Delivered by Ingenta toUniversity of Pennsylvania Library (cid 80001333),University of Pennsylvania (cid 82007452)Date: 2005..02..10..23..22..
Has the growth of the American penal sys-tem over the past thirty years transformedthe path to adulthood followed by disadvan-taged minority men? Certainly the prison boomaffected many young black men. The U.S. penalpopulation increased six fold between 1972 and2000, leaving 1.3 million men in state and fed-eral prisons by the end of the century. By 2002,around 12 percent of black men in their twen-ties were in prison or jail (Harrison and Karberg2003). High incarceration rates led researchersto claim that prison time had become a normal
part of the early adulthood for black men inpoor urban neighborhoods (Freeman 1996;Irwin and Austin 1997). In this period of massimprisonment, it was argued, official criminal-ity attached not just to individual offenders, butto whole social groups defined by their race,age, and class (Garland 2001a:2).Claims for the new ubiquity of imprison-ment acquire added importance given recentresearch on the effects of incarceration. Thepersistent disadvantage of low-educationAfrican Americans is, however, usually linkednot to the penal system but to large-scale socialforces like urban deindustrialization, residentialsegregation, or wealth inequality (Wilson 1987;Massey and Denton 1993; Oliver and Shapiro1997). However, evidence shows incarcerationis closely associated with low wages, unem-ployment, family instability, recidivism, andrestrictions on political and social rights(Western, Kling and Weiman 2000; Hagan andDinovitzer 1999; Sampson and Laub 1993;Uggen and Manza 2002; Hirsch et al. 2002). Ifindeed imprisonment became commonplaceamong young disadvantaged and minority menthrough the 1980s and 1990s, a variety of othersocial inequalities may have deepened as aresult.Although deepening inequality in incarcera-tion and the pervasive imprisonment of
Mass Imprisonment and the Life Course: Race and Class Inequality in U.S. Incarceration
Becky Pettit Bruce WesternUniversity of Washington Princeton University
Although growth in the U.S. prison population over the past twenty-five years has beenwidely discussed, few studies examine changes in inequality in imprisonment. We studypenal inequality by estimating lifetime risks of imprisonment for black and white men atdifferent levels of education. Combining administrative, survey, and census data, weestimate that among men born between 1965 and 1969, 3 percent of whites and 20percent of blacks had served time in prison by their early thirties. The risks ofincarceration are highly stratified by education. Among black men born during thisperiod, 30 percent of those without college education and nearly 60 percent of highschool dropouts went to prison by 1999. The novel pervasiveness of imprisonmentindicates the emergence of incarceration as a new stage in the life course of young low-skill black men.
AMERICAN SOCIOLOGICAL REVIEW, 2004, VOL. 69 (April:151Ð169)
#1471-ASR 69:2 filename:69201-pettit
Direct all correspondence to Becky Pettit,Department of Sociology, University of Washington,202 Savery Hall, Box 353340, Seattle, WA 98195-3350 (bpettit@u.washington.edu) or Bruce Western,Department of Sociology, Princeton University,Princeton NJ 08544 (western@princeton.edu). Draftsof this paper were presented at the annual meetingsof the Population Association of America, 2001 andthe American Sociological Association, 2001. Thisresearch was supported by the Russell SageFoundation and grant SES-0004336 from theNational Science Foundation. We gratefully acknowl-edge participants in the Deviance Workshop at theUniversity of Washington, Angus Deaton, RobertLalonde, Steve Levitt, Ross MacMillan, CharlieHirschman, and ASR reviewers for helpful com-ments on this paper.

Delivered by Ingenta toUniversity of Pennsylvania Library (cid 80001333),University of Pennsylvania (cid 82007452)Date: 2005..02..10..23..22..
disadvantaged men is widely asserted, there arefew systematic empirical tests. To study how theprison boom may have reshaped the life pathsof young men, we estimate the prevalence ofimprisonment and its distribution among blackand white men, aged 15 to 34, between 1979 and1999. We also compare the prevalence of impris-onment to other life events—college graduationand military service—that are more common-ly thought to mark the path to adulthood.Many have studied variation in imprison-ment but our analysis departs from earlierresearch in two ways. First, the risk of incar-ceration is usually measured by an incarcerationrate—the overnight count of the penal popula-tion as a fraction of the total population (e.g.,Sutton 2000; Jacobs and Helms 1996). Muchlike college graduation or military service how-ever, having a prison record confers a persist-ent status that can significantly influence lifetrajectories. Our analysis estimates how thecumulative risk of incarceration grows as menage from their teenage years to their early thir-ties. To contrast the peak of the prison boom inthe late 1990s with the penal system of the late1970s, cumulative risks of imprisonment arecalculated for successive birth cohorts, born1945–49 to 1965–69. Second, although eco-nomic inequality in imprisonment may haveincreased, most empirical research just examinesracial disparity (e.g., Blumstein 1993; Mauer1999; Bridges, Crutchfield, and Pitchford 1994).To directly examine how the prison boom affect-ed low-skill black men, our analysis estimatesimprisonment risks at different levels of edu-cation. Evidence that imprisonment becamedisproportionately widespread among low-edu-cation black men strengthens the case that thepenal system has become an important newfeature of American race and class inequality.IMPRISONMENT AND INEQUALITYThe full extent of the prison boom can be seenin a long historical perspective. Between 1925and 1975, the prison incarceration rate hoveredaround 100 per 100,000 of the resident popu-lation. By 2001, the imprisonment rate, at 472per 100,000, approached 5 times its historicaverage. The prisoners reflected in these statis-tics account for two-thirds of the U.S. penalpopulation, the remainder being held in localjails. In 1997, about a third of state prisoners in
1997 had committed homicide, rape, or rob-bery, while property and drug offenders eachaccounted for one-fifth of all state inmates. Inthat same year, more than 60 percent of Federalprisoners were serving time for drug crimes(Maguire and Pastore 2001: 519). Nearly allprisoners serve a minimum of one year, withstate drug offenders in 1996 serving just over 2years on average, compared to over 11 years formurderers. In federal prison, average timeserved for drug offenders was 40 months in1996 (Blumstein and Beck 1999:36, 49). Theselengthy periods of confinement are distributedunequally across the population: More than 90percent of prisoners are men, incarceration ratesfor blacks are about eight times higher thanthose for whites, and prison inmates average lessthan 12 years of completed schooling.RACE AND CLASS INEQUALITYHigh incarceration rates among black and low-education men have been traced to similarsources. The slim economic opportunities andturbulent living conditions of young disadvan-taged and black men may lead them to crime.In addition, elevated rates of offending in poorand minority neighborhoods compound the stig-ma of social marginality and provoke the scruti-ny of criminal justice authorities.Research on carceral inequalities usuallyexamines racial disparity in state imprisonment.The leading studies of Blumstein (1982, 1993)find that arrest rates—particularly for seriousoffenses like homicide—explain a large shareof the black-white difference in incarceration.Because police arrests reflect crime in the pop-ulation and policing effort, arrest rates are animperfect measure of criminal involvement.More direct measurement of the race of crimi-nal offenders is claimed for surveys of crimevictims who report the race of their assailants.Victimization data similarly suggest that thedisproportionate involvement of blacks in crimeexplains most of the racial disparity in incar-ceration (Langan 1985). These results are but-tressed by research associating violent and othercrime in black neighborhoods with joblessness,family disruption, and neighborhood poverty(e.g., Crutchfield and Pitchford 1997; Messneret al. 2001; LaFree and Drass 1996; Morenoffet al. 2001; see the review of Sampson andLauritsen 1997). In short, most of the racial
152ÑÐAMERICAN SOCIOLOGICAL REVIEW#1471-ASR 69:2 filename:69201-pettit