Original Papers
The Effect of Conviction on Income Through the Life Cycle 15

https://doi.org/10.1016/S0144-8188(97)00055-0Get rights and content

Abstract

Existing studies of the impact of conviction on income and employment do not consider life cycle issues. We postulate that conviction reduces access to career jobs offering stable, long-term employment. Instead, conviction relegates offenders to spot market jobs, which may have higher pay at the outset of the career but do not offer stable employment or rapidly rising wages. Thus, first-time conviction may increase the wages of young workers while decreasing the wages of older workers. We test our theory with data on federal offenders and find that first-time conviction has a positive and significant effect on income for offenders under age 25 and an increasingly negative and significant impact for offenders over age 30. These results imply that the present value of income lost as a result of conviction varies over the life cycle, reaching a maximum in the middle of the career. We find that the gains sought by these offenders follow similar profiles, suggesting that prospective offenders may be deterred by the possibility of lost future income. Because the discounted loss in future income facing young offenders may be small, our results may provide part of an explanation of youth crime.

Section snippets

I. Introduction

A growing literature indicates that arrest and conviction limit legal earning opportunities. Recent studies find that the market penalty of lost income is large, both absolutely and in comparison with fines and prison terms [Lott (1992); Waldfogel (1994a)]. Furthermore, the adverse effects of arrest and conviction are at least somewhat persistent [Waldfogel (1994b); Grogger (1992)]. The large and growing involvement of urban youth in crime is thus alarming not only because of the direct harm to

Conviction Effects

Human capital theory [Becker (1964)] predicts rising wages and stable employment in jobs where workers and employers jointly invest in firm-specific human capital. Early in the career, the worker takes some of his compensation as training, so the money wage is below the value of his marginal product. Wages rise over the course of the career as the individual accumulates human capital and experiences a declining implicit reduction in take-home pay to cover training costs. The joint investment of

III. Data

The data for this study are assembled from the administrative records of the Administrative Office of the U.S. Courts (AO). The data consist of a two-observation panel on the legal income of criminals, with one observation before, and one after conviction. Preconviction data are extracted from presentence investigation reports, and postconviction data are taken from monthly probation reports. The primary sample used in this analysis is comprised of males convicted of fraud in the U.S. federal

IV. Analysis

This section examines four empirical questions relevant to the theory advanced above. First, how does first-time conviction affect job stability over the life cycle? Second, does the effect of first-time conviction on income vary over the life cycle so that conviction raises the income of young workers and decreases the income of older workers? Third, is the conviction effect over the life cycle different for workers who already have criminal records? Fourth, do the gains sought through fraud

V. Conclusion

This paper presents evidence that first-time conviction effects vary substantially by age, whereas subsequent conviction effects do not. First-time conviction raises the income of young offenders and reduces the income of older offenders, although subsequent conviction effects reduce income at all ages. The results are rationalized by assuming that first-time conviction moves workers off of career income profiles to less steeply sloped spot market profiles. Subsequent convictions simply

References (21)

There are more references available in the full text version of this article.

Cited by (79)

View all citing articles on Scopus
15

We thank Linda Babcock, Donald Fullerton, Lowell Taylor, Joe Tracy, and seminar participants at Carnegie Mellon and Yale for helpful comments. All errors, of course, remain our own.

View full text