Poverty and Mobile Telephony in Latin America and the Caribbean

Mobile Opportunities: Poverty and Mobile Telephony in Latin America and the Caribbean

Hernán Galperin Judith Mariscal DIRSI

November 2007 Poverty and Mobile Telephony in Latin America and the Caribbean

Mobile Opportunities: Poverty and Mobile Telephony in Latin America and the Caribbean Hernán Galperin Judith Mariscal DIRSI

November 2007

This work was carried out with the financial support from a grant given to the Instituto de Estudios Peruanos (IEP) from the International Development Research Centre, Ottawa, Canada. Abstract Access to telephony for low-income groups is largely based on different strategies of mobile telephony usage. The main goal of this research project is to explore the strategies employed by the poor in Latin America and the Caribbean to access and use mobile telephony services, as well as to identify the major market and regulatory barriers for increased penetration and usage. More generally, it seeks to contribute to the discussion on how access to mobile telephony contributes to improving the livelihoods of the poor –what we call mobile opportunities.

Our results show that mobile telephony is highly valued by the poor as a tool for strengthening social ties and for increased personal security, and that it is beginning to prove useful for enhancing business and employment opportunities. Overall, the survey results suggest that the acquisition of mobile phones by the poor has an economic impact reflected mainly in improved social capital variables such as the strengthening of trust networks and better coordination of informal job markets. These findings reveal the continued need to develop innovative business models that extend the market frontier for mobile telephony. They also highlight the urgent need to rethink public policies that are premised on the mobile phone as a luxury good. Clearly, for the poor, mobile telephony has long been the most cost-effective and accessible alternative.

We acknowledge the valuable support of Andrea Molinari and Fernando Ramírez in the research process. We would also like to thank the team leaders who carried out research in each country: Antonio Botelho (Brazil), Luis Gutiérrez (Colombia), Hopeton Dunn (Jamaica), Roxana Barrantes (Peru), and Kim Mallalieu (Trinidad and Tobago). Poverty and Mobile Telephony in Latin America and the Caribbean 3

Introduction Methodology Access to communication services and to telephony Over 7,000 face-to-face interviews were conducted in particular has long been recognized as an with individuals aged 13 to 70 residing in low- important development input. In Latin America income households in the following countries: and the Caribbean, telephony penetration has Argentina, Brazil, Colombia, Jamaica, Mexico, grown exponentially in recent years, driven to a Peru, and Trinidad and Tobago. A probabilistic significant degree by mobile services. This has been sample was drawn using maps from existing particularly relevant for the poor, who for a variety georeferenced data provided by the national of reasons have had limited access to traditional fixed statistics office of each country. Individual telephony services. For those at the bottom of the respondents were randomly selected from each income pyramid, access to telephony is largely based household. The goal was to obtain a representative on different strategies for using mobile telephony. (and statistically independent) sample of low- Yet empirical studies aimed at understanding the income residents of urban areas in each country, patterns of mobile phone use by the poor, and of although in one case (Jamaica), semi-rural areas communication services in general, are rare. were also included. Fieldwork was conducted between April and June 2007. The main goal of this research project isto understand the strategies employed by the poor in Latin America and the Caribbean to access and use mobile telephony services, as well as to identify the major market and regulatory barriers for increased penetration and usage. The project also attempts to analyze the untapped commercial and public policy opportunities associated with increased mobile penetration among low-income groups (the so-called bottom of the pyramid). More generally, it seeks to contribute to the discussion on how access to mobile telephony contributes to improving the livelihoods of the poor –what we call mobile opportunities. Key results Our results show that the exponential growth in the mobile Figure 2 telephony market in Latin America and the Caribbean has Prepaid plans (% of total) had a significant impact on telephony access opportunities for the poor. With the exception of Mexico, the majority of respondents in the countries studied had used a mobile Jamaica 98 phone in the past three months, regardless of whether or not they actually owned one (see Figure 1). Interestingly, Brazil 96 the level of shared ownership was relatively low: in most cases, users own their own handset and service. The notable T&T 96 exceptions are Colombia and Peru, where a healthy service Peru 96 resale market in urban areas (with very competitive tariffs) reduces ownership incentives. Mexico 92

Figure 1 Colombia 90 Mobile usage and ownership (% of total) Argentina 74

020406080 100 Jamaica 94 90 Colombia 89 63 T&T 86 This preference for prepaid plans is based on two factors, as 83 illustrated in Figure 3. The main motive is spending control: Argentina 70 61 users value the ability to purchase credit when they have cash in hand, rather than committing to a fixed monthly charge. Peru 60 37 The other factor is cost: users perceive prepaid plans to be Brazil 53 cheaper than post-paid plans. This counterintuitive result 42 must be interpreted carefully. On the one hand, there seems Mexico 37 30 to be information problems that prevent users from making 0 20 40 60 80 100 accurate comparisons of per-minute costs between prepaid and post-paid plans. On the other hand, users are simply Owners Users referring to total monthly costs based on their particular usage basket, which tends to be well below the post-paid service packages offered by the operators. Our results also confirm the importance of prepaid service models for a population with fluctuating incomes and limited Figure 3 insertion in the formal economy. Overall, low-income Reasons for opting for prepaid mobile service (in %) mobile users prefer prepaid plans (see Figure 2). However, in a developed mobile telephony market like Argentina, the solid performance of mixed plans (which combine a low 61 T&T 29 fixed monthly fee with top-ups for extra credit) illustrate the Brazil 52 potential of commercial innovations specifically targeting 38 47 the poor. Argentina 35 Mexico 46 26 Colombia 41 31 Jamaica 41 32 Peru 34 33 01020304050 60 70 I can control my mobile spending better It is cheaper Poverty and Mobile Telephony in Latin America and the Caribbean 5

Interestingly, most users think mobile services are neither cheap nor expensive, although the percentage of users who consider the service expensive tends to be higher in markets with lower ownership rates, such as Mexico, Peru and most notably, Brazil (see Figure 4).

Figure 4 Mobile service cost perception (in %)

18 Argentina 68 14 26 T&T 57 17 35 Peru 55 10 21 Colombia 51 28 38 Mexico 45 16 34 Jamaica 35 31 54 Brazil 12 34 0 10 20 30 40 50 60 70 80 Very cheap or cheap Neither cheap nor expensive Expensive or very expensive

For the poor, service affordability remains a key barrier for increased adoption of basic as do value-added services. This is confirmed by responses from non-users, who cite cost as the main reason for non-adoption of mobile telephony, as shown in Figure 5.

Figure 5 Reasons for not owning a mobile phone (in %)

Mexico 56 20 Peru 56 28 Brazil 52 32 Jamaica 49 23 Argentina 46 32 T&T 40 34 Colombia 38 21 01020 30 40 50 60

I don’t think havingamobile of my own is necessary I can’t afford it Overall, we find that the poor highly value mobile services. Figure 7 This is not surprising given the lack of alternatives, Total outgoing calls per week (average and median) particularly traditional fixed telephony services. Most respondents are willing to spend significant amounts on 30 25 handsets, with averages ranging from USD 53 in Peru to 25 22 USD 68 in Argentina. In fact, our results show that the 17 20 informal market for second-hand handsets is relatively small 13 15 11 (see Figure 6). Rather, the real driver of increased adoption 10 7 10 7 7 10 7 is the trickle-down effect of handset replacement from the 5 5 5 top of the income pyramid: in the more developed markets 5 such as Argentina, over a third of respondents had obtained 0 T&T Jamaica Colombia Brazil Argentina Peru Mexico handsets as a gift from an acquaintance or family member. Average Median Figure 6 Owners who bought a new handset (in %) A surprising result is that the number of incoming calls is only slightly higher than that of outgoing calls, with the sole exception of Jamaica. Contrary to the common perception Jamaica 98 that poor users receive many calls but make very few, we find that volumes of incoming and outgoing calls tend to be T&T 97 closely related, as Figure 8 demonstrates. Peru 86 Figure 8 Mexico 86 Incoming calls over total calls (in %) Argentina 83 Colombia 80 Brazil 76 Jamaica 67

020406080 100 Peru 55 T&T 55 It is well known that existing tariff levels, combined with Argentina 52 the calling party pays (CPP) system, lead low-income users to limit use of mobile phones for outgoing calls. Our results Colombia 51 show that most users in Latin America average less than one Mexico 50 call per day (see medians in Figure 7), whereas outgoing call levels are higher in Caribbean countries, where tariffs are Brazil 50 more affordable. These results also explain the preference for 010203040506070 80 prepaid plans, since most operators do not offer fixed-rate plans for such low volumes of outgoing calls. In addition to making a very limited number of outgoing calls, the poor utilize other cost-control strategies that involve so-called ‘beeping,’ or simply not making outgoing calls for a period of time. In fact, in most of the countries studied, over a third of respondents had not made a single outgoing call in the week preceding the survey (see Figure 9). Poverty and Mobile Telephony in Latin America and the Caribbean 7

Figure 9 Most common cost-reduction strategies (% of users)

70 57 57 61 60 50 51 47 50 44 40 33 32 25 26 30 23 16 21 20

10

0 Argentina Brazil Colombia Jamaica Mexico Peru T&T Only used the mobile to receive calls Beeping

Our results reveal that the poor represent a significant market for mobile operators, with higher than expected average per capita expenditure. As shown in Figure 10, average monthly spending on mobile telephony services in Latin America ranges from USD 6.1 in Colombia to USD 15.7 in Mexico, while in Caribbean nations expenditure levels are considerably higher.

Figure 10 Average monthly spending on mobile telephony (USD)

18 16 16 14 12 11 12 10 8 7 6 6 4 2 0 Mexico Brazil Argentina Peru Colombia While accurate income data are difficult to obtain in this Figure 12 type of survey, our estimates reveal that average spending If the cost of using a mobile phone were halved… (in %) on mobile telephony as a percentage of income ranges from 3.2% in Brazil to 6.3% in Argentina, however, in some cases, users may spend well above 10% of their monthly income Brazil 84 on mobile telephony services. We also found that demand Argentina 83 for mobile services at the bottom of the pyramid appears to Mexico 83 be rather inelastic with respect to tariff variations, although price elasticity tends to rise with income levels. In general, Peru 82 poor users highly value the few calls they make or receive and Colombia 78 would not significantly change usage patterns in the event of T&T 77 price increases. However, for some users, there appears to be a threshold point after which they would simply stop using Jamaica 41 their mobile phones (see Figure 11). 01020304050607080 90 I would increase my phone usage Figure 11 If the cost of using a mobile phone doubled… (in %)

Although lower than expected, income elasticity of mobile 36 T&T 9 telephony usage also appears significant in some cases, such Argentina 34 as in Peru (see Figure 13). 9 Mexico 29 23 Figure 13 Brazil 24 8 If your income doubled… (in %) Jamaica 22 11 Peru 21 9 Peru 68 Colombia 20 16 Colombia 47 0510 15 20 25 30 35 40 Jamaica 43 I would not change my phone usage I would stop using my phone Brazil 42 Argentina 37

At the same time, our results confirm that current tariff T&T 34 levels represent a significant deterrent to use: most users would significantly increase their consumption of mobile Mexico 30 telephony services if the costs were reduced, as shown in 0 10 20 30 40 50 60 70 80 Figure 12. I would increase my phone usage

Meanwhile, despite increased adoption, users are not taking full advantage of the services enabled by the mobile platform. Text messaging (SMS) is the only service beyond voice that is rapidly being adopted across the region. In the Poverty and Mobile Telephony in Latin America and the Caribbean 9

more developed markets such as Jamaica and Argentina, the large majority of low-income mobile phone users have used SMS services. However, factors such as literacy levels seem to have deterred adoption in other countries, notably Brazil (see Figure 14).

Figure 14 SMS use among mobile phone users (in %)

Argentina 91

Jamaica 67

Mexico 53

T&T 46

Peru 45

Colombia 44

Brazil 36

0102030405060708090 100

As shown in Figure 15, the main reason mentioned for not using SMS services is lack of knowledge, which is not surprising given that most respondents are relatively new users (two years or less). In fact, our results suggest that adoption of services beyond voice increases over time, as users advance along the technological learning curve.

Figure 15 Main reason for not using SMS services

T&T 57 Peru 53 Jamaica 42 Mexico 39

Brazil 38 Argentina 37

Colombia 34 0 10 20 30 40 50 60 I don’t know how to use SMS In most markets, the current structure of tariffs creates Figure 17 incentives for intensive use of SMS as a cost-control Destination of outgoing calls (in %) strategy. As shown in Figure 16, the volume of SMS usage is significantly higher than the volume of voice calls, 100 particularly in markets with high SMS adoption rates, such as Argentina. 80 60 Figure 16 Total outgoing SMS per week (average and median) 40 20

40 0 33 Argentina Brazil Colombia Mexico Peru T&T 35 30 Workplace Friends Relatives in the country Others 26 25 20 Figure 18 20 14 13 Origin of incoming calls (in %) 15 10 10 8 9 8 8 5 5 5 5 4 100 0 80 Argentina Mexico T&T Jamaica Peru Colombia Brazil 60 Average Median 40

Aside from text messaging and voice services, low-income 20 users make little use of mobile telephony services. In the 0 more developed mobile markets such as Jamaica and Trinidad Argentina Brazil Colombia Mexico Peru T&T and Tobago, there is some usage related to downloading ringtones Workplace Friends Relatives in the country Others and participating in radio/TV games, but the use of more sophisticated services such as banking and government services is practically non-existent. This represents an In other words, as illustrated by Figure 19, the main interesting opportunity for the delivery of information and value associated with mobile phones is the strengthening transaction services by the government as well as market of existing ties, although increased personal security (in actors, given the relatively high level of penetration of this emergencies, for example) is also frequently mentioned transaction platform among the poor. as a major benefit. Increased business opportunities also appear to be an important factor, particularly for longer- As mentioned, the main perceived benefit of mobile use term users. In the case of Mexico and Peru, it is interesting among the poor is improved communication with family to note that those who use mobile phones for work-related and friends. This is consistent with the fact that most mobile reasons tend to have higher call volumes. Overall, our results calls are made to or received from friends and family, followed suggest that the economic impact of mobile adoption by by work-related calls, as illustrated by Figures 17 and 18. the poor is mediated by social capital variables such as the strengthening of trust networks and improved coordination of informal job markets. Poverty and Mobile Telephony in Latin America and the Caribbean 11

Figure 19 Perceived benefits associated with mobile use (in %)

Argentina Brazil

Family 71 Family 54

Friends 69 Friends 45

Emergencies 47 Emergencies 46

Work 43 Work 31 0 20 40 60 80 0 20 40 60 80 Colombia Jamaica

Family 65 Family 65

Friends 42 Friends 64

Emergencies 37 Emergencies 44

Work 43 Work 38 0 20 40 60 80 0 20 40 60 80 Mexico Peru

Family 48 Family 52

Friends 29 Friends 30

Emergencies 40 Emergencies 14

Work 27 Work 26 0 20 40 60 80 0 20 40 60 80

Trinidad and Tobago

Family 73

Friends 73

Emergencies 56

Work 49

0 20 40 60 80 Finally, our results revealed that beyond mobile networks, the poor have limited access to other communication platforms (see Figure 20). With the notable exception of Colombia, most respondents lack fixed telephony services in their homes, confirming the unequal distribution of traditional fixed networks as compared with new mobile telephony networks.

Figure 20 Mobile and fixed telephony ownership (in %)

100 96 96 90 87 81 80 80 70 70 63 62 60 50 37 43 39 40 38 31 30 15 20 10 0 T&T Jamaica Argentina Brazil Mexico Colombia Peru

Mobile owners Fixed owners

A similar pattern is found with respect to internet access. With the exception of Peru, the poor seem relatively uninterested in the internet, although usage tends to rise among the younger population. Interestingly, the factors that explain the lack of internet usage by the poor are only partly related to affordability or infrastructure availability, since in most urban areas access is widespread via public internet centres (such as cyber cafés), which offer relatively inexpensive hourly access (at least compared with per- minute mobile tariffs). Our results reveal that the poor simply perceive no benefit in using the internet, with the exception of a small percentage of youth who consider it useful for school purposes and for keeping in touch with friends. This represents a red flag for policymakers who promote universal access policies and who frequently focus on internet/PC shared-access programs. For the poor, the mobile phone has become a much more important and familiar platform than the internet. Poverty and Mobile Telephony in Latin America and the Caribbean 13

Our results also show that public phones –often overshadowed by other priorities in universal access programs such as telecentres– continue to play a significant role for the poor, often as a complement to mobile services (see Figure 21). Cost, convenience and simply the lack of other options are the reasons most often mentioned by the poor for the continued use of public phones (see Figure 22), despite the increased availability of and access to mobile telephony.

Figure 21 Use of public telephony in the last month (in %)

Peru 70 33 Brazil 64 3 Mexico 55 6 Colombia 49 33 33 Argentina 46

T&T 9 3 0 10 20 30 40 50 60 70 80 Public telephones or non-specialized commercial establishments Commercial establishments specialized in telephone service

Figure 22 Reason for using a public payphone (in %)

100

80

60

40

20

0 Argentina Brazil Colombia Mexico Peru T&T

Lower cost Easy access Privacy Lack of other options Other

Poverty and Mobile Telephony in Latin America and the Caribbean 15

Conclusion The goal of this report was to contribute tothe Since affordability is the most significant barrier discussion on how access to mobile telephony to extending the reach of mobile services, as contributes to improving the livelihoods of the poor well as the range of services offered to the poor, in Latin America and the Caribbean –what we call priority should be placed on policies aimed mobile opportunities. Our results show that mobile at reducing tariffs. Enhancing competition telephony is highly valued by the poor as a tool for through increased spectrum allocation, reducing strengthening social ties and for increased personal taxation levels, and implementing number security, and that it is beginning to prove useful for portability are among the initiatives worth enhancing business and employment opportunities. considering. Tariff reductions need not penalize Overall, the survey results suggest that the acquisition operators. Our results show ample room for of mobile phones by the poor has an economic win-win initiatives that increase overall traffic impact reflected mainly in improved social capital and create new commercial opportunities for variables such as the strengthening of trust networks operators and third-party service providers, and better coordination of informal job markets. ranging from simple information to more complex transaction services. Given their limited access to traditional fixed telephony, the poor attribute a significant Along these lines, our results also reveal that users improvement in quality of life to mobile access. are rarely taking full advantage of the services Demand for mobile services at the bottom of the offered by the mobile platform. Text messaging income pyramid appears to be rather inelastic with is the only service beyond voice that is rapidly respect to tariff variations. The significant level being adopted. Many of the applications that of expenditure on mobile handsets and services could most benefit the poor, such as m-banking found among these low-income populations is also and m-government, are still in their infancy in consistent with the numerous benefits perceived the region. by users. Nonetheless, there are still barriers that discourage the poor from acquiring mobile phones. Taking advantage of the poverty-reduction opportunities created by the widespread In countries like Mexico, Brazil and Peru, a majority adoption of mobile phones among the poor will of the poor still cannot afford a mobile phone, require a concerted effort between market actors although many rely on informal resellers and and governments. As users advance along the family or friends to make or receive calls. Those technological learning curve, and handset prices who can afford their own mobile phone make little continue to drop, these mobile opportunities use of voice and other services, tightly controlling should continue to increase. their expenditure and frequently relying on public payphones for outgoing calls.

These findings reveal the continued need to develop innovative business models that extend the market frontier for mobile telephony. They also highlight the urgent need to rethink public policies that are premised on the mobile phone as a luxury good. For the poor, mobile telephony has long been the most cost-effective and accessible alternative. Contact information: Hernán Galperin [email protected] Judith Mariscal [email protected] [email protected] www.dirsi.net

Editing: Kristin Keenan Layout: Carlos Abril Photographs: Aldo Arozena / Kim Mallalieu