Barriers to Prisoners’ Reentry into the Labor Market and the Social Costs of Recidivism

David F. Weiman Barriers to Prisoners’ Reentry into the Labor Market and the Social Costs of Recidivism ARE PRISONS CRIMINOGENIC? A L T H O U G H T H E P R IS O N W AS ORIGINALLY C O N C E IV E D FO R T H E NOBLE purpose of rehabilitating crim inal offenders, critics from its very incep­ tion worried that the cure was worse than the disease (Rothman 2005 [1971]; Garland 1993).' In m odem parlance they regarded the prison as an inherently criminogenic institution, which reinforces the criminal behaviors of its occupants. This line of criticism is m ost relevant today, w hen a potent rationale for current US crim inal justice policies, which I term mass incarceration, is the public safety benefits from incapaci­ tating actual offenders and deterring potential ones (see, for example, Levitt, 2004). If the prison experience actually hardens inm ates into more serious offenders, then the incapacitation effect is at best transi­ tory (see also Freeman, 1995; 36 and Blumstein, 1998; 130-31; on deter­ rent effects, see Nagin, 1998, Spelman, 2000, and W estern, 2006). As Jeremy Travis (2005) fervently reminds us, the vast m ajority of inm ates “all come back” to their com munities and so can in principle perpetrate m ore harm than they com m itted prior to their initial incarceration. This classic argum ent identifies a potential social cost of mass incarceration, which, if large, can negate the conventional utilitarian benefit-cost argum ent in its favor. To elaborate my point, consider the social research Vol 74 : No 2 : Summer 2007 575 not too hypothetical case of a nonviolent marginal drug offender, who is sentenced to prison rather than receiving a nonincarcerative sanc­ tion to a drug court. If the individual faces a greater likelihood of recidi­ vism because of his prison record, then the total social costs o f these “get tough” policies should also include the additional harm from his postrelease crim inal activity and the adm inistrative costs related to his subsequent arrest, prosecution, and im prisonment. Even this m ore com prehensive approach significantly under­ states the social costs of mass incarceration. It does not begin to reckon the m anifold losses directly inflicted on m arginal drug offenders, which one epidemiologist has estim ated in term s of “years of life lost” (Drucker, 2002). And it discounts other unintende d negative social im pacts, ranging from the disruptions to and burdens on families (especially children), the erosion of neighborhood social capital (and so weaker inform al social control mechanisms), and political alienation and distrust of public authority (Pattillo, Weiman, and Western, 2004; Hagan and Dinovitzer, 1999; Rose and Clear, 1998; Clear, Rose, and Ryder, 2001; Mauer and Chesney-Lind, 2002; Huo and Tyler, 2002; and Travis and Waul, 2004). Prisons can exert a criminogenic influence on inm ates through distinct causal mechanisms. The m ost direct emphasizes peer effects in which inm ates are assimilated into prison culture and gang networks w hether through a social osmosis or sheer survival instinct. As shown by Chen and Shapiro (2004), its im pact is especially powerful w hen inm ates on the m argin are assigned to higher security prisons, where they come into contact w ith more violent repeat offenders. According to their estimates, the placem ent of a m arginal inm ate into a low rather than a m inim um security facility increases his recidivism rates—mea­ sured by the risk of being arrested w ithin three years of release—by 33 percentage points. I focus instead on an indirect m echanism, the im pact of a prison record/experience on form er inm ates’ labor m arket outcom es.1 Like other formative social institutions such as marriage, the labor m arket is integral to the successful reintegration of released prisoners into 576 social research their families and com m unities (Laub, Nagin, Sampson, 1998; Sampson, Laub, and W imer, 2006; W estern, 2006). Their path away from crime and future prison spells—w hat criminologists call desistance—depends critically on em ploym ent, specifically finding and holding a good job (Sampson and Laub...