Academic journal article Contemporary Economic Policy

Does Improving School Quality Reduce the Probability of Unemployment?

Academic journal article Contemporary Economic Policy

Does Improving School Quality Reduce the Probability of Unemployment?

Article excerpt


There is an ongoing debate among economists about whether improvements in high school quality, such as increasing per pupil expenditures or lowering class size, raise post-schooling earnings (e.g., Betts 1995; Card and Krueger 1992; Grogger 1996; Hanushek 2004). This article fills a noticeable gap in the existent literature on the effects of high school quality on labor market outcomes. Previous studies have focused on employed individuals, but the effect of school quality might also affect labor market attachment in general. This article explores whether high school quality affects labor market attachment as measured by the probability of being unemployed.

There are a number of potential links between school quality and unemployment probabilities. For example, in a standard human capital model, any school quality measure that enhances human capital formation would logically affect both wages and hours worked, with unemployment being an important special case of zero hours worked. To add a little more structure, suppose one models the employment decision within a search-theoretic framework with an individual searching for a job with a wage in excess of a reservation wage. School quality could plausibly affect both the distribution of wage offers and the reservation wage, thus making the effect of school quality on the probability of being unemployed theoretically ambiguous.

School quality might also affect the matching of students to jobs. Higher school quality could plausibly be positively correlated with dissemination of better information to students about their own interests and abilities, thus leading to better employment matches. Similarly, improved school quality could lower the search cost for obtaining employment by providing better job search skills, fostering relationships with potential employers, or providing information about potential employment opportunities.

This article presents estimates of the empirical effects of high school quality on the probability of unemployment using standard specifications which are comparable to the existing literature on school quality. (1) The remainder of the article is as follows: Section II discusses the data and the estimation approach, section III presents the results, and section VI concludes.


The authors use the 1980 Sophomore cohort of the nationally representative High School and Beyond (HSB) data set collected by the National Center for Education Statistics. The HSB is appropriate for this study because it contains a rich set of family background characteristics, reasonable measures of school quality, and measures of labor market attachment just after high school as well as several years later. The authors focus on men to avoid the confounding effects that intermittent labor market experience may have for women.

The relationship between school quality and the probability of unemployment is examined at three points in time. The analysis initially examines whether school quality affects unemployment probabilities for noncollege-bound men from the sophomore class of 1980 (potentially graduating in 1982) during the time period from September 1982 to February 1984. This time period covers the first 18 months of potential labor market experience following the senior year of high school (somewhat more for those who dropped out of school before the senior year) for men who did not enroll in postsecondary education. This group of young men is interesting to analyze because the years just after high school are the time of the greatest likelihood of unemployment (Ehrenberg and Smith 1996, chap. 13), which suggests that school quality can have its largest impact on unemployment during this period. (2) The analysis also examines labor market attachment for noncollege-bound men in two later periods, September 1984-February 1986 and September 1989-February 1991. The first interview date allows the authors to track 18 months of potential labor market experience while also skipping the first summer after high school graduation. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.