Insights / Research BriefOct 19, 2023

Labor Market Impacts of Reducing Felony Convictions

Dmitri Koustas, Amanda Y. Agan, Andrew Garin, Alexandre Mas
Retroactively reducing felonies to misdemeanors leads to employment benefits forindividuals who (likely) requested their reductions, but no benefits among the larger subset of individuals whose records were reduced without their involvement.

People who experience run-ins with the criminal legal system tend to earn less than people who do not, a pattern that contributes to the pervasive inequalities that exist between racial and ethnic groups, income levels, and neighborhoods in the United States. Understanding and remediating the harm associated with the criminal legal system is an important area of inquiry among economists, who today are focused on questions such as: Can avoiding a jail sentence improve defendants’ earnings? Or, are earlier interventions more effective? Are jobseekers with criminal records less likely to get callbacks from potential employers?

In this paper, the authors study whether retroactively reducing felony convictions to misdemeanors can improve defendants’ employment prospects. California’s Proposition 47, which became effective in 2014, reclassified certain theft and drug possession felonies to misdemeanors. Under the policy, individuals with eligible offenses could petition to have their previous felonies reclassified as misdemeanors, with an estimated one million Californians eligible for a record reduction under the law. In this context, the authors study whether defendants who had their sentences reduced through Proposition 47 experienced better employment opportunities as a result.

It’s possible that the defendants who petition to have their felonies reclassified are different than those who do not in ways that would otherwise be related to labor market success, even in the absence of felony reductions. To identify the causal effects of Proposition 47 absent these other factors, the authors exploit the fact that following the passage of Proposition 47, legal agencies in San Joaquin County worked together to reduce tens of thousands of eligible felonies to misdemeanors without involvement from eligible individuals. The ordering of these publicdefender- initiated reductions was essentially random, allowing the authors to compare employment for individuals before and after their reductions among a non-selected sample.

To assess the impacts of these interventions on defendants’ labor market outcomes, the authors link data on Proposition 47 reductions in San Joaquin County to tax return data. They find the following:

  • Public-defender-initiated reductions from felonies to misdemeanors do not appear to improve defendants’ employment, selfemployment, or tax filing. There is evidence of increases in the prevalence of platform gig work, however, as well as suggestive evidence of positive employment increases for individuals whose reductions occurred less than four years after their convictions.
  • The authors also measure the impact of felony reductions among defendants whose petitions were not filed by public defenders, but most likely by the defendants themselves. Unlike for public-defender-initiated reductions, these reductions result in a 12% increase in the probability of having any wage employment in the year of the reduction.
  • There is evidence that the group of defendants who petitioned to have their own records reduced would have had better labor market outcomes than the group of defendants whose records were reduced by public defenders, regardless of whether or not the actual reduction took place. The authors show that (likely) self-petitioners experience an upward employment trajectory in the run up to their reduction, suggesting that many in the group were already experiencing relative success in the labor market when their reduction took place.

What explains these results? One possibility is that many individuals whose records were reduced by public defenders were unaware of the change, and therefore did not increase their job search efforts in response. By contrast, individuals who know about or voluntarily seek remediation of their records may feel more confident when searching for jobs. To test this hypothesis, the authors designed a randomized controlled trial in collaboration with criminal legal agencies in San Joaquin County in which they notified a random subset of individuals about their defenderinitiated reductions. Their experiment yielded the following result:

  • Individuals chosen for notification did not experience detectable improvements in labor market outcomes compared to those not chosen for notification, suggesting that a lack of knowledge about a Proposition 47 reduction is unlikely to be the main driver of results shown here.

These findings are consistent with recent research showing that defendants facing a sentence are already sufficiently detached from the labor market that merely avoiding jailtime has little effect on their long-run employment trajectory. The subjects of this study—defendants with criminal records—are also relatively disadvantaged to begin with, such that retroactive reductions have little effect.

Since the passage of Proposition 47, many other states, including Utah, Connecticut, Alaska, Rhode Island, and Oklahoma, have adopted measures reclassifying drug possession from a felony to misdemeanor, with policy advocates arguing these policies can improve outcomes for impacted individuals. While this research does not rule out the benefits of these policies, it does suggest that policies addressing upstream factors, such as other criminal legal interactions including conviction and arrest, human capital, or broader environmental and social influences may be more effective at reducing the economic harm associated with the criminal legal system.