May 5, 1998


by Andy Oram
American Reporter Correspondent

CAMBRIDGE, MASS.—Many countries, whether considered to be “developed” or “underdeveloped,” believe poverty can be overcome by Internet access. Therefore, when investigating the hope of universal service, it is well worth looking at South Africa, which combines a developed and an underdeveloped nation in one.

Whites and a growing number of entrepreneurial non-whites enjoy affluence and a high degree of education. Meanwhile, 70% of the population lacks access to running water, and at least a quarter are estimated to be malnourished. Half of the country is illiterate. A huge number of blacks received virtually no formal education, having been of school age during a time of turbulence and boycotts. Unemployment is high.

In this context, it is quite audacious—but well in keeping with its liberal populism—for the ruling African National Congress to set a goal of bringing advanced information services to everyone. Where most of the world is moving to a free market, ethnic stratification, and a growing gap between rich and poor (in short, to a situation more and more resembling South Africa of the apartheid era) the Mandela government is pressing on in the opposite direction.

An email address ending in .za, to be sure, cannot bring people food, medicine, or clean water. But it can provide some more subtle benefits: modern education, advice about finding jobs or sales opportunities, and information about disease prevention.

As in other countries, the telephone system is the substructure for providing Internet access. Telkom SA Limited, the South African telephone company, is like many such companies around the world, a technologically slow-moving monopoly. Local calls, as in much of Europe, have per-minute costs.

And now the company wants to extend its monopoly to Internet service. In its submission to the regulatory authority, the company calls the Internet Protocol a kind of transport service and claims that its exclusive right to offer all kinds of “access” to the telephone network includes Internet access. For existing Internet service providers, the submission magnanimously leaves “higher-level” or “value-added” services like hosting Web pages and routing email.

This proposal, which would bring laughter to the faces of Internet users in most of the world, hasn’t flown very far in South Africa either. The South African Telecommunications Regulatory Authority, which plays a role similar to the FCC in the United States, has denied the request. It is also opposed, of course, by the two dozen members of the country’s Internet Service Providers Association.

Furthermore, a government White Paper on telecommunications, released on February 5, lays out the goal of gradually lifting the Telkom monopoly on various services and explicitly grants monopoly (or “exclusivity”) only on the basic telephone network; Telkom’s proposal goes squarely in the opposite direction.

But Telkom has an interesting argument: it claims that a monopoly over Internet service would let it bring such service to poor and rural areas faster—in other words, to carry out the government’s universal service goals. Its major tool is cross-subsidization, notably “geographical averaging,” in which urban areas pay more so that rural areas can pay less. Telkom accuses the competing Internet providers of playing the opposite game, cherry-picking (aiming at highly-concentrated, high-paying customers).

There are no grounds for these arguments. Monopolies in technologically developing industries are almost always less efficient and more costly than competitive markets. The ISPA, in fact, charges that Telkom charges artificially high costs for basic telephone service (not subject to competition) in order to offer unfairly low costs for Internet access (which is subject to competition). The providers point out that most of the world is moving away from cross-subsidization because it causes “pricing distortions.” They suggest that if costs were fair, they could extend their service throughout the country.

But SATRA has to find a positive note with which to counter Telkom’s argument. Their reply promises, “a regulatory environment that will require Internet service providers (ISPs) to contribute to the goals of universal service.” However, no details have been offered. One strategy, not enunciated by the government but consistent with its statements, would be to make ISPs pay a percentage into the universal service fund.

The Internet service providers, in turn, promise that, “The ISPA is committed to establishing projects which provide services to historically disadvantaged communities” (that is, non-whites).

South Africa is having trouble finding the way forward, but no more than richer countries like the United States. Our own universal service plan for Internet access is relatively minimal—and even so, it’s in trouble.

The U.S. plan involves payments from telephone companies’ universal service fund to projects chosen by the government to wire up schools, libraries, and health care providers. In short, a tiny fraction of the prices charged for phone calls are supposed to go to these public institutions that benefit the public directly.

Even that modest plan has been challenged in court by the telephone companies and called in question by Senators from largely rural states. Furthermore, many thoughtful analysts who have no ideological objection to income transfers say that universal service is not achieved efficiently by such cross-subsidies.

If knowledge is power, access to advanced information systems is useful to the poor. Even one terminal in a rural community (perhaps hooked up in one of the few buildings that boasts electricity) can be a valuable resource.

In fact, the way to universal Internet access in the U.S. as well as South Africa will probably be through community centers that offer both training and computers. Many cities in the U.S. already have such community centers. A line to every home is a valid ideal, but we can get to universal service much faster with a more public approach.

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Editor, O’Reilly Media
Author’s home page
Other articles in chronological order
Index to other articles