Almost 60% of the world’s population — most of whom live in developing countries — are offline. Close to 70% of households in the developing world do not have Internet access, and while Internet penetration rates have increased dramatically in recent years, the pace of change seems to be slowing.

Most of those who are not connected simply cannot afford to be. While many studies note that access prices are falling around the world, we find that the cost of fixed broadband remains about 40% of an average citizen’s monthly income across the 51 countries covered in this study, while the price for an entry-level mobile broadband package hovers at just above 10% of monthly incomes. Other issues, such as lack of relevant content, and limited digital and language literacy, combine to entrench this divide even further.

The Alliance For Affordable Internet’s 2014 Affordability Report is an effort to identify and quantify some of these challenges, and to identify what policy and regulatory drivers can lead to enhanced affordability. We aim to understand why some countries have succeeded in making Internet access more affordable, accessible and universal, and what others can do to catch up quickly.

At the heart of the report is the “Affordability Index”. This unique composite Index scores each of the 51 countries included in the report on a scale of 0 – 100, based on both current penetration and usage rates and the policy and regulatory environment in place that could lead to further progress. A statistical analysis proves that high scores on the Affordability Index are strongly correlated with lower broadband prices. To help assess drivers in more detail, the overall Affordability Index is broken down into two sub-indices — (1) the infrastructure sub-index and (2) the access sub-index.

This year, Costa Rica tops the overall rankings, followed by Colombia, Turkey, Malaysia and Peru — all middle-income countries. Rwanda secures the top spot among developing countries, followed by Nigeria, Morocco, Uganda and Kenya.

In this year’s report, we also note that certain groups are far less likely to be able to connect to the Internet affordably. Specifically, our research finds that:

• In the 51 countries that we surveyed, there are approximately two billion people earning less than $2 a day ($60 per month), according to World Bank data. Depending on the country in which they live, these individuals have to spend anywhere between 5.5% and 114.5% of their average monthly income in order to access an entry-level broadband package. At present, not a single emerging or developing country can claim to meet the affordability benchmark set by the United Nations (UN) Broadband Commission of broadband priced at less than 5% of monthly income for those potential users surviving on less than $2 a day.
• Across the board, women are far less likely to be able to access the Internet affordably than men. Research has shown that women, on average, earn 30% – 50% less than men. This income disparity diminishes the ability of women to afford to access, adopt, and benefit from a broadband connection. The Internet access gender gap is apparent throughout the world, although the extent of the gap varies from region to region.
• Those living in rural areas are often unable to secure affordable access to the Internet. This is for two primary reasons: (1) incomes tend to be lower in rural areas, resulting in a higher real cost to connect; and (2) challenges associated with infrastructure deployment in rural areas result in limited opportunities for access — particularly under current regulatory environments — or in access prices that are significantly more expensive than those in urban areas.

Using statistical tools, we have analysed the relationships between policies and regulations, and the affordability environment in each country. We have identified five common success factors.

These are:

• Effective broadband strategies: Clear and comprehensive national broadband plans, which allow for increased private investment, remove barriers to infrastructure deployment, and encourage public-private partnerships.
• Enhanced competition: A level playing field that encourages innovation and gives consumers a range of choices as to their service provider; a unified licencing framework is of particular value here.
• Efficient spectrum allocation: Spectrum allocated in a fair and competitive way, with innovative spectrum tools considered.
• Infrastructure sharing models: Laws and partnerships designed to embed open access methods and reduce sunk infrastructure costs.
• Universal access to affordable Internet services: Development of shared services, available at community centres, schools, libraries and other anchor institutions; especially important in rural areas.

Given the transformative potential that affordable access to the Internet holds, we recommend that affordable access to broadband Internet is enshrined as a sustainable development goal (SDG) by the United Nations General Assembly when it meets later this year. Close collaboration between key players — including government decision makers, international development partners, private sector, academia and civil society — on the five areas identified above will be essential to make this a reality.

In 2014, almost 60% of global households remained unconnected to the Internet. In developing countries, fewer than one in three people are online; in the world’s 49 least developed countries, that figure plummets to just 5%.

There are also ominous signs that progress may be decelerating. According to a recent study by McKinsey & Company, the worldwide growth in Internet users has slowed from a three-year compound annual growth rate (CAGR) of 15.1% from 2005–2008, to 10.4% from 2009–2013.

Clearly, significant barriers to online access persist. The most pernicious is the high cost to connect. In 2011, the UN Broadband Commission set a target for entry-level mobile or fixed broadband to cost no more than 5% of average monthly incomes (Gross National Income (GNI) per capita), by 2015. We will fall woefully short of that target. According to the International Telecoms Union (ITU), at the end of 2013, the average price for an entry-level fixed broadband connection in the developing world represented more than a quarter of an average citizen’s monthly income. Meanwhile, the price for an entry-level mobile broadband package hovers between 8% and 11.5% (depending on the plan chosen) in developing countries. By contrast, the cost to connect in many developed countries is near negligible. Citizens of rich nations pay on average just 1-2% of their monthly income to connect. In the 51 countries we analysed, the cost of fixed broadband remains about 40% of an average citizen’s monthly income, while the price for an entry-level mobile broadband package hovers at just above 10% of monthly incomes. Other issues, such as lack of relevant content, and limited digital and language literacy, further entrench this divide.

These access challenges are felt more acutely among certain populations as a result of geographic, economic, gender and socio-cultural factors, with marginalised or vulnerable groups often the hardest hit. Rural Internet users have reduced access when compared with their urban counterparts; low-income populations are disproportionately underrepresented online; and persistent income gaps, coupled with engrained social and cultural norms, keep women and other marginalised populations both from being able to afford Internet services and from being able to use the Internet freely.

The Internet has the potential to improve the quality of life — particularly for poor and marginalised communities — in myriad ways. By increasing access to information and making communication more efficient, we can observe beneficial economic and social impacts in areas like entrepreneurship, health, education, and more. There have been multiple studies looking at the benefits of enhanced access, but perhaps one recent statistic from the Copenhagen Consensus Centre is the most telling: increasing mobile broadband penetration levels threefold across the developing world would provide a return of $17 for each dollar spent. This considers both the economic growth that would be stimulated by increased access, as well as the cost savings that governments could achieve as a result.

The unprecedented potential of the Internet to empower marginalised groups and deliver social and economic benefit has not gone unnoticed by policy makers. Indeed, many are advocating that affordable access should be at the heart of the UN’s forthcoming sustainable development goals (SDGs), which will set the global development agenda for the next fifteen years. At a country level, governments are stepping up efforts to improve access to and affordability of broadband Internet. Increasingly, governments are recognising that connecting the unconnected requires a well-rounded approach to address barriers to access, including investment in new technologies, policy and regulatory reforms that reduce industry cost structures, and sustained efforts to bring affordable access to households, small enterprises, and communities. However, while technology advances at a breakneck pace, the process of updating policy and regulatory frameworks remains slow. All too often, innovative technologies with the power to reduce the cost to connect are held back by poor or out-dated policies. The policy development environment varies widely across countries, and often does not take advantage of the private sector’s ability to create competitive and innovative edges, if given the right incentives.

The Affordability Report represents an ongoing effort to understand why some countries have succeeded in making Internet access more affordable, accessible and universal, and what others can do to catch up quickly. The report assesses the policy and regulatory environments in 51 countries, and analyses the effectiveness of government programmes in addressing the divides and barriers that prevent affordable access to the Internet. At the heart of the report is the Affordability Index, a unique composite Index that assigns countries a score of between 0-100 based upon the drivers of affordability observed in the country.

This is the second edition of this annual study, which was first produced in 2013. This 2014 edition features five new countries (Myanmar, Mozambique, Dominican Republic, The Gambia and Haiti), and benefits from an improved research methodology, including streamlined and clearer survey questions based on feedback gathered last year.

To produce our country rankings, we blend secondary data, from reliable, established sources, with primary data and in-depth country research based on surveys conducted by country experts.

The Affordability Index is a composite index, composed of two sub-indices that measure the impact of two drivers critical to affordability: infrastructure and access.

• The infrastructure sub-index measures the current extent of infrastructure deployment and operations, alongside the policy and regulatory frameworks in place to incentivise and enable cost-effective investment in future infrastructure expansion.
• The access sub-index measures current broadband adoption rates and the policy and regulatory frameworks in place to encourage growth and ensure provision of affordable and equitable access.

The Index scores each country on a scale of 0 to 100. Higher scores indicate higher current penetration levels, combined with strong policy and regulatory conditions for advancing Internet affordability now and in the future.

The Index deliberately does not directly measure prices and affordability in each country; there are numerous stand-alone surveys and reports which do this — most notably, the annual ITU publication “Measuring the Information Society”. Instead, the Affordability Index measures progress toward increased broadband adoption, and the policy and regulatory environments that lead to affordability. We then analyse the Affordability Index scores against prices as measured by the ITU. In fact, our regression analysis proves that there is a strong, statistically significant causal relationship between Index scores and broadband prices. Strong scores on the Index indicate better current levels of affordability and great potential for prices to fall further in the future. (See Section 4 for a detailed regression analysis.)

Why have we chosen to focus on both infrastructure and access measures? By combining these two broad components, the Affordability Index provides a measure of the impact that each of these drivers — both individually and combined — has on affordable access outcomes. Of course, there are some important areas (e.g., tax reduction) where it is currently impossible to source comparable indicators across countries. We continue to research these important areas and hope to include them in future reports.

Finally, in order to conduct fair comparisons among all 51 countries covered in the research, we have identified each nation as either an emerging or developing country — as defined by the World Bank and determined by income levels — and have provided a separate analysis for each group. Comparing countries with similar income levels allows us to analyse the timing and patterns of decision-making as they relate to the level of economic development of each country.

This year, to reflect the Alliance for Affordable Internet (A4AI) set of best practices, we asked our expert assessors in each country to pay close attention to five particular sub-areas:

• Overarching broadband policies and programmes;
• Policy and regulation for effective competition, including the level of transparency in, and effectiveness of, the licensing process;
• Spectrum policy and regulation;
• Infrastructure sharing and access to rights of way; and
• Government-led programmes to fund or subsidise universal Internet access.

So why is Costa Rica atop this year’s Affordability Index rankings? The country has embraced broadband as a catalyst for economic growth and social inclusion and, as a result, has invested significant resources in improving infrastructure and expanding access and affordability. Costa Rica has a relatively modern telecommunications infrastructure. An estimated 88% of Costa Rica’s population can already access the Internet and declining mobile broadband access prices will enable more of the population to come online. The government has created a National Telecommunications Fund (FONATEL), which has focused on providing broadband Internet to under-served schools, public healthcare institutions, and other public entities. The government has been using FONATEL to expand fibre optic Internet service to rural towns and to install cell towers where fibre cannot reach, in order to provide access to wireless broadband services. These efforts are reflected in Costa Rica’s 8th place ranking (with a score of 48) on the infrastructure sub-index.

Where Costa Rica really shines, however, is on measures of access and affordability. The country’s score of 77.5 on the access sub-index is nearly 10 points higher than the next highest ranked country. This high score is underpinned by the nation’s National Broadband Strategy, which was launched in 2012 and outlines a strategy for increasing broadband penetration through 2016. Mobile broadband prices were about 1% of the GDP in 2014 — one of the lowest in the Latin American region. In fact, the price of an entry-level monthly mobile broadband package halved over the past year, from US$17.8 in 2013 to just US$8.79 in 2014.

Despite all of this recent progress, the country is not resting on its laurels — the government is further committed to using connectivity as a catalyst for growth and development, and is focusing both on increasing broadband penetration and Internet access, and on narrowing the access, adoption and use divide among its entire population.

Colombia (see box out) has also made concerted efforts to advance policy, regulatory and infrastructure development, with a focus on improving Internet access at the household level. Colombia not only comes out on top of the infrastructure sub-index, but also scores well on the access sub-index, placing fourth among all 51 countries covered. These scores reflect Colombia’s combined efforts to push infrastructure investment to rural and under-served areas, and to increase ICT literacy and provide subsidies to households that cannot otherwise afford Internet access. This approach has resulted in substantial increases in Internet adoption, with about 51% of Colombians using the Internet. Strong government leadership, comprehensive broadband policy and plans, increased competition, and shared infrastructure to lower costs have all proven to be key ingredients in Colombia’s successful recipe for achieving affordable universal access.

With a relatively urban, young and technically literate population, Turkey has witnessed a dramatic increase in Internet users. Almost half of the Turkish population (46%) are currently Internet users. More recent initiatives — including the Prime Minister’s National Broadband Vision, sector policies in education and e-government, and community ICT projects — have further extended access to broadband Internet. Mobile broadband is widely used by Turkey’s Internet-savvy population and the country is emerging as a noticeable front-runner in the nascent mobile payments sector, supported by high rates of credit card adoption and use. Turkey’s mobile broadband prices are among the most affordable found in emerging countries — a 500MB prepaid handset-based mobile broadband plan costs the average citizen just over one percent of their annual income.

Despite this strong affordability environment, many have raised concerns about the Turkish government’s propensity to use the Web for surveillance and censorship. The country ranked 52nd out of 86 countries on the “Free and Open” measure of the Web Foundation’s 2014-15 Web Index, and Freedom House’s 2014 “Freedom on the Net” Report noted the country’s frequent violations of user rights.

Malaysia — which topped the Affordability Index in 2013 — remains the top performing Asian nation, and is continuing its efforts to bring affordable Internet access to under-served areas. Mobile broadband prices have decreased even further over the past year, falling to 1.39% of GNI per capita from 3.2% in 2013.

Malaysia continues to do well when it comes to promoting affordable Internet, pursuing many of the policies that earned the country the top spot on last year’s Affordability Index. (Malaysia’s lower score on this year’s Index results primarily from the more recent nature of ICT sector investment and development in Colombia and Costa Rica, which allowed these two countries to score higher than Malaysia in the infrastructure sub-index.) Malaysia’s government recognises the importance of broadband networks and has continued to improve access in under-served areas. The High Speed Broadband Network (HSBB1) project that launched in 2010 and connected 1.5 million users was upgraded in 2014, and is expected to connect 4.8 million broadband fibre subscribers — a development that would make the country’s high-speed broadband connection rate one of the highest in Asia.

Peru has also seen a steady increase in broadband penetration. Approximately 40% of the Peruvian population used the Internet in 2014 — a rate that is lower than those found in Colombia and Costa Rica, but one that is growing fast. Peru has adopted a progressive policy framework and has undertaken concerted initiatives to bridge the rural-urban Internet access gap. An ambitious National Broadband Plan, drafted in 2011, calls for an 11,000km open access fibre backbone network, delivered through public-private partnerships.

Peru has introduced a unified licensing regime for the provision of all telecom services — including fixed-line and mobile broadband, pay TV, and Internet — under one license. The single concession regime includes a national register of telecom services under which each operator must notify the regulator of each new service it intends to offer. Unified concessions are no longer area-specific, but automatically cover the whole country. This streamlined licensing regime enables a wide range of players, providing a variety of applications and services to users, to enter into the market, increasing competition and eventually enabling a reduction in the cost of access and usage. Peru’s mobile broadband prices are stubbornly high but the government expects that recent policy and regulatory developments will result in further price reductions.

The five countries at the bottom of the table — Yemen, Ethiopia, Haiti, Sierra Leone, and Malawi — are all politically or economically fragile nations. Our analysis indicates that policy choices by these governments have acted as a primary contributor to their low ranking in the Affordability Index.

Yemen’s ICT infrastructure is one of the poorest in the Arab region. While political, developmental, and domestic security challenges have contributed to the Internet’s local under-development, the country’s policy and regulatory environment is also hampering the growth of broadband. Yemen’s mobile phone sector has been liberalised (i.e., new entrants are, in principle, allowed to enter the market), yet the government’s continued control of the sector has prevented the positive effects of increased competition and deregulation that generally accompany sector liberalisation. Until recently, state-owned Yemen Mobile was the only operator allowed to provide mobile broadband services. Similarly, state-controlled companies PTC and TeleYemen are the only providers of fixed Internet service in Yemen, and a significant portion of the subscriptions are still for dial-up services. Yemen’s Ministry of Telecommunications and Information Technology (MoTIT) acts as the de facto regulatory authority for the sector, responsible for the issuance of licenses, management of radio frequencies, and deployment and operations of telecommunications infrastructure. This has resulted in an unpredictable regulatory environment, with limited incentives for private investment.

As a consequence, Internet and mobile broadband prices in Yemen are some of the highest in the world and remain unaffordable for the vast majority of Yemen’s 24.5 million people. This is particularly the case for the 38% of the population that lives under the $2/day poverty line, who would have to spend 22-24% of their average monthly incomes to purchase entry-level broadband Internet access.

Despite significant ICT sector investment in recent years, Ethiopia (see box out) remains far behind other countries in the region when it comes to providing its citizens with access to quality Internet services. The country has yet to introduce any competition in its ICT sector. The Ministry of Communication and Information Technology makes policy, regulates the sector, and owns the incumbent provider, Ethio Telecom. This means that both fixed and mobile telephone and Internet services are tightly controlled under a government monopoly.

Sierra Leone faces considerable economic hardships. These hardships, combined with inadequate regulatory environments, have hampered the provision of affordable Internet. Sierra Leone’s entry-level mobile broadband price represents about 25% of GNI per capita, among the highest in West Africa.

Like Sierra Leone, Haiti has experienced political and economic difficulties. As a result, economic and social indicators remain far below the averages found in the other countries of Latin America and the Caribbean. Recent natural disasters and years of political and economic turmoil have stifled most sectors of Haiti’s economy, including the telecom sector. Entry-level mobile broadband prices spiked from US$23 in 2013 to US$39 in 2014, making access unaffordable for many citizens — particularly for the 77% of the country’s population earning less than $2 a day, for whom entry-level mobile broadband represents about 38% of GNI per capita. Haiti’s low score on the Affordability Index reflects the country’s inadequate legal and regulatory environment, limited international connectivity, and the absence of an effectively regulated wholesale broadband regime.

While infrastructure in Malawi has improved in recent years due to its access to submarine cables via Mozambique and Tanzania, prices remain very high. As of early 2013, the monthly price of fixed-line Internet access was around US$16.50, while a monthly mobile 3G data plan cost about US$24 for 1.5GB of data. The ITU estimates that entry-level mobile broadband prices are equivalent to about 28% of GNI per capita, one of the highest prices found in southern Africa.

The high cost to connect in Malawi is also caused by the country’s significant power crisis — costs associated with the need for mobile operators to power base stations are passed on to consumers and increase the cost for them to access the Internet. The country has one of the lowest electrification rates in the world, with only about 9% of the population having access to electricity. Half of the formal sector enterprises in Malawi have backup generators — twice the rate found in other low-income African countries. Malawi’s experience further indicates the crucial importance of a stable and reliable electricity supply for increasing affordability of and access to the Internet.

The 2014 Affordability Index also ranks countries based on economic development levels. Unsurprisingly, countries classified as “emerging economies” by the World Bank top the Index. Our analysis shows that the emerging economies that score higher in the Affordability Index have not only been pursuing policy and regulatory frameworks that promote healthy and competitive markets for infrastructure expansion, but have also been creating incentives to stimulate the demand for broadband services for quite a long time (at least two to three decades). Developing economies, such as Rwanda and Nigeria, are following suit, but their effort are more recent.

Rwanda is the top-ranked developing country in the Affordability Index, reflecting the success of its progressive policies, which have been designed to leverage the ICT sector as an engine for economic and social development. TheRwanda ICT Policy and Master Plan — also called the National Information and Communication Infrastructure (NICI) Plan and lately referred to as the SMART Rwanda ICT Master Plan — is divided into five phases of five years each. Phase III of the NICI plan (2011-2015) saw the government roll out a national high-speed fibre optic backbone (2565 Km of four-ring cable) that is shared with the private sector. The Rwandan government exclusively owns the infrastructure and uses one of the four ducts; the remaining three are available for the private sector on open access terms.

The Rwandan government also plans to expand ICT infrastructure to rural and under-served areas where fibre cannot reach. In March 2013, the government announced an agreement with Korea Telecom Corporation, under which the latter will invest US$140 million to deploy a Long-Term Evolution (LTE) access network over a period of three years. It is expected that this will lead to 95% of the population being within range of a high-speed mobile broadband network. However, many have expressed concern regarding this type of deployment model and the possibility that it may deter competition at the retail level. The long-term impact remains to be seen.

On the demand side, the Rwandan government has launched several ICT e-Government applications for services including e-immigration and e-health, and has also encouraged the use of mobile applications to deliver agricultural information. The new Smart Rwanda ICT Master Plan (2015-2018) aims to build on these gains.

Nigeria comes second in the Affordability Index’s ranking of developing economies — scoring higher than other African developing economies like Kenya, Morocco and Uganda, and higher even than some emerging economies, including Mexico, South Africa, Thailand and Tunisia. The backbone infrastructure in Nigeria has improved significantly over the last decade, with multiple players, including Phase 3, Glo 1, Suburban Telecom, Multilink and MTN, building fibre networks that crisscross the country. Nigeria’s regulator, the Nigerian Communication Commission, plans to award seven licenses to regional infrastructure companies to extend broadband infrastructure nationally. The first two of these were awarded in early 2015 to MainOne and IHS Communications to provide services in Lagos and North Central states, respectively. The government is also working to improve infrastructure sharing among these operators, who have traditionally built overlapping fibre networks.

The nascent “Smart States” initiative, which sees states committing to reduce the cost of broadband access by reducing taxation and simplifying regulation, is also a positive step. Nigeria’s mobile broadband penetration rate stands at just 10% — despite the fact that close to 40% of Nigerians use the Internet — and the government has put in place policies to increase this penetration level to 30% by 2018. To increase the ability of mobile operators to serve more Nigerians, plans were recently announced to auction spectrum in the 2.6 GHz band.

Morocco — the third-highest ranked developing country overall — has the top score among all developing countries in the infrastructure sub-index. Ongoing reforms in the communications sector and growing investment in fibre optic national backbone networks have had a dramatic impact on prices. The price of a monthly entry-level mobile broadband package dropped fourfold from US$49.4 in 2013 to US$11.9 in 2014, and mobile broadband penetration currently stands at 15%, while fixed broadband penetration remains at 10%. According to the regulatory agency, 88% of the country’s 8.5 million mobile users are also mobile Internet subscribers, but the majority of Internet users are in urban areas.

Uganda has seen improvement in Internet access in recent years due to increased competition in the telecom sector, a reduction of international bandwidth prices, and the availability of wireless and mobile technologies, such as WiMAX, EV-DO, HSPA and LTE, throughout the country. Though high entry-level mobile prices (18% of GNI per capita) have caused the Internet penetration rate to remain at a low 8.6% for mobile broadband and 7.4% for fixed broadband, prices continue to drop due to fierce competition and mobile market consolidation, which ultimately may result in increased access and adoption.

Uganda’s ranking as one of the top five developing countries is a result of the nation’s high score on the access sub-index. This score is due, in part, to the consistent support of Uganda’s government in promoting universal access, including shared access to Internet in rural and under-served areas through its Rural Communications Development Fund (RCDF). The project connects rural counties by subsidising small commercial Internet services at the district level, and providing resources for Internet access at schools, post offices, hospitals, libraries, community centres and health facilities.

Kenya’s ranking is driven by its vibrant ICT sector. The country has one of the fastest growing ICT markets in Africa, and the use of mobile services to access the Internet and complete financial transactions has grown significantly in recent years. Two-thirds of Kenya’s 22 million Internet users have access to broadband wireless Internet. Kenya’s mobile broadband prices have been falling in recent years — standing at around 7.5% of GNI per capita — due to competition between key players like Safaricom, Telekom (Orange), Essar (Yu) and Airtel.

Kenya has also seen a surge in Internet bandwidth due to enhanced access to undersea cable capacity, and an improved national backbone that has resulted in increased availability of high-speed broadband. The quality of broadband services is steadily improving, with speeds of over 10 Mbps now available to many citizens.

Our analysis suggests that three groups are the least likely to be able to access affordable Internet:

• Those living in extreme poverty
• Women
• Those living in rural areas

It will come as no surprise that those living in poverty are the least likely to be able to access the Internet affordably. Yet the scale and scope of the challenge is staggering. In the 51 countries that we surveyed, there are approximately two billion people earning less than $2 a day ($60 per month), according to World Bank data. Depending on the country in which they live, these individuals have to spend anywhere between 5.5% and 114.5% of their average monthly income in order to access an entry-level broadband package.

The UN Broadband Commission has set a target of entry-level Internet access priced at less than 5% of GNI per capita. While 23 out of the 51 countries have met this target on an aggregate basis, at present, not a single emerging or developing country can claim to meet this benchmark for those two billion potential users that survive on less than $2 a day.

While many countries — including Botswana, China, Dominican Republic, Namibia and Zimbabwe — made headway in their efforts to reduce Internet access prices between 2013 and 2014, prices in many other countries remained relatively constant. Some countries — including India, Jamaica and Peru — even saw small increases in their broadband access costs. Clearly, policy and regulatory action is urgently needed if governments are serious about using ICT and access to the Internet as a vehicle for poverty reduction and delivery of government services.

Women are far less likely to be able to access the Internet affordably than men. Research has shown that women on average earn 30% – 50% less than men — a disparity which diminishes the ability of women to afford, adopt, and benefit from broadband access. The Internet access gender gap is apparent throughout the world, although the extent of the gap varies from region to region — in parts of Europe and Central Asia, research has found that 30% fewer women than men access the Internet; in Sub-Saharan Africa, this figure jumps to 45%. The gap widens in rural areas — in some rural and remote areas of Asia, for example, it was found that men’s access to the Internet outnumbers women’s access by 50%.

A simple analysis of the gender pay gap and its impact on women’s ability to afford Internet services clearly illustrates that the price of entry-level mobile broadband service is significantly higher for women worldwide. For example, entry-level mobile broadband costs at least 8% of women’s GNI per capita, compared with 5.6% for the average Nigerian. In Malawi, entry-level mobile broadband is at least 40% of women’s GNI per capita, compared with 28% of the income of the average Malawian. (See table below for additional details.)

A Research ICT Africa Network survey also found that women “generally have less access to ICTs than men and this increases as the technologies and services become more sophisticated and expensive, requiring greater levels of income and education to access and to operate.”

The implications of women’s limited ICT access are significant — both for women, as well as for society at large. Limiting women’s access denies them the tools, resources and opportunities available through the Internet, which in turn slows economic growth and social development opportunities. More than 70 % of Internet users surveyed for Intel’s 2013Women and the Web study consider the Internet “liberating”, and 85 % believe that it “provides more freedom”. Access to the Internet has been shown to have personal, social and market benefits — from improving education and digital literacy levels, to increasing individual productivity and earning power, household resources and social capital.

Despite the fact that the United Nations has repeatedly underscored the Internet’s transformative potential, only 21% of women and girls in developing countries have access to the Internet. The possibilities for women’s empowerment via access to affordable Internet demonstrate the need for evidence-based research based on how women currently access and use the Internet. The Alliance for Affordable Internet is working closely with the World Wide Web Foundation’s gender gap research programme to understand better the nature of the gender digital divide. We hope that next year’s Affordability Report will shed more light on the extent of differential access to and use of the Internet by women and men.

Those living in rural areas are often unable to secure affordable access to the Internet. This is for two primary reasons. First, incomes tend to be lower in rural areas, making the real cost to connect higher. Second, the challenges of infrastructure deployment in rural areas can result in access being impossible, or significantly more expensive than in urban areas.

The 2012 research study by Research ICT Africa covering 12 countries in Africa found that the urban and rural divide is quite significant among Internet users. Of the 23.8% of the population found to be Internet users (defined as individuals that had used the Internet at least once in the preceding 3 months) in Uganda, 17.3% were based in urban areas, compared with just 6.5% in rural areas. In Mozambique, only 3.2% of the rural population was found to use the Internet, compared to 26% in urban areas.

As noted in Section 2, the Affordability Index does not directly measure cost, instead focusing on those areas broadly agreed to be the drivers of affordability. However, there is a strong relationship between Index scores and broadband affordability. A regression analysis (below) shows that high scores on the Affordability Index are strongly correlated with lower broadband prices as a percentage of GNI per capita. Therefore, we can conclude that rising scores on the Affordability Index are associated with lower broadband costs in that particular country.

So what factors are really driving higher scores on the Affordability Index, and therefore progress toward affordable Internet across the globe? While each country is different, we have been able to identify a number of common success factors.

These are:

• Effective broadband strategies;
• Healthy competition;
• Non-discriminatory access to spectrum;
• Infrastructure sharing; and
• Strategies to deliver universal access to rural and under-served populations.

These factors are the mutually reinforcing pillars that drive affordability. As the figure below illustrates, these pillars need to work in harmony since they cannot support and nurture universal affordable access on their own.

All of the countries that top the Affordability Index have prioritised at least four of these five key drivers of affordability. Of course, these actions cannot take place in a vacuum and their efficacy is enhanced by progress made in other areas, such as regulatory capacity, regulatory independence and increasingly transparent policy-making processes.

A closer look at the scores (with 1 being the lowest and 10 the highest) received by the top five emerging and developing countries reveals that policymakers in Index-leading countries are indeed making concerted efforts to push forward the policy and regulatory areas that will have a significant impact on affordability outcomes (see table below).

In addition to these specific policy areas, our experiences and observations across this year’s Affordability Index research have reinforced the critically important role of strong political leadership to ensure that these affordability pillars work in harmony. Leadership is the foundation of affordability, essential to achieving the ultimate objective of reduced broadband connection prices and affordable access for all. The experiences of Costa Rica, Colombia, Malaysia, Rwanda and Nigeria all demonstrate the importance of leadership — it not only encourages the telecom sector to carve a path toward broadband expansion, adoption and use, but also provides certainty in the market by pushing for and supporting the forward-looking legal and regulatory systems that will support the vision of ICT and the Internet as engines for economic growth. It further promotes competition by supporting the development of incentives for new market entrants and, at the same time, clarifies the role of government and other players as partners working toward the same objectives.

Comprehensive broadband strategies that consider both the supply of and demand for broadband services, and that reflect a partnership-based approach to decision-making and implementation, are more likely to be successful in increasing universal access to and use of affordable Internet. Notably, the mere presence of ICT infrastructure in a country does not guarantee affordability. For example, though Argentina, Indonesia and South Africa all rank highly on the Internet infrastructure sub-index, their rankings on the overall Affordability Index are much lower as a result of stubborn obstacles to increased affordable access. These countries have recently attempted to develop new and comprehensive broadband plans, which, once implemented, could have the kind of impact that has been seen in higher scoring countries.

A regression analysis shows that there is a direct and significant relationship between the cost of fixed and mobile broadband access and the existence of government-led national broadband plans that provide guidance regarding best practices for the implementation of infrastructure expansion strategies. In other words, the extent to which national broadband Internet plans set clear, time-bound targets and interventions for increasing penetration has significant implications for the reduction of fixed and mobile broadband costs. Currently, about 12 of the 51 countries covered in the study do not have any form of ICT or broadband plan or policy; many others have an plan in place, but one which has not been updated to reflect the broadband era.

The figure below shows, for example, the relationship between mobile broadband prices and the presence of a broadband plan or strategy.

For broadband plans to be truly effective, they must enhance supply, stimulate investment and create sustained user demand. Efficient broadband infrastructure is a necessary condition for access to affordable Internet, but it needs to be deployed within a framework where incentives to make investment cost effective are in place. Demand stimulation can be tackled by enhancing user awareness of the benefits of affordable Internet access, driving digital literacy programmes, supporting content developers as new entrepreneurs in the digital economy, and promoting the electronic provision of government services. When these factors work in combination, countries can develop a sustainable ecosystem rooted in affordable access.

Within this framework, it has also become clear that broadband expansion strategies must be coordinated with developments in the energy sector. In fact, several of the governments of the top Affordability Index countries are now collaborating with their respective energy ministries, and stakeholders within the energy sector, to coordinate infrastructure expansion plans with a view of reducing unnecessary costs and increasing shared infrastructure options across the sectors. Our analysis shows that electricity is an essential infrastructure variable in the path toward affordability — the lower the electrification rate, the higher mobile broadband prices are, and vice versa (see figure below). This is especially the case in sub-Saharan Africa, where operators are compelled to install independent generators that power base stations and other elements of the network. According to GSMA research, by 2012, there were around 165,000 base stations across sub-Saharan Africa lacking a reliable electricity supply and powered with diesel generators. This amounts to approximately 79% of all base stations in the sub-Saharan region. This situation is quite concerning and again, requires urgent collaboration between energy and communications ministries to coordinate infrastructure development policy and efforts that will lead to increased and reliable electricity supply across the countries.

It is well documented that a liberalised market with an open, competitive environment leads to better prices and service options for users. As seen in Ghana, Brazil, India and many other countries across the globe, competition has the power to fuel the growth of affordable mobile voice services. Competition also fuels the growth of affordable access to the Internet, but, as pointed out above, competition alone is not a sufficient condition to ensure affordable access. It is therefore necessary to have the right policy and regulatory framework in place to ensure that competition is indeed effective (i.e., results in better choices and lower prices for users), and that any anti-competitive behaviours or actions are addressed by regulators with the capacity to enforce clear and transparent rules, based on robust market analysis and established within the existing framework.

One of several elements that contributes to an effectively competitive market is the establishment of a licensing regime that provides clear rules for market entry while also allowing businesses the flexibility to make the most appropriate and cost-effective decisions about the technology used and the type of services offered (i.e., allowing for technology and service-neutral licenses). Such licensing regimes are based on the concept of unified licensing frameworks (ULFs), which minimise the administrative and formal requirements to enter into the market, generally providing technology and service-neutral licenses as a way to encourage and enhance competition at all service levels.

Our primary research indicates that those countries that have adopted ULFs, and those that have independent regulators with the capacity to enforce rules, have made better progress toward market growth and affordability (e.g., Kenya and Turkey). By contrast, countries with vertical market structures, weak regulatory institutions, and anti-competitive environments tend to suffer from high prices and low connection rates, leading them to languish at the foot of the Affordability Index (e.g., Ethiopia and Yemen).

There are 2.3 billion mobile broadband users around the globe — with only 21% of global mobile broadband subscriptions belonging to users in developing countries. As the demand for mobile and wireless broadband continues to rise, especially in developing countries, there is increasing competition for spectrum (especially in the digital dividend band of 700 MHz, the Advanced Wireless Services (AWS) band of 1710-1755/2110-2155 MHz, and the 2.5 G band (2500-2570/2620-2690 MHz)). TV White Spaces in the Ultra High Frequency (UHF) band have also been identified as an option to increase affordable access to broadband services, especially in rural areas.

Countries that top the Affordability Index have made spectrum available on a competitive and non-discriminatory basis (see case of Peru below). Radio frequency spectrum is fundamental to expand networks to last mile areas and to provide higher data throughputs for wireless broadband services. Making licensed spectrum available to operators on a competitive market basis will bring down the cost to access wireless networks, while offering more license-exempt spectrum bands to users will foster innovation.

Regulators in developing countries need to develop spectrum policies that are responsive to the high demand for mobile broadband spectrum. Not only should they focus on making the Digital Dividend bands (i.e., 700 MHz and 800 MHz bands) available as soon as possible, in addition to other possible options referred above, but they should commit to implementing transparent and non-discriminatory assignment procedures, while at the same time aiming to lower spectrum prices. Columbia and Peru provide useful experiences in this area.

There is emerging evidence to suggest that making mobile broadband spectrum available to operators and users will speed up universal, affordable access to the Internet, thereby boosting jobs, productivity, and sustainable growth. Astudy by the GSMA in Latin America indicates that licensing the Advanced Wireless Service (AWS) band in Argentina, Ecuador, El Salvador, Nicaragua, Guatemala, Panama and Paraguay will stimulate investment of about US$30 billion in the ICT sector, and will indirectly contribute US $23 billion across other industry sectors. The other major benefits of competitive allocation of the AWS spectrum include the creation of new jobs, increasing development-oriented mobile app and content. A similar study by the GSMA indicates that releasing the 700 MHz, 800 MHz and 2600 MHz bands in six countries in Africa (Ghana, Kenya, Nigeria, Senegal, South Africa and Tanzania) — in other words, the Digital Dividend and 2.6 GHz bands — would have a US$ 33.6 billion impact on the GDP of these nations between 2015 and 2020 and would lead to the creation of 14.9 million jobs. Policy makers and regulators must urgently implement measures to ensure that their countries can reap the benefits of increased mobile broadband spectrum availability.

Infrastructure sharing mechanisms could cut broadband costs significantly – by up to 80% of current deployment costs, according to a forthcoming study by the Association for Progressive Communications. Infrastructure sharing reduces the capital costs of network deployment and therefore supports expansion and increased geographical coverage. It also reduces operating costs (e.g., tower maintenance and operation) by allowing operators to share these costs. If new market players can gain access to existing infrastructure at competitive rates, entry barriers will be minimised, promoting competition that can result in reduced prices.

Infrastructure sharing occurs at different levels — through opening up access to the existing copper network, through joint building and operation of shared backbone infrastructure, or through coordination among linear infrastructure providers, like power lines, gas pipelines, or roads. The participation of all market players in creating a special purpose vehicle for aggregating, building and marketing backbone networks under open access principles has been gaining momentum in Africa in recent years and, as evidenced by the case of Burundi, has had a positive impact on network expansion and affordability.

Our research shows that countries that have instituted shared infrastructure in the backbone market fare better than those with limited initiatives for cooperation between operators. Countries that have implemented infrastructure sharing mechanisms have generally seen improved access at affordable prices (e.g., Kenya, Malaysia, Ghana and Nigeria). It is therefore important to promote commercially driven sharing, based on open access principles, and encourage collaboration among linear infrastructures providers, like power, gas and railway firms, to coordinate the building of broadband networks and leverage their rights of way and other assets to reduce access costs.

Successful experiences in Colombia, Malaysia, Pakistan and Nigeria demonstrate that governments in developing countries have a critical role in facilitating broadband infrastructure investment and Internet adoption in the digital age. These experiences further illustrate that universal access and service funds can be highly effective in expanding Internet access and use when they are updated and adapted to support national broadband strategies and targets. Our research shows that each of the top 15 countries has made a concerted effort to promote shared access at both the community and institutional level. They have carried out public investments to expand broadband networks (especially in rural areas), created community broadband access centres, encouraged enterprises to provide services through cybercafés, and extended Internet access through government offices and community anchor institutions like schools, libraries and hospitals. Such steps are vital in order to extend broadband access to very poor or marginalised communities.

In Thailand, for example, the government established the Broadcasting and Telecommunications Research and Development Fund to support universal service for broadcasting and telecommunications and promote community services. The fund was used to create public WiFi networks in over 30,000 centres with a total of 150,000 access points. The government plans to increase broadband wireless coverage to 80% of the population by 2016.

Peru provides another example of where a USF is being used to extend broadband networks to under-served communities. Peru’s Fondo de Inversión en Telecomunicaciones (FITEL) is hailed as one of the most successful programmes to extend access to communication networks in rural areas using the competitive subsidy scheme. While the original plans focused on narrow band network and voice communications, FITEL has now begun promoting access to broadband by improving wireless broadband and the rollout of fibre networks in under-served areas of the country. Municipalities, educational institutions, and health centres have all been targeted, and communities in under-served areas are increasingly gaining access to broadband.

Mexico’s Ministry of Communications and Transport (SCT) has launched a project dubbed “Mexico Conectado” (“Connected Mexico”), which seeks to boost broadband access across the country. The project covers the deployment of broadband lines in over 250,000 public spaces, including schools, government institutions, and hospitals, by 2018. The project has provided high-speed Internet access to over 11,000 schools and community centres and 9,000 public access spaces. The government plans to expand Internet penetration to 60% of the population by 2018.

This report is being released six months before the UN General Assembly gathers to debate and agree the sustainable development goals (SDGs) that will guide the global development agenda for the next 15 years. Our research has clearly demonstrated both the beneficial developmental impact that affordable Internet can have, as well as the common, replicable success factors that can drive prices down.

Our key recommendation, therefore, is:

“To enshrine affordable access to broadband Internet in the sustainable development goals, and require all stakeholders to work toward achieving affordable, universal access in the coming years through a blend of infrastructure investment and policy and regulatory reform.”

Of course, this will not be a simple task. It requires a multi-pronged strategy — one that depends on close collaboration among key players, including governments, policymakers, private sector, academia and civil society. Crucially, policies and regulations should address both the demand and supply side of Internet development.

In this final section, we present our key recommendations, grouped by the success factors we have identified and broken down into action points for governments, the private sector, and civil society.

If gaps in broadband infrastructure remain, the poor and those who live in remote areas will remain offline. In addition, as the demand for quality broadband increases, the need for robust in-country infrastructure and international connections surges. Addressing the infrastructure gap demands clear understanding of the gaps, increased private investment, and the development of public-private partnerships — all tied together under a clear and holistic plan.

Many challenges can be addressed by intensifying competition, in particular through the adoption of a unified licensing regime. Access to resources such as spectrum and right of way must take place in a market-based, non-discriminatory and transparent fashion. This also calls for enhancing the capacity of regulators and the adoption of evidence-based regulatory principles.

The Affordability Index data show few operators are sharing passive and active components of their networks or spectrum. Infrastructure sharing is an opportunity that can be tapped into to reduce the cost of Internet access; therefore, commercially driven sharing should be accelerated.

Wireless technologies provide cost effective means of Internet access in remote and rural areas, but their impact depends on the availability of spectrum on competitive, open and fair terms. In order to ensure their availability, policy makers, regulators and industry should agree to abide by transparent spectrum plans.

Those users who cannot afford access to the Internet need targeted subsidies through universal access funds. These subsidies can be provided directly or through shared infrastructure. Demand-side initiatives, such as e-government services and national research and education networks, can have a significant impact on Internet usage. Governments need to commit resources to increase access at community centres and local public institution facilities, such as libraries and schools, and promote initiatives that stimulate the uptake of e-services.

This report was written by Sonia Jorge and Lishan Adam, with contributions from Kojo Boakye and valuable advice and editorial support from Dillon Mann, Lauran Potter and Anne Jellema of the World Wide Web Foundation.

The Affordability Index research was directed by Hania Farhan and Siaka Lougue (African Institute of Mathematical Sciences), with support and advice from the Web Foundation’s research team. A wide range of experts were involved in the primary research and we thank them for their contribution.

The A4AI Affordability Working Group — composed of member representatives from Alcatel-Lucent, the Association for Progressive Communications, Cisco, DIRSI, Ericsson, Google, LIRNEasia, the Internet Society, Microsoft, Research ICT Africa, USAID and the Web Foundation — provided valuable inputs and advice during the Affordability Report review workshop that took place in Cape Town in November 2014 and on earlier drafts.

Finally, we are grateful for the support of A4AI’s global sponsors — Google, the UK Department for International Development and USAID — and that of the Alliance’s entire membership.

The Affordability Index is a composite measure that summarises in a single (average) number an assessment of the drivers of Internet affordability in various countries. Benefiting from the research framework established by the Web Index, the Affordability Index covers 51 countries and focuses on two key aspects driving affordability: communications infrastructure and access.

Methodology

Two types of data are used in the construction of the Index: existing data from other data providers (“secondary data”), and new data gathered via a multi-country expert researcher survey (“primary data”). The survey consists of a set of questions — scored on a scale of 0 – 10 — on issues regarding policy, regulation, and various other aspects around broadband and affordable access to the Internet. The questions were specifically designed by the Alliance for Affordable Internet, the Web Foundation, and its advisers. These primary data, based on and aligned with the A4AI Best Practices, attempt to assess the extent to which countries have achieved a policy and regulatory environment that reflects the best practice outcomes.

Survey questions were scored based on predetermined criteria by country experts. Three country experts were asked to provide evidence and justification that supports each score. The scores were checked and verified by a number of peer and regional reviewers.

Data sources and data providers

The sources of the secondary data that we use are highly credible organizations that produce consistent and valuable data in various fields. We are grateful to those organizations for allowing us to use and reproduce their data. A complete list of data sources and the individual indicators used from each is available on the A4AI website (adi.a4ai.org).

Indicator inclusion criteria

We searched a very large number of international databases to find indicators that measure or proxy the dimensions under study.

Before an indicator is included in the Index, it needs to fulfil five basic criteria:

• Data providers have to be credible and reliable organisations, which are likely to continue to produce these data (e.g., theirs is not a one-off dataset being published).
• Data releases should be regular, with new data released at least every three years.
• There should be at least two data years for each indicator, so that basic statistical inference could be made.
• The latest data year should be no older than three years back from publication year.
• The data source should cover at least two-thirds of the sample of countries, so that possible bias — introduced by having a large number of indicators from one source that systematically does not cover one-third or more of the countries — is reduced.

Index Computation

There are several steps in the process of constructing a composite Index. Some of those involve deciding which statistical method to use in the normalisation and aggregation processes. In arriving at that decision, we took into account several factors, including the purpose of the Index, the number of dimensions we were aggregating, and the ease of disseminating and communicating it in an understandable, replicable, and transparent way.

The following seven steps summarise the computation process of the Index:

1. Take the data for each indicator from the data source for the 88 countries covered by the Web Index for the 2007-2013 time period (or 2014, in the case of the expert assessment survey). Impute missing data for every secondary indicator for the sample of 88 countries over the period 2007-2013.Some indicators were not imputed, as it was not logical to do so. None of the primary data indicators were imputed. Hence, the 2014 Affordability Index is very different from the 2007-2013 Indexes that may be computed using secondary data only. Broadly, the imputation of missing data was done using two methods, in addition to extrapolation: country-mean substitution if the missing number is in the middle year (e.g., have data for 2009 and 2011, but not for 2010), or taking arithmetic average growth rates on a year-by-year basis. For the indicators that did not cover a particular country in any of the years, no imputation was done for that country/indicator.
2. Normalise the full (imputed) dataset using z-scores (z=(x-mean)/standard deviation), making sure that for all indicators, a high value is “good” and a low value is “bad”.
3. Where applicable, cluster some of the variables (as per the scheme in the tree diagram), taking the average of the clustered indicators post-normalisation. For the clustered indicators, this clustered value is the one to be used in the computation of the Index components.
4. Compute the two sub-index scores using arithmetic means, using the clustered values where relevant.
5. Compute the min-max values for each z-score value of the sub-indices, as this is what will be shown in the visualisation tool and other publications containing the sub-index values (generally, it is easier to understand a min-max number in the range of 0 – 100 rather than a standard deviation-based number). The formula for this is: [(x – min)/(max – min)]*100.
6. Compute overall composite scores by averaging the sub-Indexes (at z-score level).
7. Compute the min-max values (on a scale of 0-100) for each z-score value of the overall composite scores, as this is what will be shown in the visualisation tool and other publications containing the composite scores.

Choice of weights

This year, given the feedback and advice from various experts and the Working Group, we have assigned equal weights across all indicators and sub-indexes.