A little reality check on business and the internet in developing countries

Internet usage has been increasing at a rapid rate over the last decade or so. For example, according to World Development Indicators (WDI, World Bank), internet subscribers equaled 13% of the total population in low and middle income countries in 2007. This is up from a mere 1.5% in 2000, implying on average an increase of over 109% per annum in the proportion of internet subscribers.

There is a small but growing body of work that shows that the spread of internet can have a fairly strong positive effect on economic efficiency, functioning of markets and economic development in general (see, for example, Freund and Weinhold 2002, 2004).

A good number of these studies use a general purpose measure of internet penetration such as the number of internet subscribers, portals, websites, hosts, etc. For these measures to make sense, it is important they correlate well with internet usage for business or commercial purposes. Simply exchanging, for example, personal emails with friends is unlikely to have any significant impact on economic growth.

To check on the degree of correlation, I define internet penetration among businesses as the percentage of firms in a country that use email to communicate with clients/suppliers (Enterprise Surveys). The measure is available for over 80 countries. Next, I use the traditional measure of internet penetration: the number of internet subscribers per capita in a country (WDI, World Bank). The figure below shows that the two measures are reasonably well correlated: about 60% of the variation across countries in internet usage for general purposes (WDI measure) can be attributed to internet usage for business purposes (Enterprise Surveys measure).

While this is a good start, two concerns remain. First, existing studies are not picking up about 40% of the variation in internet usage by businesses.  Second, commercial usage of internet may extend beyond firms communicating with clients and suppliers and it is not clear how well the WDI measure can serve as a proxy for these other commercial usages of the internet. Both these problems add to the noise in the data, rendering current estimates of the relationship between internet usage and economic development potentially imprecise.

Source: Enterprise Surveys and World Development Indicators, World Bank. The Enterprise Surveys data for the various countries were collected during 2006-2009. The WDI measure of internet subscribers is averaged over 2005-2008. Country acronyms are taken from WDI, World Bank.