The Impact of Incarceration on Wage Mobility and Inequality

(2002) - Bruce Western

Journal: American Sociological Review
Link:: http://www.jstor.org/stable/3088944?origin=crossref
DOI:: 10.2307/3088944
Links::
Tags:: #paper #LifeCourse #SocialTheory
Cite Key:: [@Western2002]

Abstract

A life course perspective on crime indicates that incarceration can disrupt key life transitions. Life course analysis of occupations finds that earnings mobility depends on stable employment in career jobs. These two lines of research thus suggest that incarceration reduces ex-inmates' access to the steady jobs that usually produce earnings growth among young men. Consistent with this argument, evidence for slow wage growth among ex-inmates is provided by analysis of the National Longitudinal Survey of Youth. Because incarceration is so prevalent-one-quarter of black noncollege males in the survey were interviewed between 1979 and 1998 while in prison or jail-the effect of imprisonment on individual wages also increases aggregate race and ethnic wage inequality.

Notes

“A life course perspective on crime indicates that incarceration can disrupt key life transitions.” (Western, 2002, p. 526)

“that earnings mobility depends on stable employment in career jobs.” (Western, 2002, p. 526)

“effect of imprisonment on individual wages also increases aggregate race and ethnic wage inequality” (Western, 2002, p. 526)

“The prison boom of the 1980s and 1990s coincided with growing polarization of the American labor market” (Western, 2002, p. 526)

“inequality increased during these decades, and wage de-” (Western, 2002, p. 526)

“dines were particularly large among men with little education (Bernhardt et al. 2001” (Western, 2002, p. 526)

“A small research literature thus finds that incarceration reduces earnings (see the review by Western, Kling, and Weiman 2000).” (Western, 2002, p. 526)

“increases in wage inequality through the 1980s and 1990s, however, the low earnings of ex-convicts may be an artifact of widespread wage stagnation among men with little schooling.” (Western, 2002, p. 526)

“strong causal inference about the negative effect of imprisonment on wages is also threatened by the fact that men with few economic opportunities may turn to crime” (Western, 2002, p. 526)

“The life path of ex-inmates diverges from the usual employment trajectory in which earnings mobility for young men is generated by steady jobs with regular career ladders (Spilerman 1977).” (Western, 2002, p. 527)

“my analysis also controls for declining wages among men with little schooling.” (Western, 2002, p. 527)

“If incarceration slows wage growth at the individual level, the prison boom may have increased wage inequality in the aggregate” (Western, 2002, p. 527)

“Most research relating the criminal justice system to wages focuses on estimating a main effect-a constant decrement in wages attributed to, say, criminal conviction or incarceration.” (Western, 2002, p. 527)

“Longitudinal studies of careers find that internal labor markets in large firms, public sector pay schedules, on-thejob training, and union seniority provisions all contribute to job continuity and earnings growth among young men (DiPrete 1989; Spilerman 197” (Western, 2002, p. 527)

“While life course research on occupations ties earnings growth to employment in career jobs, a life course perspective on crime treats incarceration as a turning point that disrupts key transitions, restricting access to This content downloaded from 109.175.170.229 on Mon, 14 Feb 2022 20:03:32 UTC All use subject to https://about.jstor.org/terms” (Western, 2002, p. 527)

“528 AMERICAN SOCIOLOGICAL REVIEW such jobs (Sampson and Laub 1993).” (Western, 2002, p. 528)

“Three mechanisms explain why prison or jail time is linked to slow wage growth. Incarceration is stigmatizing, and it erodes human and social capital.” (Western, 2002, p. 528)

“Incarceration also erodes job skills. Time out of employment prevents the acquisition of skills gained through work experience” (Western, 2002, p. 528)

“Finally, the social contacts that provide information about job opportunities may be eroded by incarceration” (Western, 2002, p. 528)

“Thus Sampson and Laub (1993:153-68) found that time served in prison by youths aged 17 to 25 was negatively related to continuity of employment and work commitment at ages 25 to 32. Urban ethnographers similarly report that the prison system provides a pathway to secondary labor markets and informal economies (Duneier 1999; SanchezJankowski 1991:281; Sullivan 1989; Hagan 1993” (Western, 2002, p. 528)

“The collective effect of the penal system is captured by Garland's (2001:2) term "mass imprisonment." In his formulation, the incarceration rate is so high for some” (Western, 2002, p. 529)

“groups that its influence is felt not just by individuals, but by broad demographic groups.” (Western, 2002, p. 529)

“Wacquant (2000) argues that the prison, alongside the ghetto, has become a system of forced confinement that marginalizes minority communities from mainstream economic life” (Western, 2002, p. 529)

“Several biases may result. If earnings are only observed for ex-inmates who get jobs in the formal economy, analysis may include just those with successful experiences of re-integration.” (Western, 2002, p. 530)

“Survey data are rarely used because few surveys include institutionalized respondents or ask about imprisonment” (Western, 2002, p. 530)

“To trace mobility in earnings, data on log hourly wages, yi, is analyzed for respondent i in year t for the period 1983-1998. The hourly wage rate is measured for the respondent's current or most recent main job” (Western, 2002, p. 532)

“I discard observations with zero wages and a few outliers greater than five times the median wage (see Bernhardt et al. 2001” (Western, 2002, p. 532)

“. Because illegal earnings are likely missed, the analysis provides information about the effects of incarceration on ex-inmates' legitimate economic opportunities.” (Western, 2002, p. 532)

“regression models are built around three main predictors. First, the log of respondent's age, Ai,, captures the nonlinear age-earnings profile.” (Western, 2002, p. 532)

“Second, a dummy variable, Pi,, records whether the respondent previously served time in prison or jail” (Western, 2002, p. 532)

“Third, another dummy variable, Ci,, measures current incarceration status” (Western, 2002, p. 532)

“In this analysis, I adopt three different strategies to control for the nonrandom selection of men into prison and jail. First, several sources of selectivity are explicitly controlled” (Western, 2002, p. 533)

“Second, a more general model of respondent characteristics introduces fixed effects to capture the influence of time-invariant, observed and unobserved characteristics” (Western, 2002, p. 533)

“Finally, the selectivity of inmates is also examined by restricting the comparison group against which the incarceration effect is evaluated” (Western, 2002, p. 534)

“Nonrandom sample attrition can bias the analysis of panel data covering a long time period.” (Western, 2002, p. 536)

“Analysis of the NLSY provided mixed support for these claims. There is strong evidence that incarceration reduces the wages of ex-inmates by 10 to 20 percent.” (Western, 2002, p. 541)

“incarceration was also found to reduce the rate of wage growth by about 30 percent.” (Western, 2002, p. 541)

“Although the effects of incarceration on wage inequality were relatively small, the true effect in the population may be larger.” (Western, 2002, p. 541)

“More generally, the penal system has never occupied a central place in the study of American inequality and has been relegated instead to a specialty interest among criminologists” (Western, 2002, p. 542)