72 Pages Posted: 27 Aug 2014 Last revised: 5 Feb 2017
Date Written: February 6, 2017
An estimated one in three American adults has a criminal record. While some records are for serious offenses, most are for arrests or relatively low-level misdemeanors. In an era of heightened security concerns, easily available data and increased criminal background checks, these records act as a substantial barrier to gainful employment and other opportunities. Harvard sociologist Devah Pager describes people with criminal records as “marked” with a negative job credential.
In response to this problem, lawyers have launched unmarking programs to help people take advantage of legal record clearing remedies. We study a random sample of participants in one such program to analyze the impact of the record clearing intervention on employment outcomes. Using methods to control for selection bias and the effects of changes in the economy in our data, we find evidence that: (1) the record clearing intervention boosts participants’ employment rates and average real earnings, and (2) people seek record clearing remedies after a period of suppressed earnings.
More research needs to be done to understand the durability of the positive impact and its effects in different local settings and labor markets, but these findings suggest that the record clearing intervention makes a meaningful difference in employment outcomes for people with criminal records. The findings also suggest the importance of early intervention to increase opportunities for people with criminal records. Such interventions might include more legal services, but they might also include record clearing by operation of law or another mechanism that does not put the onus of unmarking on the person with a criminal record.
Keywords: criminal records, ban the box, reentry, mass incarceration, legal services, clean slate
Suggested Citation: Suggested Citation
Selbin, Jeffrey and McCrary, Justin and Epstein, Joshua, Unmarked? Criminal Record Clearing and Employment Outcomes (February 6, 2017). Available at SSRN: https://ssrn.com/abstract=2486867 or http://dx.doi.org/10.2139/ssrn.2486867