Politicians and economists in developing countries searching for new technologies to create jobs and spur economic growth need look no further than their desks. The most vital technology for sparking development is a familiar and unglamorous one: the telephone. In many poor nations, telephone service is available only in large cities-at a price few can afford- and the more widely available mobile phone service remains expensive. As a result, at least 1.5 million villages in poor nations lack basic telephone service.

Guatemala has just 65 telephones for every 1,000 people; Pakistan, 23; Nigeria, 5; and Burma, 4. By comparison, the United States has 667 telephones per 1,000 people. Manhattan alone boasts more telephone lines than all of Africa.

During the 1990s economic boom, many developing nations invested in laying fiber-optic lines, building satellite relay stations, and connecting to transoceanic cable-the high-capacity ''backbone'' elements of telephone networks that transport data. So why does the 128-year-old telephone remain out of reach for more than 3 billion people? In part, because the cost of bridging the ''last mile'' from national network to local customer vastly exceeds potential returns in countries such as Colombia, where annual per capita spending on telecommunications is just $231 (in the United States, it's $2,924).

Two new technologies offer a potentially quick solution: wireless-fidelity networks (Wi-Fi) and voice calling over the Internet (VoIP). Wi-Fi uses small, low-power antennas to carry voice and data communications between a backbone and users at schools, businesses, and households, all without laying a single wire, greatly reducing the cost of traversing the last mile.

Laying land lines can cost up to $300 per foot. Wi-Fi hardware is fitted to existing structures for about $10,000 per base station-a reasonable sum, considering that one Wi-Fi station can provide access to thousands of residences within two miles and that the antennas that attach to customers' homes cost less than $100.

VoIP technology sends telephone calls over the Internet inexpensively by transforming people's voices into data ''packets.'' Conventional phone service requires an open line at either end of a call-an expensive service, not least because every conversation pause wastes bandwidth. By chopping words and pauses into tiny packages that are routed through the least congested part of the Internet, computers make VoIP calls much cheaper. In the United States today, a phone call using VoIP service costs less than half of a call made using traditional telephony; these savings can be duplicated in developing countries.

Together, Wi-Fi and VoIP can make telephone service affordable and accessible in poor countries. But for developing nations to benefit, their governments must rethink who owns the telecommunications networks. Put simply, it's a bad idea to have a monopoly, whether government or private, both control the network backbone and provide retail services to consumers. Such arrangements lead to higher prices and less competitive services.

Consider Telkom, the owner and operator of South Africa's telephone network, a formerly state-owned monopoly that was privatized between 1997 and 2003. Despite enjoying an advanced network backbone, Telkom does not offer basic telephone service to a majority of South Africans. Because it depends on revenues from phone calls, Telkom has little incentive to offer cheap VoIP service. South African law dictates that only Telkom and ''under-serviced area licensees'' (small firms in rural areas) are allowed to offer VoIP, yet the government has not approved a single under-serviced area licensee. So today, for a variety of regulatory reasons, only Telkom can provide VoIP. For competitive reasons, it does not.

Developing countries can break such strangleholds by renationalizing their network backbones, liberating them from the retail business of servicing consumers. Although state monopolies provided infamously poor service, running a network core is easier than providing retail services. State-owned network backbones can operate on a non-profit basis, providing access to private companies that compete to service local customers in villages and towns. It's not that the ordinary bias favouring private ownership and free markets is misguided. Nor are telecommunications networks too critical a public service to be left to free markets. Rather, networks in developing countries have never been subject to real competition. Ironically, a publicly owned backbone would level the playing field and increase competition among retail providers, leading to innovative services at lower prices.

One model for success can be found in Utah, where authorities in Salt Lake City and 17 surrounding towns have formed the Utah Telecommunications Open Infrastructure Agency (utopia), building a high-speed network for 250,000 households and 35,000 businesses. The government owns the backbone, but does not sell Internet or VoIP service directly to customers. Instead, utopia is open to anyone wishing to sell broadband service.

Can the same model work in the developing world, where money and accountability are more elusive? Yes, for two reasons. First, Wi-Fi and VoIP flip the traditional telecommunications model on its head. The network backbone has only one objective (delivering data via a small set of universal procedures), leaving governments with a simpler job. Delivering local service is harder. Traditional telecommunications models are the opposite: The telephone is simple; the circuit-switched network is complex. And while a private monopolist has every incentive to charge an exorbitant price and increase profits at the expense of consumers, a public monopoly lacks that impulse. Nonetheless, to ensure that consumers benefit, an independent, nonprofit organization could jointly administer the backbone network with a government agency. To increase efficiency, the daily operations of the backbone could be leased to a private entity.

Such renationalization of network backbones would be expensive for developing countries, but the costs are not insurmountable. Governments could buy back network backbones using long-term debt funded by revenues flowing from the operating lease. A properly structured public debt issuance would assuage foreign investors' fears of a broader nationalization campaign.

In developing countries, telecommunications lead to more jobs, improved health care, and higher levels of education. The renationalization of telecommunications backbones is analogous to the state-funded building of roads. Roads and highways increase a nation's wealth by enabling commerce. In poor nations, the same can be true of the information superhighway, if politicians choose technology over ideology.

April 21, 2004.

Chris Sprigman is a fellow at the Stanford Law School Center for Internet and Society.
Peter Lurie is general counsel at Virgin Mobile USA.

[This article has appeared in Foreign Affairs March/April 2004.]