June, 2001, Vol. 124, No. 6
Computer use and older workers
Biology of unions
Précis from past issues
Computer use and older workers
Workers who use computers on the job retire later than those who don’t. This raises the following question: Do workers with computer skills choose to delay retirement or do workers who plan on later retirement choose to learn computer skills?
Leora Friedberg of the University of Virginia attempts to discern the effects of computer skills on retirement in "The Impact of Technological Change on Older Workers: Evidence from Data on Computer Use" (National Bureau of Economic Research Working Paper 8297). She examines data on computer use from the monthly Current Population Survey (CPS) and from the longitudinal Health and Retirement Study (HRS).
The CPS data come from the October surveys of 1984, 1987, 1993, and 1997, which asked workers: "Do you directly use a computer at work?" Trends in the evolution of computer use that are apparent in the CPS data suggest that impending retirement plays an important role in computer use among older workers.
With the HRS data, Friedberg estimates the impact that computer use has on retirement. Her results indicate that computer use itself leads to later retirement. She finds that, although they are not precise, the estimates imply that computer use has an independent effect on retirement, increasing the likelihood of continuing to work by up to 25–30 percent during a 4-year period.
Biology of unions
In "A Biological Model of Unions" (National Bureau of Economic Research Working Paper 8257), Michael Kremer and Benjamin A. Olken of Harvard University take the principles of evolutionary biology and apply them to the study of unions. The researchers note that "biological models suggest that selection pressure often works against organisms that are too harmful to their hosts." Kremer and Olken apply this concept to the relationship between unions and firms. They argue that unions that obtain the level of wages that are optimal for their current members will be displaced in competition with unions with more moderate wage policies; the more restrained policies of such unions allow their firms to live longer.
Kremer and Olken’s model also indicates that industries with high turnover rates for firms will have low unionization rates. This may help explain the higher unionization rates now found in the public sector relative to the private sector.
To test their model, the authors analyze data on union membership by industry from the 1983–91 Current Population Survey and data on industry characteristics from the 1992 Census of Manufactures. They find support for their model—for example, they report that "across a variety of specifications, higher firm exit rates are associated with lower unionization rates."
Kremer and Olken’s model has implications for the future of unions in the United States. If business is becoming more competitive and if this means higher turnover rates for firms, then the decline in unionization that has been seen in the past two decades could be intensified.
When economists have spoken about the importance of skills, the tendency has been to skip straight to measures of cognitive skill, such as years of schooling or test scores. In fact, however, cognitive skills can determine only part of a person’s success in the labor market or other aspects of economic life. A session of the 2001 annual meeting of the American Economic Association presented some of the exploratory research addressing the issue of accounting for the benefits of skills, both cognitive and noncognitive. The four papers recently appeared in the May 2001 Papers and Proceedings of American Economic Review.
In "The Importance of Noncognitive Skills: Lessons from the GED Testing Program," James J. Heckman and Yona Rubenstein find that the General Educational Development (GED) testing program provides a mixed signal. Dropouts who take the GED do have higher levels of cognitive skill than those who do not, but once ability is measured and controlled for, GED recipients actually have lower earnings than other dropouts. They attribute this to unmeasured differences in noncognitive skills, such as reliability and persistence.
Rachel Dunifon, Greg J. Duncan, and Jeanne Brooks-Gunn, in "As Ye Sweep, So Shall Ye Reap," examine the role of organization and efficiency in affecting earnings and other outcomes over two generations. Using a measure of household cleanliness as a proxy for those traits, they find that "the cleanliness rating of one’s home is predictive of: (1) one’s own earnings 25 years later; (2) children’s subsequent completed schooling; and (3) children’s earnings measured 25 years later."
In "Incentive-Enhancing Preferences: Personality, Behavior, and Earnings," Samuel Bowles, Herbert Gintis, and Melissa Osborne look at the impact personality traits might have on earnings. For example, among women, a one standard-deviation increase in "fatalism" as measured by the "Rotter ‘locus of control’ scale" has the effect of reducing wages by 6.7 percent.
The final paper, "Understanding, Speaking, Reading, Writing, and Earnings in the Immigrant Labor Market," by Anthony P. Carnevale, Richard A. Fry, and B. Lindsay Lowell looks at a wider selection of cognitive skills than earlier, Bureau of the Census-based research has been able to do. While the Census survey asks only about speaking ability, the data from the National Adult Literacy Survey allows analysis of reading, writing, and understanding as well. The authors find that only one of these four aspects of language acquisition skills—the ability to understand the spoken word—has a significant impact on wages.
We are interested in your feedback on this column. Please let us know what you have found most interesting and what essential reading we may have missed. Write to: Executive Editor, Monthly Labor Review, Bureau of Labor Statistics, Washington, DC. 20212, or e-mail MLR@bls.gov
Within Monthly Labor Review Online:
Welcome | Current Issue | Index | Subscribe | Archives
Exit Monthly Labor Review Online:
BLS Home | Publications & Research Papers