The Internet, Wages, and Consumer Welfare
Broadband's deployment created between $8.3 and $10.6 billion of new GDP in 2006.
The advent of the Internet has had an undeniable impact on the landscape of American business. From dial-up to broadband, with each evolution the Internet has been hailed by some economists and downplayed by others. Widespread evidence indicates that investment in information technology (IT) in the 1990s produced gains in U.S. productivity and economic growth at the national, industry, and firm levels. Equally substantial evidence raises questions about whether the benefits of IT investment were experienced everywhere. In particular, new IT investments had the greatest effects on productivity for industries that were already IT-intensive and for workers with more education and skills. Yet, those findings do not address the issue of the effect of the Internet on regional wages in the United States. Specifically, did wages converge or diverge because of frontier use of IT?
This question had special relevance in the 1990s because the new IT investments of that era-and particularly the rise of the commercial Internet-facilitated long-distance communication. One view hails the Internet as a great enabler of economic growth, particularly for low-density regions, because increased communication breaks the link between local investment, productivity, and wage growth, leading to wage convergence across regions. A contrasting perspective casts the Internet as a technology that exacerbates existing inequalities in wages between urban/rural and frontier/mainstream users of IT, and consequently leads to wage divergence across regions.
In The Internet and Local Wages: Convergence or Divergence? (NBER Working Paper No. 14750), authors Chris Forman, Avi Goldfarb, and Shane Greenstein closely examine the relationship between business use of advanced Internet technology and regional variation in U.S. wage growth between 1995 and 2000. They find that business use of advanced Internet technology is associated with wage growth, but they find no evidence that the Internet contributed to regional wage convergence. Their most interesting finding suggests that the Internet instead caused a divergence of wages. Wage growth and advanced Internet use were more strongly correlated in counties that already were doing well on a variety of measures. In particular, advanced Internet use was especially correlated with wage growth in the 180 counties that, as of 1990, had a population over 100,000 and were in the top quartile in income, education, and fraction of firms in IT-intensive industries.
Overall, while the Internet explains just 1 percent of the wage growth in the average county in their sample, it explains 25 percent of the difference in wage growth between the 180 counties that were already doing well and all other counties. The authors also find little evidence that use of advanced Internet technologies was associated with growth in either employment or establishments.
Because these results suggest a considerable divide in the benefits of advanced Internet use across urban and rural areas, the researchers feel that the debate about the economic impact of IT must focus on regional variation. Efforts to subsidize rural Internet development would have little impact, they claim, because there is little evidence that the Internet has much impact in rural areas. This runs counter to the motivation for a wide array of policies encouraging Internet business use outside of urban areas, including policies to subsidize rural broadband development.
In a related paper, The Broadband Bonus: Accounting for Broadband Internet's Impact on U.S. GDP (NBER Working Paper No. 14758), Shane Greenstein and Ryan McDevitt examine broadband's economic contribution through its replacement of dial-up Internet access. In September 2001, approximately 45 million U.S. households accessed the Internet through a dial-up connection, while only 10 million used a broadband connection. By March 2006, approximately 47 million households had broadband connections, while 34 million used dial-up. The authors find that while broadband accounted for $28 billion of GDP in 2006 (out of $39 billion in total for Internet access), approximately $20 to $22 billion of that was associated with household use. Of that amount, the authors estimate, broadband's deployment created between $8.3 and $10.6 billion of new GDP.
Greenstein and McDevitt also find that increased broadband use raised consumer surplus by between $6.7 and $4.8 billion. Consumer surplus is the benefit to consumers from purchasing a product at a price that is less than they would be willing to pay. In both cases, this benefit is above and beyond what dial-up would have generated. The authors' estimates of the consumer benefits generated from upgrading to broadband are much lower than those typically quoted by Washington-based policy analysts. These estimates also differ from the CPI (Consumer Price Index) for Internet access. The findings of this study correct a historically inaccurate inference about the pricing of Internet access and lead to the conclusion that the official index's timing of price decline is actually several years too late.
Finally, the authors help to inform understanding about why the national policy of the last decade has had the effects it did. Initially, most federal policy sought to subsidize the deployment of dial-up technologies to less-served areas and users; but, at the outset of the millennium, policy changed. The new policies relied largely on the private incentives of private actors to deploy broadband technologies, without subsidy or any regulatory intervention. In retrospect, these policies seem to have promoted wire line-based broadband diffusion. Yet, this outcome is puzzling in light of the lack of price change measured in the CPI. The authors' findings resolve this puzzle. Price indexes undervalued the gains to users, and these gains were precisely what motivated the upgrade at many households.
-- Lester Picker
The Digest is not copyrighted and may be reproduced freely with appropriate attribution of source.