Wednesday, April 09, 2003


My colleague Charles Kenny (and his collaborator, Carsten Fink) pointed out to me that while the numbers of fixed and mobile telephone lines per capita have diverged between developed and developing nations, as have the numbers of internet users per capita, the rates of growth of both kinds of connectivity are larger in developing than developed countries. How can this be? Easy, the developed countries started with larger numbers. Ten percent of 50 is more than 20 percent of five. Still if developed and developing country connectivities keep growing at the same rates as in recent past, developing country connectivity will soon equal that of developed countries.

Of course it may not be the case that the recent trends will continue. The market for fixed phone lines and mobile phones may saturate in the rich countries. When the available computers with phone lines are all linked to the Internet in Africa, the growth of the Internet may slow? But supposing that the growth trends do continue, how are we to interpret them?

Kenny also points out that people in developing countries tend to spend more of their income on telephone services than do people in developed. (There is a countervailing trend for rich countries to spend more of their GDP on ICT than do poor countries). Again, how are we to interpret this information.

I suspect that there is a limited need for personal telephone service, at least measured in fixed lines per person. If so, the market can be saturated, and the growth in connectivity reduced.

(I also suspect that the use of communications by things will soon eclipse that by people, and it may be that the rate of growth of connections to the Internet will grow exponentially for a far longer period than one might suspect. What if our homes and cars have Internet links? Our refrigerators and home appliances? Every item of clothing? All the individually wrapped packages in retail trade? All of these possibilities are already discussed. One implication, is that I don’t think Internet connectivity of a can of beans in the supermarket should count as much as of a person.)

Returning closer to the point, I would point out that poor people (who are not going hungry) often spend a larger portion of their incomes on food staples than do rich. How much bread, or rice, or beans can you eat, or do you want to eat. As people’s income increases, they spend a greater portion on “luxuries”, and the basic necessities come to represent a smaller portion of income. It could be that, similarly, telephone use (where it is available) is a basic consumer service that everyone uses, but that the marginal propensity to consume phone service declines with income.

Clearly people in the business of building the ICT infrastructure should be counting how many people it serves. Connectivity measures are important. Indeed, so are issues of cost, availability, accessibility, and utilization.

The Digital Divide is often discussed in terms of connectivity. Indeed, there is a divide in connectivity now. But I think that is not the Divide that we in international development should emphasize.

Ideally, perhaps we should focus on the divide in benefits accruing to people as a result of the information revolution. I feel sure that the world’s richest man, Bill Gates?, is benefiting much more from the Internet than I do, and this is true whether or not he is personally connected or whether he personally surfs the net. The indirect benefits that result from all his investments and all of his staff are clearly greater than mine. Ideal maybe, but impractical to try to measure such benefits.

Kenny and Fink point out that the air conditioning divide between the US and India is about 74 fold, or the same magnitude as the Internet connectivity divide. They do so to make the (valid) point that the Digital Divide is a slogan rather than a policy principle, and that there is little more evidence that bridging the gap in Internet connectivity is important for development and poverty reduction, than there is that bridging the air conditioning gap would be important.

But they may have it exactly wrong. Thomas Hughes suggested that it is best to think of electricity in terms of a technological system. The economic benefits from Edison’s inventions only really became visible in the economic statistics when electric motors had been connected to the electrical grid in addition to lights, and when those motors began to be used in trollies, factories, refrigerators, etc. I would point out that computers can be seen as representing a further extension of that system.

The value of electricity is not measured in terms of the number of buildings connected to the grid, but rather in terms of the number of appliances. Better, it is measured in terms of the economic value of the activities made possible or more efficient by those appliances. So it may be that value of the air conditioners, personal computers, refrigerators, electric stoves, radios, TVs, and other appliances better measure the “electricity divide” than do the number of electrical connections to houses.

It may well be that the Digital Divide will best be measured in terms of the differences of economic value of the activities made possible or more efficient by the national information infrastructures – e-commerce, e-government, telemedicine, distance education, etc.

In my previous posting on “common” versus “uncommon” knowledge, I tried to point out that our common experience is a poor guide to the nature of some of the more important ICT applications to society. The satellite remote sensing that helps predict weather and agriculture productivity is an example, as are the supercomputers that are used to design high performance aircraft or automobiles, and the computerized tomography devices used in teaching hospitals. Again, the total value of such technology might well be a better measure of the digital divide, than the numbers of household telephones.

I suspect further that advanced rich countries will add value in these ways faster than most poor countries, and that the Digital Divide measured in these ways is real and will be with us for a while!

No comments: