as health care and education) and e-government in the country. Innovation and Entrepreneurialism : this urges the country s authorities to increase the focus on R&D and innovation, on fostering a culture of innovation, and on creating infrastructure platforms (such as M2M technology, Big Data and cloud computing) that are needed to support innovative enterprise application development. 20. Futuro Digital Chile 2014 (ComScore, August 2014), p Tango, Samba and Mariachis: The Latin American Mobile Market (behaviour, 2014): akacao= & akcnt=6a08b7ae& akvkey=9bf2&utm_source=akna&utm_medium= &utm_campaign=newsletter+oi+corporativo+-+05%2f06%2f %3a08#.vbiugusqe6k 23. Mobile Privacy: Consumer research insights and considerations for policymakers (GSMA, February 2014) Projeto Brasil Digital Inovador e Competitivo (Telebrasil / LCA, September 2014) 42 The social and economic impact of mobile in Latin America

420 THE MOBILE ECONOMY 2014 LATIN AMERICA Throughout the region, there are a number of centres emerging with a concentration of technology startups that government and international bodies are seeking to support through subsidies and incentives. For example, in Chile, a smart city project has been launched in the Gran Concepción conurbation, with support from the World Bank and the country s Ministry of Transport and Telecommunications 26. In Brazil s financial capital São Paulo, a dynamic start-up ecosystem has grown, which has received support from both equity funding and government programmes 27. The start-ups in the city are strongly focused on the social needs that should be served through technology innovation, which in turn will help realise the enormous growth potential. As Yuri Gitahy, founder of the business accelerator Aceleradora states: There is a deep economic change with the rise of the lower-income classes, which generates a massive consumer market that will evolve over the next 10 or 20 years. So it s not about the ecosystem, it s about the growing number of market opportunities Mobile commerce: low uptake to date but huge potential With mobile becoming the primary means of accessing the internet for a growing number of users, the opportunities for mobile commerce (m-commerce) are also considerable. Mobile terminals and billing platforms are also increasingly used as payment methods by populations across the region. A recent report showed that the share of e-commerce sales transacted over mobile devices in Brazil nearly doubled from 2.5% in early-2013 to 4.8% at the year end, and this strong growth trend is likely to continue. 28 While the share of e-commerce transactions completed over mobile devices and apps is still low, it is growing fast in the context of an overall ecommerce market that is also enjoying rapid expansion. According to one recent survey, Latin America is now the second-fastest growing e-commerce market worldwide after China, with the web sales of the region s top 500 retailers increasing by 23% to $17.94 billion in In addition, the C2C and B2C e-commerce platform MercadoLibre, which operates across the region, reported that it expects m-commerce to account for 50% of its business by 2019, from 12% of unit sales as at May Reasons for the relatively low level of m-commerce uptake to date include: lower level of smartphone and mobile broadband penetration than in other developed and developing regions; security concerns such as those relating to app usage referred to above, and to mobile money discussed later in this report; the low level of financial inclusion and high share of the population without bank accounts or credit cards; and the absence of m-commerce apps, content and products in local language and of relevance to the Latin American consumer. The different regulatory regimes in markets across the region is a further barrier. However, all of these factors are now being addressed by mobile ecosystem players and internet retailers alike. As described elsewhere in this report, smartphone penetration is expected to more than double out to 2020, and mobile broadband availability is expanding rapidly as 3G and 4G networks are deployed. In addition, e-commerce players are addressing the issue of local language and content. For example, ebay launched Spanish- and Portuguese-language apps for the first time in 2014, with content and offers tailored to the local markets; and Amazon which launched an e-commerce site in Brazil in 2012 started selling printed books in Portuguese during Mobile payment platforms are being integrated with commerce platforms and apps, so that m-commerce transactions will increasingly be carried out using mobile money accounts, mobile wallets and e-money via mobile devices. M-commerce, combined with mobile money services, offers the important benefits of convenience, and enhanced access to affordable goods and services, to Latin American consumers. Moreover, the mobile medium also taps into the Latin American consumer s habits of carefully managing and budgeting expenses and bills, as mobile money services and apps provide an easy means to track all of the user s spending and accounts. The different legal and regulatory frameworks across the region, which often fail to take the specificities of web-based commerce into account, may take longer to harmonise. However, at an individual country level, progress is being made regarding the responsibilities of payment service providers, guarantees for mobile money users and interoperability of mobile money platforms (several of which are present across the region) Report by local consultancy Ebit, referred to by Business Monitor International, America Telecommunications Insight (May 2014), p The social and economic impact of mobile in Latin America 43

421 THE MOBILE ECONOMY 2014 LATIN AMERICA 3.3 M2M services are growing rapidly across the region One area of mobile-driven innovation that is showing rapid growth across the region is the number of M2M connections and services. The Internet of Things (IoT) will offer a range of innovative new services to consumers whilst also helping to address some of the broader challenges in the region. The GSMA s vision for the IoT is the Connected Life. The term Connected Life refers to a world in which consumers and businesses use many different devices to experience compelling new services and ubiquitous internet access delivered via mobile networks. These devices include the next wave of smartphones, tablets and consumer electronics, as well as machines, vehicles, monitors and sensors equipped with M2M communications. There were under 16 million cellular M2M connections in Latin America as at September Growth is expected to be strong over the next few years, at a CAGR of 25% out to 2020 and with the total reaching 66 million connections by that date. Source: GSMA Intelligence M2M connections in Latin America (M) The GSMA Intelligence M2M connections data used in this report refers exclusively to a SIM connection that enables mobile data transmission between machines. It does not count SIMs used in computing devices in consumer electronics such as smartphones, dongles, tablets, e-readers, routers and hotspots 44 The social and economic impact of mobile in Latin America

422 THE MOBILE ECONOMY 2014 LATIN AMERICA As at the second quarter of 2014, Brazil was the fourth-largest M2M market worldwide, with 9.1 million connections, representing around two thirds of the Latin American total and 5% of the global total. The country is projected to remain in fourth position by 2020, by which time the number of connections will have risen to 41.9 million. Across the region, M2M accounts for only around 2% of all connections; but this will rise to 7% by There could be substantial upside to these forecasts if a number of growth factors are successfully enabled by both industry players and governments in the short to medium term, as highlighted in recent research by GSMA Intelligence 32. The areas identified as possible growth stimulators that are of particular reference to Latin America include the following: Government policies to facilitate wider deployment of cellular M2M in key sectors such as utilities, smart cities, automotive and healthcare; The standardisation of technical solutions and components; and The development of new business models by operators, particularly those that move beyond the provision of simple connectivity. Supportive regulatory frameworks can play an important role in stimulating the deployment and adoption of new M2M applications and services. One example of a progressive government policy is the tax break introduced in 2014 for M2M connections in Brazil. M2M connections normally generate a significantly lower ARPU than personal connections, so a reduction in the tax charges associated with M2M should play an important role in market development going forward. Another M2M service with a significant regulatory driver is the stolen vehicle-tracking SIMRAV project (Sistema Integrado de Monitoramento e Registro Automático de Veículos) in Brazil, otherwise known as the CONTRAN 245/07 regulation. CONTRAN requires that all new vehicles be fitted with the capability to be tracked and remotely disabled in case of theft. The project has been postponed several times already, mainly because vehicle manufacturers do not want to add to production costs. In May 2014, DENATRAN, the Brazilian national transit department, announced that the project would be delayed by a further two years. The first of these two examples in particular illustrates the importance of gaining buy-in and co-operation from all the public and private stakeholders right from the planning stage of any new public-interest M2M deployment. This allows the elaboration of business models and technical specifications, which in terms means that the cost-benefit for all of the actors can be clearly understood. 32. https://gsmaintelligence.com/analysis/2014/9/cellular-m2m-forecasts-and-assumptions /446/ The social and economic impact of mobile in Latin America 45

423 THE MOBILE ECONOMY 2014 LATIN AMERICA Automotive leading the way in Latin America Despite the difficulties encountered by the CONTRAN regulation, Brazil is nonetheless leading the way in terms of automotive M2M deployments, with around 3.5 million connections, compared with 8.8 million across the region at end-2013, according to Machina Research 33. The majority of deployments to date are in the areas of vehicle and fleet tracking, and vehicle security, with the need to improve the efficiency of logistics networks being an important consideration in the region. These segments are expected to grow strongly out to Other applications will also see rapid growth, including more premium services such as pay-as-you-drive insurance and in-car infotainment services. In September 2014, BMW launched its ConnectedDrive concept in the Brazilian market, in the shape of its i3 electric car. This model comes equipped with a built-in SIM card providing connectivity to the BMW cloud via a smartphone interface. Using this connection, a number of services are delivered, such as: emergency calls (and soon, break-down calls); vehicle-performance information; satellite navigation; and route planning and distance calculation to minimise battery or fuel consumption, and determine whether the driver will need to charge up or re-fuel to reach their destination. A smartphone app also enables the driver to control other vehicle functions internally (without connecting to the cloud), such as air conditioning, assisted driving and the rear-view camera. 33. Machina Research s M2M numbers employ a broader definition than GSMA Intelligence s M2M projections, which are limited to cellular network connections 46 The social and economic impact of mobile in Latin America

424 THE MOBILE ECONOMY 2014 LATIN AMERICA Smart grids and smart cities Latin America has four out of the 15 largest urban conglomerates in the world: Mexico City (ranked 3 rd ), São Paulo (6 th ), Buenos Aires (12 th ) and Rio de Janeiro (14 th ). All four of these are already classified as megacities, with populations in excess of 10 million. The region is expected to have 585 million urban citizens by 2030, and by 2050, over 90% of the population in the region is forecast to be living in urban areas. 34 In the developing world, the primary challenge for cities is typically over-congestion (both urban density and traffic volume) caused by the unprecedented rapidity of urbanisation in the past years. A recent survey by the Inter-American Development Bank highlighted how these challenges impacted individuals in Mexico City, São Paulo and Buenos Aires. It found that, for example, over 28 million people in these three cities alone commuted for over 90 minutes every day, while lower income households tended to be more exposed to electricity blackouts and voltage fluctuations than higher income households. Smart city services and applications can help to move urban growth onto a more environmentally sustainable footing, as well as reduce the impact of issues such as congestion and pollution. In addition, smart grid and smart metering systems offer the prospect of reducing the waste, costs and pollution of inefficient energy use and energy fraud, thereby improving the reliability of utility services in the region s cities, and allowing the benefits of modern power and water supply to be extended to poorer, rural areas. In Chile, Smartcity Santiago was launched in early 2013 by electricitydistribution company Chilectra. This is a prototype urban environment incorporating a range of technologies, including: smart metering/smart grid technologies; renewable power generation; energy-efficient homes incorporating smart home automation; electric vehicles; and automated security systems. The goal is to demonstrate to local government and the city s residents the benefits that these new technologies can provide. There have been a number of other smart city initiatives in Santiago, most notably a system for variable, automated toll charging on some of the city s most congested highways, which uses short-range, non-cellular wireless tags to handle communications between registered vehicles and the system. There has also been a number of smart city initiatives in São Paulo. Three of these have involved connectivity supplied by the local Telefónica mobile subsidiary Vivo: a system for tracking the city s waste-collection fleet; a realtime flood warning and prevention system; and a traffic-management system for the city s public bus service. The latter service enables PC or smartphone users to track buses on their proposed routes, and to access estimated pick-up and arrival times for their journeys The social and economic impact of mobile in Latin America 47

425 THE MOBILE ECONOMY 2014 LATIN AMERICA The role of the GSMA in realising the potential of the IoT The initial focus of the Connected Living programme is to address key barriers and challenges that are prohibiting the development and growth of M2M/IoT connections and services across the world. This will be achieved by industry collaboration and appropriate regulation, as well as by developing key enablers to support the growth of the M2M market and new services and applications. The GSMA works with its ecosystem partners to establish guidelines for how machines should communicate via the mobile network in the most intelligent and efficient way. The goal is to define future network requirements and capabilities that are required to allow operators to support IoT devices and services. The GSMA s vision is to unite all industry stakeholders behind a single, common embedded SIM specification to help accelerate the growth in M2M connections. The GSMA recently announced an agreed specification for a machine (embedded) SIM that addresses remote provisioning requirements. The embedded SIM specification project has the support of a number of the world s leading mobile operators. The GSMA is working on a range of business enablers that aim to create a sustainable regulatory and policy environment that enables operators to deliver the consumer and business benefits of the IoT. The GSMA is also working to identify and develop enablers that will support future network capabilities such as in the areas of end-to-end security, non-gsm device authentication and billing and charging. The GSMA also engages with partners across the broader ecosystem to help drive adoption and facilitate the development of new services in markets including automotive, health, utilities and transportation. In the automotive field, the GSMA automotive special interest group allows automakers and operators to share information and enable industry cooperation, with the goal of resolving barriers to connected car deployments and to speed the adoption of telematics and infotainment services. 48 The social and economic impact of mobile in Latin America

426 THE MOBILE ECONOMY 2014 LATIN AMERICA 3.4 Delivering access to basic services across Latin America There are a number of common challenges facing developing economies across the world, and a number of these apply in Latin America. High levels of population growth can bring challenges in ensuring access to basic infrastructure and services such as electricity, education, healthcare and banking. In contrast, the rapid growth in urban areas is creating increasing challenges in relation to congestion and the efficient provision of public services. There is a clear potential for the mobile ecosystem to make a meaningful contribution to addressing these challenges. Mobile networks and services have the potential to help to facilitate access to basic services in areas such as health care, education and agriculture. The number of deployments in Latin America in these areas has increased steadily over recent years. However, the number of services in the region still trails behind other developing regions such as Sub-Saharan Africa, and there is clearly the potential for mobile to make a greater contribution to the region s social challenges in the future. Source: M4D Impact tracker, GSMA Intelligence Mobile-enabled products and services in Latin America Number of deployments (cumulative) Pre Health Learning Money Others The social and economic impact of mobile in Latin America 49

427 THE MOBILE ECONOMY 2014 LATIN AMERICA A recent survey by the GSMA has shown that mobile operators in the region do want to do more to enable social innovation services in these basic areas 35. Operators have a clear perception of the social needs of the markets where they operate, and these needs in turn are identified as presenting opportunities for new service launches. Operators regard education and health care as among the greatest challenges in the region, but also as presenting some of the strongest opportunities for new service launches. However, a key finding of the survey was that operators feel constrained from doing more in these areas by a lack of proven business models. Policymakers and regulators have a critical role to play in creating the right regulatory frameworks, the funding streams, and the engagement of relevant professionals and institutions necessary for these new business models to be elaborated and the full potential for social innovation services to be realised. Co-ordination between public and private stakeholders has often been deficient in Latin America, whereas some of the region s governments have imposed an onerous tax burden on operators to raise funds for public sector-led social projects, thereby running the risk of stifling innovation by starving operators of development and investment funding. The GSMA is working with regulators, governments, operators, investors and other stakeholders to provide policymakers with precisely this sort of hard data to inform their decisions and contribute to more constructive collaboration with operators. For instance, in June 2014, the GSMA announced it was working with the InterAmerican Development Bank (IDB) on the development of the Centre for Advanced Studies in Broadband Development (CEABAD), based in Managua, Nicaragua. The Centre aims to provide in-depth training to government officials from the region s Spanish-speaking countries on the socio-economic benefits of mobile broadband 36. Operators themselves are attempting to stimulate a constructive dialogue with policymakers and to lay out their vision of the steps that need to be undertaken to fulfil the potential for mobile communications to support basic services. The Brazilian operators federation Telebrasil s plan for an Innovative and Competitive Digital Brazil 2015 to 2022 was discussed earlier in this report. Under the heading Basic services and strategic sectors, Telebrasil lists a number of sectors where it believes telecoms can make an important contribution. In addition to education, health care and agro-industry discussed later in this section these include: energy; sanitation, environmental management and urban mobility; social welfare; and public security. The plan also envisages the creation of an Institutional Council for Communications Policy, which will bring leading players in the communications, media, culture and education sectors together with representatives from ministries across government to co-ordinate policy formation see also 50 The social and economic impact of mobile in Latin America

428 THE MOBILE ECONOMY 2014 LATIN AMERICA mhealth There are a range of different health challenges facing populations and their governments in countries across Latin America. The emergence of a prosperous middle class across the region has led to growing problems around non-communicable and lifestyle diseases. The latter include diabetes and heart disease, with the former now amongst the top five causes of mortality in the region, whilst rates of child obesity have also increased sharply. In more developing countries in the region, such as a number of those in Central America, the challenges revolve around increasing access to basic healthcare and disseminating information that can save lives. Even in some of the more developed economies in the region, there are issues around providing healthcare in more rural and remote regions. There are a number of successful mhealth projects and commercial services throughout the region. However, the extent of the deployments to date is relatively limited and far from meeting the scale of the social challenges across the region. Progress is being made to drive policy formation so as to create the regulatory, institutional and financial frameworks that will allow operators to innovate services and establish business models in this field. The Telebrasil plan discussed previously in the report lays out four action points that need to be addressed by Brazilian policymakers in order for the potential of mhealth services to be realised more fully: More effective scheduling of online medical examinations and consultations Creation of centralised data platforms, and open access for developers to make use of private applications and services Interconnection of transplantation centres Telemedicine as a tool of preventive health care, diagnosis and access to medical specialists. Examples of good practice in mhealth are becoming more widespread across the region, and we highlight a number of examples below: In Brazil, the mpers solution (mobile Personal Emergency Response System) is providing remote, mobile monitoring of patients with heart conditions and high blood pressure. The system comprises a wearable hub that collects data from sensors such as blood pressure monitors, pulse oximeters and electrocardiograms, and transmits it to remote servers for health monitoring and emergency response. The user s location is tracked via GPS or by means of wireless-signal triangulation when indoors. Other features include reminders to take medication, and integration with home-automation systems to automatically trigger an emergency response if the user becomes non-responsive. In Chile and Peru, the MiDoctor (My Doctor) system has been deployed to provide automated monitoring of chronically ill patients via voice calls, SMS and a webbased medical record. The patient receives SMS alerts to make sure that they attend their medical appointments or, if they are not able to attend, that the clinic is notified and can re-use the appointment for another patient. The system is also able to alert physicians of any risk factors, and ensure patients are followed up if they are failing to adhere to their treatment and monitoring programme. The social and economic impact of mobile in Latin America 51

429 THE MOBILE ECONOMY 2014 LATIN AMERICA meducation Many countries in Latin America face a range of challenges in providing education to their youthful and growing populations. The key issues include high drop-out rates, especially in secondary education; high rates of illiteracy; limited access to education; low educational quality; and insufficient teacher training programmes. These problems are especially pronounced in low income urban and rural populations. Increasing the quality of education and broadening access to schooling and skills based learning is therefore a key policy area for governments. Mobile-enabled services have the potential to address many of these challenges and make education more widely available. The scale of the challenge and of the opportunities in the field of education are similar to those in health care, and a joined-up approach on the part of policymakers, the mobile industry and the education sector is just as essential. The Telebrasil plan for digital innovation and competitiveness, discussed above, sets out five action points for the sector that address some of the main challenges: Developing new teaching methods incorporating ICT, promoting an increase in both the efficiency and quality of teaching; Improving teachers skills in ICT as a basic teaching tool, enabling the propagation of active learning programmes for pupils and stimulating the creation of new educational content; Deploying infrastructure and connectivity for educational projects, by building out broadband infrastructure to all schools (for instance, via a Public-Private Partnership model), enabling schools to install ICT equipment and Internet access adequate to their needs; Increasing the range of educational content available and spreading the benefits of knowledge across society through the creation and management of content aligned with the school curriculum, and the development of distance learning platforms; and Creating a programme of digital literacy, thereby enabling the rapid inclusion of the most vulnerable and excluded sections of the population. 52 The social and economic impact of mobile in Latin America

430 THE MOBILE ECONOMY 2014 LATIN AMERICA The Telebrasil plan also recognises that meducation objectives cannot be successfully realised merely through the provision of better broadband connectivity and access to ICT equipment, essential though these are. It is equally important to improve the ICT literacy and skills of both teachers and the most deprived sections of society, who represent one of the main target groups and potential beneficiaries from meducation programmes. Similarly, access to ICT equipment and the internet must be used to provide a stimulus for the development of higher-quality, and more relevant, educational content, along with better teaching and learning methods. These ideas are supported by the findings of a study of the socio-economic effects of broadband connectivity in Latin America published in July 2014 by DIRSI, a regional ICT policy research network. This includes a study of the effects on educational achievement (as measured by standard test results) of programmes designed to equip schools with computers and broadband connectivity, in Brazil, Chile and Peru 37. The investigation found there was relatively little improvement in performance and even some negative effects in cases where deficient teacher training allowed broadband availability to divert attention from the school curriculum. The more positive effects were seen where broadband was used to provide more engaging teaching methods. Examples of this include the use of news stories to teach grammar and maths, or where the availability of broadband motivated older students to continue attending and improved the learning environment by helping to reduce discipline problems in class. There were some indirect positive effects, including for example the digital inclusion of disadvantaged pupils. However, for broadband to really improve educational performance, there is a requirement for better teacher training and methods, as well as more relevant and engaging educational content. Some of the more promising meducation initiatives in the region seek to exploit mobile as a tool to improve the quality of teaching, learning and content: Puentes Educativos (Educational Bridges) is a programme in Chile focused on training teachers to use ICT more effectively to improve the teaching and learning process, and outcomes for 5th and 6th graders in schools in economically vulnerable areas, particularly in maths, science and English. Schools are equipped with Nokia C7 smartphones, and the scheme is sponsored, among others, by Nokia, the educational publisher Pearson and Telefónica. Guia do Estudante (Student Guide) is a service from Telefónica s Brazilian subsidiary Vivo which provides daily test questions to school students via MMS. Students respond and receive their test results via SMS. The tests involve content developed especially for the service by educational publisher Abril and are available in a range of curriculum subjects. 37. The Internet and Poverty: Opening the Black Box (DIRSI, July 2014) The social and economic impact of mobile in Latin America 53

431 THE MOBILE ECONOMY 2014 LATIN AMERICA 54 The social and economic impact of mobile in Latin America

432 THE MOBILE ECONOMY 2014 LATIN AMERICA 3.5 ewaste One consequence of the mobile industry s expansion across the region is a growing problem of ewaste. According to UN estimates, the volume of ewaste generated in Latin America is set to increase from 4,226 kilotons in 2012 to 4,968 kilotons in By 2014, the region will be responsible for 8.6% of the global production of ewaste 38. The mobile industry is of course not the only source of waste electrical and electronic equipment (WEEE), as it is also known, which also includes computer and TV equipment, and general electrical devices, for instance. However, as the average lifecycle of a mobile phone is between three and five years, and given the rapid rate at which basic and feature phones are now being replaced by smartphones across the region, the mobile industry is a major contributor to this issue. The region s mobile operators are working hard to address the problem of ewaste through a range of projects: Developing initiatives to recycle network equipment; Implementing reverse logistics schemes to store and manage unused mobile devices, chargers and batteries; and Conducting awareness campaigns to alert people to the risks of inappropriate ewaste management. Efforts to make progress in this area have been impeded by rigid regulatory regimes, which often make operators solely responsible for the waste generated by mobile devices. This is in contrast to regulatory regimes in other regions that incorporate the principle of extended producer responsibility, whereby the manufacturer is responsible for the entire lifecycle of the product, especially postconsumption. In addition, Latin America lacks much of the infrastructure needed to manage the WEEE it generates, meaning that hazardous materials such as mobile phone batteries need to be shipped abroad to be processed. This both results in considerable extra costs and in a lost economic opportunity from recovering the large volume of precious materials, including silver and gold, contained in electronic devices. The GSMA is working with operators, governments, regulators and the broader mobile ecosystem to promote a more comprehensive and transparent regulatory regime, and a more eco-efficient production and recycling chain, for ewaste in the region. 38. See ewaste in Latin America: The contribution of mobile operators in reducing electronic waste (GSMA, May 2014) The social and economic impact of mobile in Latin America 55

433 e-waste in Latin America Electronic waste growth in Latin America KT = 1Kiloton = 1,000 Tons 3,511KT 4,226KT 4,968KT Source: UNU WEEE in Latin America, selected countries El Salvador Argentina 2015 Panama Ecuador Mexico Brazil Uruguay Chile ,000 1,200 1,400 1,600 1,800 KILOTONS 3-5 years is the average lifespan of a smartphone Source: MIT/StEP Initiative In 2015, Latin America will produce 8.6% OF THE GLOBAL WEEE From 2012 to 2015 Latin America WEEE will increase 17.5%

434 7KG is the average amount of e-waste produced by every person on the planet Source: StEP Initiative What s in a smartphone 45% plastic 10% ceramic 20% copper 20% gold, aluminum and other metals 5% non-metals Source: TIM Recarregue o Planeta A ton of gold ore contains 5g of gold whereas a ton of mobile phones contains 400g OF GOLD US$21M a year in gold, silver, and other metals can be extracted from WEEE Source: e-waste Academy (EWAM) Results of the recycling process BATTERIES Composition: cobalt / nickel copper Batteries, magnets, stainless steel, metal oxides, ink pigment, and salts. COMPONENTS Composition: gold / palladium copper Jewelry, musical instruments, electronic equipment PLASTIC Composition: plastic Traffic cones, chairs, tyres, pet food containers and other everyday objects. Latin America mobile operators willingly develop: Reverse logistic schemes Awareness campaigns Recycling programs Standards adoption

435 THE MOBILE ECONOMY 2014 LATIN AMERICA 3.6 Delivering financial inclusion across Latin America Roughly 60% of adults in Latin America remain unbanked, ranging from 86% in Nicaragua and El Salvador, to 44% in Brazil 39. This equates to around 170 million people who are still denied access to basic financial services. The formal financial services sector has, in many countries, failed to serve lower income segments, largely because the economics and demographics of bricks and mortar banking do not work for most private players in the region. This is particularly true of the populations in rural areas, as traditional banking infrastructure is often restricted to urban centres. With increasing levels of mobile penetration and an existing retail infrastructure, mobile money services provide a clear opportunity to expand financial access. Greater financial inclusion would not only yield economic benefits for the region, but also enhance the social welfare of millions of Latin Americans currently unbanked or under-banked. While there are a growing number of mobile money deployments in Latin America today, the rate of adoption has tended to lag behind that seen in other emerging markets such as Africa and parts of Asia. The GSMA s 2013 State of the Industry report indicates that as of December 2013, Latin America was home to just 13% of the world s live mobile money deployments, but more encouragingly had almost 20% of the planned deployments. Percentage of live and planned MMU deployments by region 51.7% 41.6% 1.4% 7.1% 6.0% 13.3% 11.5% 8.8% 13.3% 18.6% 16.0% 10.6% Europe and Central Asia MENA East Asia and Pacific LatAm South Asia Sub-Saharan Africa Percentage of live deployments Percentage of planned deployments 39. Global Financial Inclusion (Global Findex) Database The social and economic impact of mobile in Latin America

436 THE MOBILE ECONOMY 2014 LATIN AMERICA As at June 2014, there were 36 live mobile money deployments across the region, compared with 27 one year earlier. In addition, there were 19 planned deployments at June While Latin America has seen some successful mobile money deployments, the full potential of mobile money remains untapped in part due to regulatory barriers in a number of countries. However, progress has been made over the last year or more in elaborating the regulatory and legal frameworks to be applied to mobile money services in countries across the region, and this is enabling a variety of business models and commercial partnerships to be developed, and new services launched. Consumer concerns over security are also likely to be allayed as the legal responsibilities and liabilities of players in the mobile money ecosystem are clarified. Three broad regulatory, legal and commercial approaches to mobile money are emerging: Mobile as a channel for traditional financial services, for banked consumers. Countries with traditional financial regulatory frameworks, allowing banks to use the mobile network as one of the channels for their already financially included population. In these countries, like Argentina and Venezuela, mobile financial services are provided directly by banks using mainly advanced apps for smartphones. Mobile Money for the Unbanked (MMU) through existing banking or remittance laws. Countries with financial inclusion regulation and provisions (such as flexible identity verification requirements and the use of alternative distribution channels like mobile, the internet and agents) but without a specific nonbank financial licence that allows new players to enter the market. This is the situation of Guatemala, Chile and, until recently, Colombia, for example. In these countries, the mobile money deployments are usually either bank-led or joint ventures between banks and mobile operators, but always using the bank licence. In some countries, like Guatemala, mobile operators offer mobile over-the-counter, person-to-person funds transfer under non-financial licences (usually a domestic remittance licence). The creation of new legal entities for a dedicated payments company or e-money issuer. Countries like Mexico, Colombia, Peru, Bolivia, Uruguay, Paraguay and Brazil have recently adopted modifications to their legal frameworks in order to allow the entry of new non-bank (light financial) institutions to the financial transaction market, including, in some instances, the ability to issue e-money. The level of requirements for these institutions varies among these countries. The minimum required capital for these financial licences is normally under US$1 million, except in Colombia (US$3 million) and Mexico (US$14 million). The social and economic impact of mobile in Latin America 59

437 THE MOBILE ECONOMY 2014 LATIN AMERICA In Mexico, the new non-bank entities, called niche banks, are required to be fully integrated with the existing financial infrastructure (such as payment cards and ATMs) and are subject to the full suite of banking regulation and law. Transfer, a joint venture between América Móvil subsidiary Telcel and the Banamex/Citibank and Inbursa banks, operates under the banks existing licences. Transfer offers deposits (for example at correspondence bank outlets), cardless withdrawals from ATMs, person-to-person transfers, prepaid top-ups and an optional companion card. As at September 2013, there were reportedly 2.17 million active mobile banking accounts of this type in Mexico, representing around 2.5% of the adult population 40. Increasing regulatory clarity has contributed to a growing wave of e-money service launches throughout the region in 2013 and After the mobile payments bill was published in Brazil in May 2013, for instance, three mobile money schemes involving mobile operators were launched: Zuum a partnership between Vivo (Telefónica) and MasterCard; Oi Carteira ; and Meu Dinheiro Claro developed by Claro in partnership with Banco Bradesco. The fourth national mobile operator, TIM Brasil, has also publicly announced plans to launch a mobile wallet in partnership with the government-owned bank Caixa Econômica Federal and MasterCard. New mobile money services are also expected to launch in Peru in the near future, with the mobile operators as active participants in the ecosystem. An initiative by the Peruvian Bankers Association (ASBANC) to offer a shared e-money platform captured headlines in 2014, and a pilot is expected to be launched in the first half of Regulatory Approaches to Mobile Financial Services in Latin America (Alliance for Financial Inclusion, August 2014) 60 The social and economic impact of mobile in Latin America

438 THE MOBILE ECONOMY 2014 LATIN AMERICA In Colombia, the government recently passed legislation to approve e-money issuers, and further regulation is expected to implement the financial inclusion law. In the interim, both Claro and Móvistar are offering mobile money services and do so effectively as a channel for banks or credit card companies in a similar way to Mexico. The Colombian bank Bancolombia acquired an MVNO ( Ahorro a la Mano ) to offer mobile financial services independently of the country s mobile operators. One successful operator-led mobile money service is Millicom s Tigo Money, which is active in five Latin American markets where Millicom operates networks under the Tigo brand (Bolivia, El Salvador, Guatemala, Honduras and Paraguay). The service can be used for remittances, person-to-person transfers, cash withdrawals, payments and prepaid top-ups. Any mobile phone user can access the service, which functions via SMS. In Bolivia, it was reported in October 2014 that Tigo Money is responsible for money flows of nearly US$4 million per month, and has around 700,000 customers 41. Other countries are likely to adopt regulations enabling mobile money specific licences in the near future. Such is the case of El Salvador: a Financial Inclusion bill has been drafted and presented to Congress. This will bring legal stability to the mobile money services already deployed, promoting its expansion. Among the major economies, Brazil, Colombia and Mexico have adopted financial inclusion strategies; and Uruguay is also preparing a Financial Inclusion Act. One major challenge for MMU services in the region going forward is to make the different national payment and e-money platforms of banks and non-banks interoperable, as this will greatly enhance the market reach and hence take-up of vital services such as international remittances, funds transfers, and m-payment. The way to achieve more interoperability will become clearer as the regulatory frameworks are established in each country, and as the most effective business models and players emerge. An important lesson from other markets where mobile money is growing is that the precondition for developing a successful regulatory framework is to establish an open dialogue and consultative process between policymakers, regulators and the private sector. Many regulators will agree that by establishing open channels of communication with mobile money providers, they can design regulation that embodies a financial inclusion objective whilst also providing effective oversight of the mobile money sector. Considerable progress has now been made in the region towards creating policy frameworks and business models that allow both bank and non-bank mobile money providers, including mobile network operators, into the market. Anecdotal evidence, commercial lessons and international regulatory principles all defend opening the market to providers with different value propositions. Equally, evolving mobile money regulation and legislation in the region are incorporating internationally established principles and compliance systems for safeguarding customer money, and enhancing efforts to combat money laundering and the financing of terrorism (AML/CFT). The market is growing in confidence and in scale. In one industry survey in 2014, 62% of those polled expected mobile payment to become a mass-market service in the region within two to five years 42. It is hard to overstate the social and economic benefits for the region as a whole that could be brought about by the vastly expanded levels of financial inclusion that would be enabled by a mobile-money mass market. This has the potential to get the whole economy of the region going again, by providing access to formal financial services for disadvantaged and marginalised sections of society for the first time, and thereby creating new opportunities for commerce, investment and job creation. This will also have a knock-on effect in terms of digital inclusion, as the socio-economic benefits of mobile money will provide a major incentive to subscribe to a mobile service (along with the actual means to do so) for the many millions in the region for whom the cost of mobile communications and broadband in particular has hitherto represented a barrier. And at the same time, the value that mobile money provides for mobile users will help strengthen subscriber loyalty. Mobile money should therefore be an important win-win, for financial inclusion and mobile adoption alike Enfoque Latinoamericano: Cuál es el próximo paso para los pagos móviles?, Survey of attendees of the Congreso Latino Americano de Pagos Móviles (Open Mobile Media, 2014) The social and economic impact of mobile in Latin America 61

439 THE MOBILE ECONOMY 2014 LATIN AMERICA 4 Mobile policy and the role of the industry There is no doubt that the market and regulatory landscape is changing in Latin America. New telecom laws have been passed in Mexico, as well as the Internet Civil Rights Framework Law in Brazil, while new bills are currently under discussion in Argentina, Ecuador and Perú. All of these developments are likely to significantly transform mobile market conditions. Consolidation is also now a matter for discussion in Brazil, Mexico and Chile. Data-hungry consumers are putting pressure on networks and impacting their quality of service, while regulators are sometimes addressing these demands with a political rather than technical approach. Regulatory intervention has increased across Latin America creating questions about the future of the industry. The convergence of mobile with the IP world is also creating challenges for the sustainability of the operators business model. With reduced margins, increased competition from adjacent industries and higher regulatory pressure, sustaining growth will very much depend on having the right regulatory environment. Having a transparent, predictable, consultative and well-aligned regulatory regime will be key for promoting the right incentives that would unleash future investments in mobile telecom networks and services. 62 Mobile policy and the role of the industry

440 THE MOBILE ECONOMY 2014 LATIN AMERICA A clear roadmap for spectrum allocation as well as the removal of municipal obstacles for antenna deployment will be key to satisfying the increasing demand from end users for mobile broadband. Despite these challenges, there are opportunities ahead for mobile operators to expand their businesses and to further increase their social and economic contribution. A key factor in delivering this potential, and further extending the reach new services and solutions such as M2M services, mobile money, mobile commerce, will be reducing current levels of taxation. The Brazilian case of reduced M2M taxes can be a good example for other countries to follow. Latin American operators have demonstrated their commitment to the local market in which they operate by sustaining high levels of investment and more recently by implementing new joint consumer-centric initiatives under the GSMA s We Care campaign focused in addressing social needs. 4.1 Spectrum: A key enabler to promote investment and foster innovation Spectrum assigned for mobile services in Latin America have reached almost 300MHz on average during 2014, an increase of around 25% compared to This increase is as a result of the licensing of the so-called 4G spectrum bands (AWS MHz, 2.6GHz and 700MHz) in the whole region. This means that on average 25% of the spectrum currently in the hands of the regional mobile operators has been allocated for 4G. However some countries have not specifically allocated 4G spectrum as yet. Among the countries with a higher proportion of 4G spectrum as a share of the total spectrum allocated, Argentina, Colombia, Chile and Ecuador stand out with over 40% of the total assigned spectrum in the 4G spectrum bands. Bolivia, Honduras, Mexico, Peru and Brazil are next with over 20%. Countries with less than 15% include Paraguay, Dominican Republic, Uruguay and Costa Rica. Finally, the countries yet to award 4G spectrum include Venezuela (that has an auction scheduled for end-2014), Guatemala and El Salvador. It is worth mentioning there are some operators that decided to launch LTE services in 1900MHz band, such as in Costa Rica, Paraguay and Venezuela. There were 47 LTE networks in 19 Latin American countries as of October AWS consolidated as the main band for 4G services Governments across the region have understood that spectrum is a key factor to sustaining and promoting investment in infrastructure and ICT services, while at the same time fostering innovation and bridging the digital divide. Over the course of 2014, spectrum licencing has mainly focused on the 4G bands discussed previously. Some countries also made movements towards licencing the 2.6GHz band. In May 2014, the GSMA launched a study that estimated the economic benefit of licensing the AWS band for mobile would equate to an additional US$53 billion in Latin American countries where it has not been entirely licensed yet, resulting from the increased uptake of mobile broadband this would enable 43. The countries in question include Argentina, Ecuador, El Salvador, Guatemala, Nicaragua, Panama and Paraguay: several of the countries with the lowest overall Internet penetration rates in the region. During the first half of 2014, Bolivia and the Dominican Republic assigned part of the AWS band to mobile operators. By the end of 2014, it is expected to see AWS spectrum licensed in Argentina, Venezuela and Ecuador, which will ensure AWS position as the 4G band in the region, with 15 countries using the band to deliver fast and reliable mobile broadband services. 43. The AWS Situation in the Americas (Convergencia Research, May 2014): Mobile policy and the role of the industry 63

441 THE MOBILE ECONOMY 2014 LATIN AMERICA In addition to its economic impact, the aforementioned study highlights that licensing this band is important in other key aspects: Innovation, as it fosters LTE deployments in the region. LTE is the preferred technology in almost all the countries where this band is being used. Network deployment, as it facilitates relatively quick 4G deployment due to the fact that LTE ecosystem in this band is robust for devices and networks. Quality of service, as the band will help to move heavy users out from the traditional 2G and 3G bands and accommodate them in the new 4G network. By consequence, this move will also benefit 2G and 3G users. New applications ecosystem, as many governments are promoting local entrepreneurs, and the deployment of LTE services enables a whole new applications ecosystem in many areas such as mhealth, mgovernment, mautomotive and Smart Cities, among others. New marketing strategies, as this band could help operators drive their commercial strategies to offer innovative plans to specific niches, including the bottom of the pyramid. Regional harmonisation of the best mix of capacity and coverage bands, AWS as a capacity band, together with 700MHz as a coverage band, could be the perfect combination for universalising 4G deployments in the region as 850/1900MHz bands did with 2G and 3G services. The report also highlights that in spite of the governmental efforts to introduce new competitors when licensing this band, there is no direct relationship between new entrants and competition. That being the case, incumbent operators should always be allowed to compete for new spectrum. In reality, secondary spectrum markets could help much more than reserving spectrum for new entrants. The AWS assignment process, as in any other band, needs to be carefully planned, and consulted with the industry so as to maximise its benefits and minimise its risks. On average it takes at least one year from the decision by the government to license the band and its assignment, and then another year to see 4G deployments in the country. This shows the importance to regulators of developing spectrum roadmaps when planning their spectrum strategy. Source: Convergencia Research Direct and indirect economic benefits of licensing AWS band in 2014 (US$ M) Paraguay US$333 US$646 Panama US$1,815 US$1,727 US$53 Billion Nicaragua Guatemala El Salvador US$1,017 US$3,152 US$1,716 US$1,305 US$2,810 US$1,495 Ecuador US$6,294 US$5,882 Argentina US$14,569 US$9,913 Indirect benefits US$23.3 Billion Direct benefits US$29.5 Billion Direct benefits Indirect benefits 64 Mobile policy and the role of the industry

442 THE MOBILE ECONOMY 2014 LATIN AMERICA The year of Digital Dividend Starting with auctions in Chile in February and ending up with Brazil in September and Argentina in October, a number of Latin American governments have made the first moves towards seeing the 700MHz band (Digital Dividend) being used to deliver high-speed mobile broadband services to their populations. Incumbents in Chile, Panama, Jamaica, Brazil and Argentina were all able to participate in the licensing processes. In addition, most Latin American countries have confirmed their decision to adopt the APT band plan. The Chilean government has been an early mover when it decided to licence 60MHz (out of the 90MHz available) of the Digital Dividend band to mobile services through a beauty contest model. Subsequently, Panama assigned 40MHz to two operators as part of its license renewal process and Jamaica then concluded a long awaited auction of 24MHz. In September of this year, Brazil concluded the biggest 700MHz auction in the region, assigning 60MHz on a national basis to three national incumbents and the rest divided in regions. Total bids for the 700MHz auction in Brazil were US$2.39 billion (BRL5.8 billion), excluding an extra cost of US$1.47 billion (BRL3.6 billion) to be paid by the operators for the clearing of the band. Claro, TIM and Vivo acquired one lot of 2x10MHz each, while one lot remained unsold. Oi did not participate. Finally, Argentina auctioned the whole 90MHz of the Digital Dividend band together with the 90MHz in the AWS band in October also collecting US$2.3 billion. This trend indicates that the 700MHz band will be licensed in the remaining Latin American countries between 2015 and 2016, thus consolidating its role as the key band for universalising internet services in the region. In a majority of countries, there is no need to wait for the analogue TV switch over, which expected to happen only after 2018, for allocating digital dividend spectrum. Source: GSMA Latin America 4G spectrum gaining traction (Total MHz licensed to mobile operators by country) 634 2G/3G 4G Brazil Chile Colombia Argentina Uruguay Peru Dom. Republic Nicaragua Paraguay Bolivia Costa Rica Honduras Venezuela* Mexico Guatemala El Salvador *Venezuela will license up to 160MHz in AWS and 2.6GHz by year-end 2014 Mobile policy and the role of the industry 65

443 THE MOBILE ECONOMY 2014 LATIN AMERICA Spectrum licensing conditions should be carefully drafted so as to avoid hindering investment Although governments and regulators are making good progress in terms of licensing spectrum to mobile services in the Latin America, there is a wide concern on the conditions being imposed in these processes. Governments are setting very high reserve prices, challenging coverage conditions and short timeframes to deploy services. Some countries are even including the handing in of tablets, smartphones and laptops as part of their obligations for acquiring spectrum. To ensure a successful licensing process, governments need to consider that trying to maximise revenue and increasing the burden of the conditions imposed produces a negative impact in the investment capabilities of the market. The GSMA recently commissioned Ovum to analyse the licensing obligations being imposed across the region and to assess how each condition could impact on a range of factors. These include for example consumer service prices, the narrowing of the digital gap, and investments incentives. Some preliminary conclusions of this report indicate: High spectrum prices can represent a negative net present value for the companies interested in participating, resulting in a failure to attract bidders. A 10% increase in rural coverage obligations can impact up to 25% of project s value for operators, due to high capital expenditure investments. Each additional year in the license term increases the value of the project by up to 25%. Allowing RAN sharing can increase the value of the project by up to 30%. It is key that governments and regulators understand and take into account the industry and investors perspective when licensing spectrum in order to fully achieve the benefits of using this scarce resource in favour of users, especially of those who are still not enjoying the benefits of internet services. 66 Mobile policy and the role of the industry

444 THE MOBILE ECONOMY 2014 LATIN AMERICA Network competition is preferred above single wholesale networks Policymakers in a number of countries have been considering establishing a single wholesale network (SWN) instead of relying on competing mobile networks to deliver 4G mobile broadband services, but there are currently no nationwide initiatives in action. However, the GSMA believes that moving away from traditional mobile network competition models towards SWNs can stifle innovation, restrict investment and take-up of mobile broadband services and ultimately be against consumer interests. A recent report commissioned by t