Is the Internet stuck?

Is the digital divide growing or shrinking? In a relative sense, at least, the divide seems to be shrinking. The number of Internet users around the world grew from 381 million in 2000 to 795 million in 2004, an annual average growth rate of 20% (see figure 1).

Growth was higher in developing countries, so that their share of the world’s Internet users rose from 21% in 2000 to 39% in 2004. The number of Internet users per 100 inhabitants in developing countries was 1.6 in 2000, rising to 5.9 by 2004. As a result, the relative difference in Internet penetration between developed and developing countries dropped from 19.5 to 8.5 between 2000 and 2004. These numbers suggest good reason for optimism.

However, although the relative difference between developed and developing countries has decreased, the absolute difference has increased.
The difference between Internet users per 100 inhabitants in developing and developed countries rose 13.6 points between 2000 and 2004, from 30 to 43.6. One alarming statistic is that the annual growth rate in the number of Internet users has been declining since 1998 (see Figure 2). This is certainly cause for concern, considering the intense focus governments around the world have been according to information and communication technology (ICT). For the moment, at least it seems clear that this attention has not translated into a dramatic rise in growth. Instead, the Internet seems to risk “getting stuck.”

If this decline in growth rates continues, it seems unlikely that the World Summit on the Information Society (WSIS) goal “to ensure that more than half the world’s inhabitants have access to ICTs within their reach” will be met by 2015¹. Indeed, at the current rate, less than a fifth of the developing world will be online by 2015 (Figure 2).

The absolute gap in access to ICT gives rise to striking discrepancies. There are more Internet users in the South Korean capital of Seoul than in all of Sub-Saharan Africa (excluding South Africa), and more Internet users in London than all of Pakistan. Switzerland, host of the first WSIS, has five times the Internet penetration rate of Tunisia, host of the second WSIS.

The divide within

Discrepancies are just as striking between developing regions as within developing countries. There is as big a relative difference in Internet penetration between the region of Europe and Central Asia (ECA) and the region of Sub-Saharan Africa (SSA) as there is between developing and developed countries. The number of Internet users per 100 inhabitants in ECA was 12.3 in 2004, compared with just 1.4 in SSA – almost 9 times fewer (see Figure 3).

One problem for statisticians, however, is that it has always been difficult to generalize about the digital divide, since in many so-called “third world” countries have pockets of “first world” infrastructure and access. A case in point: there are more households with Internet access in Brazil than in Norway; furthermore, the penetration rate among wealthy Brazilian homes is higher than Norway (Figure 3). But in low-income Brazilian households—which make up 85% of all homes in the country—only 5.1% of the population has Internet access.

Two speed Internet

If the differences in Internet access between developed and developing countries are striking, the contrast between the speeds at which that resource is accessed is even more shocking.

Developing countries accounted for just 20% of the 153 million broadband subscribers around the world in 2004. If China is excluded—the second largest broadband market in the world—then the developing country share drops to only 6% (Figure 4). This translates to a broadband penetration rate of 0.24 per 100 inhabitants in developing countries — 50 times less than the rate of 12 per 100 inhabitants figure in developed nations.
Looking even further, discrepancies in international Internet bandwidth — the critical infrastructure that dictates the speed at which web sites in other countries can be accessed — are nothing short of astounding. Tiny Denmark has more than twice the international Internet bandwidth than Latin America and the Caribbean combined (see figure 4).

The high cost of international bandwidth is cited as a major constraint, with developing countries often having to pay the full cost of a link to a developed country hub. More than 40 countries have less than 10Mbps of international Internet bandwidth, whereas in Belgium, a 9Mbps consumer ADSL package is available for just EUR 60 a month.

This contrast between available bandwidth, which in many developing countries remains scarce and expensive but which in the developed world is now a commodity which is declining rapidly in price, is a major barrier to bridging the digital divide. Users in many developing countries are simply locked out of the full Internet experience, since they lack sufficient bandwidth to adequately access the growing number of multimedia and graphics-rich sites. At present, the norm in developing countries for those privileged enough to afford to go online at all is slow and unreliable dial-up. In the world’s richest nations, modern broadband is now often as cheap as old dial-up.

Access to applications

For users, quality of experience is linked to applications and sophistication of use. But most developing countries trail far behind when it comes to indicators measuring application use. A United Nations index measuring the availability of e-government applications shows that developed countries score much higher than developing ones (see Table 1). And whereas most developed nations have connected nearly all their primary and secondary schools to the Internet, just 38 percent of developing countries have done so — and fewer than 1 percent of many African countries.

Differences in the number of secure Internet servers — essential for the availability of e-commerce — are similarly stark. While developed nations have more than 300 such servers per 1 million people, developing nations have fewer than 2. Canada, for example, has more secure servers than all the world’s developing countries combined.

Table 1: Measures of electronic government, education, and commerce in developing and developed countries, 2004

NOTE — Based on incomplete data, with 142 countries for e-government, 68 for e-education, and 122 for e-commerce. n.a. - not applicable. Sources: WDI 2004; Netcraft 2004; UNPAN 2004.

What about mobiles?

Much has been made of the success of mobile cellular technology in bridging the voice gap in developing nations.

High-speed mobile technologies such 2.5G (GPRS, CDMA2000 1x) and 3G (WCDMA, CDMA EV-DO) could provide an alternative for Internet access, particularly where fixed lines are in short supply. In some developing countries, mobile Internet is proving effective as an solution to bridging the digital divide. In Peru, 40% of Internet subscriptions are mobile using Wireless Application Protocol (WAP) (see figure 5). In Romania, 73% of broadband subscriptions use IMT-2000, ITU’s global 3G standard.

However, these countries are exceptions. 2.5G, and particularly 3G, are not widely implemented in developing countries. Where they are provided, they are not always available as part of prepaid packages (which form the majority of users in developing nations), or tariffs are simply too expensive.

One development that could cause concern in the realm of Internet access is new initiatives focused on producing cheap handsets for developing countries. While this is a welcome development for spreading basic voice communications, there is a worry that such handsets, stripped of advanced capabilities, will in their own way represent a barrier to future Internet access.

The way forward

What can be done to boost Internet penetration and put developing countries more firmly on the information society path?

Predictably, perhaps, the problem is a complex one, with no single factor that applies equally to all countries. Rather, sustained growth will depend on a number of areas:

• Leveraging the capability of broadband wireless technologies. This includes deploying high-speed mobile technology on the back of the existing cellular networks that are already so successful in developing nations. Countries should encourage the launch of 3G networks by eliminating high licence fees, making the required frequencies available and allowing current operators to upgrade or deploy new networks using their existing spectrum allocations. Countries should also support the use of high-speed mobile for bridging the digital divide through innovative universal service schemes with mobile operators. Another promising wireless technology is WiMax. As with high-speed mobile, governments should make the requisite spectrum available quickly and as cheaply as possible, and avoid the temptation to charge high licence fees.
• Countries should open their telecom markets to the fullest extent possible. The resulting competition will help drive additional infrastructure deployment and reduce prices, making services more affordable for citizens. Competition is particularly important for Internet access and critically, for international connectivity. Where there are still regulatory obstacles to freeing the international bandwidth market, governments should consider alternative ways of reducing wholesale prices. s.
• Encourage use of the Internet, particularly in those markets where low Internet access is not primarily the result of lack of infrastructure or affordability. This can be done through government and industry partnerships to raise awareness of the benefits of the Internet, and programmes to train potential users. Governments can also attract more Internet users by developing e-government services that citizens regularly need to use, and ensuring that these applications are available nationwide through community access centres.
• Finally, the international community has a critical role to play, particularly in assisting the 49 UN-designated Least Developed Countries, where for the moment the economics of Internet access simply cannot foster widespread access because of extremely low per capita incomes. The some 30 countries with an Internet penetration of less than one should be a particular focus of attention.

Shock measures are needed to dramatically increase Internet access in developing countries. Without them, it is unlikely developing nations will reach the WSIS connectivity goals by 2015.

¹ See WSIS Plan of Action, “B. Objectives, goals and targets” available here.