By Charles Kenny
May 16, 2011
Save your money, United Nations -- the developing world doesn't need broadband Internet to get ahead.
Expanding broadband access is international development's latest cause célèbre, and it's easy enough to see why. The incredibly rapid spread of cell phones has enabled residents of the developing world to text crop prices to market and arrange bank credit; the Arab Spring has been fueled, filmed, and mobilized using Facebook, Twitter, and YouTube. It's a tempting story line: If basic communications technologies like mobile phones are good, and narrowband Internet has even more uses, surely broadband Internet is the secret to unlocking global progress. If micro-entrepreneurs in rural Africa can access global markets through their flash-enabled websites, and if education and medical care can be imparted through streaming video technology, surely the world's poor can leapfrog into prosperity.
Indeed, that's the conclusion of a high-level U.N. panel that issued a report on the subject during last September's United Nations summit on the Millennium Development Goals. The Broadband Commission for Digital Development, co-chaired by Mexican telecom magnate Carlos Slim -- the richest man in that country's history -- and Rwandan President Paul Kagame, issued a "Declaration of Broadband Inclusion for All." Members of the wide-ranging commission -- including billionaire Richard Branson, the heads of UNESCO and other international organizations, the Senegalese singer Youssou N'Dour, and high-profile development types such as Jeffrey Sachs and Muhammad Yunus -- all joined in the chorus for broadband. "In the 21st Century, broadband networks must be regarded as vital national infrastructure -- similar to transport, energy and water networks, but with an impact that is even more powerful and far-reaching," the report states. Many development organizations have jumped on the broadband bandwagon as well, declaring public subsidies for the spread of fiber-optic cable across regions from Africa to Central Asia a potential funding priority.
But with such a wide spectrum of the world's thought leaders backing the cause, it might be a surprise that the broadband report is rather thin on specifics and evidence. Its 69 pages have space for 58 photographs of commission members, but not one footnote to an academic article. In fact, the distance between the report's optimism and the existing literature on broadband's impact on development could be measured in astronomical units. To suggest the commission's conclusions were half-baked would be an insult to the frozen dinner roll.
It's true that information and communication technologies are having a big impact in the developing world. Ten African markets alone generate over $1 billion in mobile-service revenues each year, and the total for the continent is about $45 billion. Already, more than half of Internet users and nearly half of all broadband subscribers live in low- and middle-income countries. Information technology and business process offshoring (things like operating call centers and records management overseas), was close to a $100 billion industry in 2009, according to the U.N. Conference on Trade and Development (UNCTAD) -- and much of those revenues were captured by developing countries. India's share of the industry alone has grown from $1 billion in 1990 to almost $60 billion in 2009.
But just because information and communication technologies in general are having a big development impact, and broadband in particular is spreading rapidly, doesn't mean that digging trenches and laying fiber-optic cables are the fastest way to a world free of poverty. For a start, there is little evidence that ubiquitous speed and nationwide broadband networks is necessary for firms to benefit from the opportunities offered by offshoring through the Internet. India's booming IT industry claims by far the largest share of the global business process offshoring market: 35 percent, according to UNCTAD. But India also ranks 114 in the world in terms of average connection speed, according to a survey by global consultancy Akamai.
Second, there is precious little evidence linking broadband rollout to economic growth. Sadly, in the commission's report one of the few brief nods to actual studies of the impact of broadband notes that "estimates suggest that for every 10 per cent increase in broadband penetration we can expect an average of 1.3 per cent additional growth in national gross domestic product." That is indeed what one unpublished study claims, but the analysis supporting it is tenuous. The estimate is based on economic growth rates and broadband penetration from 1980 to 2006. Countries with more broadband also grew more rapidly on average, reports the study. But for most of that period, broadband didn't even exist. So, the considerably more plausible interpretation of the analysis is that countries that grew faster between 1980 and 2006 could afford more rapid rollout of broadband as it became available in the new millennium.
Most of the more robust studies of broadband's economic impact have taken place in the United States. In 2006, broadband accounted for about $28 billion in U.S. Internet service-provider revenue. Economists Shane Greenstein and Ryan McDevitt estimate that consumers would have been willing to pay $5 billion to $7 billion more than they did for the benefit of access. This "consumer surplus" and Internet service provider revenues together were equivalent to just over 0.1 percent of U.S. GDP -- this at a time when nearly half of American households had broadband. With regard to broadband subsidies in particular, Ivan Kandilov and Mitch Renkow at North Carolina State measure the impact of a U.S. government loan program that provided subsidized capital to telecom companies to roll out broadband in rural areas. They can find no evidence that the program had an impact on employment, payrolls, or business establishment.
The U.N. report suggests potential impacts far beyond economic growth, however. For example, it suggests that "given that there are rarely enough health practitioners to serve everyone in need of healthcare … broadband is needed to enable doctors to share images and diagnose patients hundreds of miles away using technologies such as video-conferencing." One might question the relative importance of "advanced e-health" when fully one-third of the 10 million child deaths in low-income countries that occur each year could be prevented with technologies as simple as oral rehydration therapy, breast-feeding, and insecticide-treated bed nets.
Again, broadband "offers a potential solution in the ability to deliver education in developing and developed countries alike," according to the report. But you have to wonder whether paying for fiber optics and computer labs is worth it in countries where, for 50 cents per child per year, deworming pills could reduce the incidence of illness-causing parasites and improve school absence by 25 percent. And you have to wonder even more when existing studies actually suggest that more widespread broadband use in schools may correlate with lower test scores. It's not much of a surprise, given that the studies found that the most popular activities for boys online were email, chat, Myspace, YouTube, music, and games. Broadband rollout to households is also associated with lower test scores -- almost certainly for the same reason.
Don't expect governance to get much better with broadband either. Jed Kolko at the Public Policy Institute of California analyzed U.S. survey responses in 2006 and found that while broadband users were more than twice as likely than dial-up users to say they had downloaded music, gone to a social networking site, or visited an adult entertainment site, the chance that they reported visiting a government website was only 6 percentage points higher (at 26 percent). We don't have good survey evidence on broadband usage in poor countries, but University of California, Berkeley's Kentaro Toyama suggests that the dominant use of the Internet in cybercafes and public Internet facilities in the developing world "is by young men playing games, watching movies, or consuming adult content." That's great for the entertainment industry, but it is hard to see how it's going to improve health, education, or the quality of developing-country institutions.
Undaunted, the U.N.'s broadband commission calls for public investments and subsidies in the technology to help reach the Millennium Development Goals. But there are a number of reforms that developing countries can introduce to extend access to broadband networks more cheaply: McKinsey estimates that a combination of adding to available spectrum, encouraging infrastructure- and spectrum-sharing, and reducing regulatory burdens could collectively cut broadband costs by as much as 50 percent. Meanwhile, the public money that the commission calls for would be much better spent on the simpler, cheaper, proven, and effective ways to meet global targets for health and education -- many of which today go unfunded. Diverting already scarce resources to ensure universal access to high-definition Colbert Report repeats on YouTube is not a solution to global poverty.