A new World Bank study - Policy Reform, Economic Growth and the Digital Divide - seeks to quantify the impact of the internet on developing countries, in Africa particularly. The topic is much discussed: development practitioners believe that the internet will have little impact in Africa because of poor education and weak telecoms infrastructure, while internet enthusiasts see the worldwide web as a major source of competitive advantage for developing countries. However, there are scant hard data on the topic.
Using statistics compiled by Pyramid Research, a telecoms research company owned by the EIU, the authors seek to address this. They believe that data on income levels will show if education is a major constraint on internet usage, while the impact of government competition policies can be measured by establishing whether internet services are more readily available where telecoms are in the hands of state monopolies. Urbanisation is included in the study to ascertain whether internet intensity grows more quickly in urbanised than in rural societies.
Their research suggests that two of the three factors - urbanisation and competition policy - are indeed major influences explaining the spread and growth of internet usage. Surprisingly, however, there appears to be very little link between incomes and the internet. In other words, rates of economic growth do not appear to have a significant impact on internet intensity. Indeed, African, Asian and Latin American countries appear to have higher internet intensities than their OECD counterparts.
As the writers point out, "This result is obviously not consistent with the conventional view of the digital divide", although they accept that the results could mean that low-income countries are experiencing rapid - but temporary - internet growth among their high-income "elites". Once this relatively small market is saturated, there could be a marked slowdown in the growth of internet use. Whatever the explanation, the researchers conclude that competition policy plays a far greater role in driving internet intensity than income per head.
However, there is a crucial distinction between internet intensity, which is defined as internet subscriptions per telephone line, and internet connectivity - internet usage per head. Income levels appear to have little impact on intensity but a major impact on connectivity. In other words, where tele-penetration is high, reflecting easy access to telecoms services, usage of the internet per head will be far greater than where telephones are scarce.
This means that there is nothing terribly new about the digital divide: it merely reflects a "long-standing disparity in telecoms access between rich and poor" rather than an additional handicap for developing countries.
Developing-world access to telecoms is at least improving: all across the world penetration rates have risen dramatically over the past decade thanks to privatisation and the deregulation of state monopolies. China's telecoms penetration rate, for example, rose 25-fold during the 1990s, while in Sub-Saharan Africa (SSA) the rate almost trebled, rising from 5 to 13 per 1,000 population.
The most significant feature of this growth has been the spread of mobile telephony. From a near-zero base in 1990 mobile subscriptions have risen to 45% of fixed subscriptions in East Asia (excluding China), 40% in South Africa, and 30% in China, SSA (excluding South Africa), and the Middle East and North Africa.
Here, very clearly, incomes and economic development are crucial. There is a strong link between growth in income per head and expansion of mobile telephone subscriptions. Urbanisation, too, is strongly correlated with mobile subscription growth. In other words, rapid growth in income per head and in urbanisation will lead to much greater telecoms access, which in turn leads to increased internet utilisation.
In an effort to quantify their research, the authors outline two scenarios.
The baseline projection assumes unchanged current values in each country in terms of growth and competition policy, while a separate projection assumes a GDP growth rate of 5% per year allied with an improved policy framework, along the lines of those already in place in countries like Botswana, Malawi and Uganda.
The data are striking. They show that even with the optimistic scenario of policy reform and far stronger economic growth than most African countries have managed for the past 25 years, tele-penetration rates per head will in general be very low in 2009 (the main exceptions being South Africa, Botswana, Cameroon, Ghana and Namibia). On the same assumptions, penetration rates in Latin America will be far higher, at 69% in Argentina, 73% in Uruguay and 50% in Chile. To make matters worse for African states, the 5% growth rate looks to be unattainable even if policy reforms are put in place, since HIV-AIDS is likely to have a substantial negative impact on GDP growth, in Southern Africa especially.
In pure number terms, however, growth in mobile phone subscriptions will be spectacular. Even without policy enhancement and more rapid income growth, the number of subscriptions--outside South Africa--is set to grow tenfold between 1998 and 2009. If the right policies are put in place, and faster income growth can be achieved, subscriptions are set to increase 44-fold over the period, from 441,000 in 1998 to more than 191m. Under such circumstances, it is hardly surprising that companies are willing to pay huge premiums for cellphone operator licences in countries like Nigeria.

SOURCE: EIU via DigAfrica