Date created: | Last Updated:

Category: Project

Description: Direct evidence establishes that the existence of a criminal record, even a record devoid of convictions, constitutes an impediment to stable housing and employment. Equally direct evidence establishes that lack of stable housing and employment are risk factors for recidivism. Together, these two truths establish that the existence of a criminal record is a barrier to the finalization of a successful reentry strategy for justice-involved individuals. The unresolved question is, what if anything can be done about it? One proposal that has gained recent momentum is the destruction or suppression of official records of an individual’s criminal justice involvement. Popular across the political spectrum, the concepts of “expungement” and “sealing” has drawn the attention of state legislatures, who are extending eligibility criteria and the scope of records subject to suppression or destruction, and legal aid providers, who are dedicating more of their scarce resources to record-clearing. But does record-clearing work? Can it serve effectively as the final stage of successful reentry by reducing housing insecurity, employment insecurity, and thus recidivism? There is reason for hope, but there is also reason for doubt. All this leads to two questions: First, does expungement or sealing cause stabilization of housing and employment with concomitant reductions in recidivism? Second, should oversubscribed legal services providers dedicate their scarce resources to meeting the vast demand for assistance in record-clearing under state law? This study proposes to answer both of those questions with a two-site randomized control trial (“RCT”) analyzing both expungement and sealing as potential useful reentry tools. In the analysis of expungement, individuals eligible for expungement under state law will be randomized to different levels of service (self-help materials or attorney representation) from oversubscribed legal services providers. In the analysis of sealing (applicable only under the law of one site), individuals eligible for automatic sealing will be randomized into receiving notification of impending sealing or no notification. We anticipate expungement outcomes will differ based on service level. We similarly anticipate that individuals will be more motivated to improve their employment and housing outcomes when reminded of the automatic sealing of their criminal record. This allows us to deploy an instrumental variables design to infer the effect of expungement and sealing on recidivism, housing stability and employment. If the expungement or sealing remedies are successful tools in a reentry strategy, then the present momentum toward greater access to these tools should continue. We propose a seven-year timeline which includes eighteen months of enrollment and randomization of eligible participants and five years of follow up to determine recidivism, housing and employment outcomes. Our sites in this RCT will be Western Pennsylvania and the state of Kansas. Sealing is only available under Pennsylvania law and thus will be the only site involved in the sealing portion of the study. Our partners will be Kansas Legal Services (Kansas) and Neighborhood Legal Services Association, the University of Pittsburgh School of Law, and Duquesne University School of Law (Pennsylvania).