Academic journal article Journal of Economic Development

The Effect of Youth Labor Market Experience on Adult Earnings

Academic journal article Journal of Economic Development

The Effect of Youth Labor Market Experience on Adult Earnings

Article excerpt

(ProQuest: ... denotes formulae omitted.)


Explaining wages and their distributions has been, and continues to be, a central theme in the labor economics literature. Recently, the effects of youth and early adult labor market experiences has increasingly become a concern as the current recession has had severe negative impacts on labor markets with the concentration being among younger individuals. Moreover, the effects of the recession seem to have exacerbated a prior trend: the employment to population ratio in the U.S. for youths aged 16-19 declined from 45.2% in 2000 to 34.8% in 2007 then further declined to 25.8% in 2011 (Fernandes-Alcantara, 2012). Early studies of the effects of working while young (particularly while in school) found substantial gains in adult earnings (see Hotz et al,. 2002, for a review). However, a reassessment by Hotz et al. (2002) found these earlier results were spurious, and after controlling for selection, the effects were either dramatically reduced or disappeared altogether. This paper takes a second look at these effects with a newer cohort and extends the literature by focusing on the number of unique youth jobs rather than pure experience. The findings indicate that even after controlling for time spent in the labor market as a youth, having multiple jobs leads to increases in adult earnings, though a possible spurious explanation cannot be completely ruled out.

There are several reasons to suspect working as a youth should have a positive causal effect on adult outcomes. It likely develops responsibility, some appreciation for work, and some idea of what will be expected of them in future working environments. Furthermore, having multiple jobs might increase this return as individuals not only learn about their skills and interests, but also gain experience in actual job search thus increasing efficiency of search in the future. Alternatively, working as a youth might direct inputs away from educational attainment that would have had larger benefits. In addition, a stable work experience with fewer jobs might lead to greater specific skill development implying multiple jobs as a youth might have negative impacts on adult earnings. This last point is supported by a study by Neumark (2002) who finds that job stability in early career leads to increased adult earnings later in life.

This paper estimates the effect of the number of youth jobs on adult earnings using data from the 1997 National Longitudinal Survey of the Youth (NLSY). Initially I investigate this relationship with multiple regression specifications controlling for important covariates. However, a central issue in addressing the question of the causal effect of youth labor experience is self-selection. Though a clear correlation shows up in the data under multiple specifications that control for youth experience and number of adult jobs, it may simply be that individuals who undertook many youth jobs have other unobserved characteristics which themselves improve individuals' incomes in adult life. As noted by Schoenhals, Tienda and Schneider (1998) a complex array of background characteristics affect youths' decisions to work and it is possible these have lasting effects on adult earnings. Such endogeneity concerns are common. To address this concern this paper also employs a partial identification method stemming from work by Manski (1989, 1990, 1997) and Manski and Pepper (2000) and is organized as follows. Section 2 introduces the data. Section 3 reports and discusses regression results. Section 4 discusses the partial identification strategy. Section 5 discusses estimation and bounding results and section 6 concludes.


The data used in this study comes from individual respondents from the 1997 NLSY. The NLSY is a nationally representative sample of nearly 9,000 men and women in the U.S. born between 1980 and 1984 with minorities over-represented. This is newer data and a younger cohort than used in earlier studies on related topics (both Hotz et al. …

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.