1. Inicio
  2. node
  3. Statement of Devah Pager, Professor of Sociology at Princeton University

Statement of Devah Pager, Professor of Sociology at Princeton University

Meeting of November 20, 2008 – Employment Discrimination Faced by Individuals with Arrest and Conviction Records

Madam Chair and Commissioners,

I am pleased to have the opportunity to participate in this meeting on employment discrimination issues faced by individuals with arrest and conviction records. Today, I will be presenting the results of a large-scale field experiment I conducted with Bruce Western in New York City investigating the effects of race and a prison record on employment. In this study, teams of black and white men were matched and sent to apply for hundreds of low-wage jobs throughout the city, presenting equivalent resumés and differing only in their race and criminal background. The results of this study demonstrate a large negative effect of a criminal record on employment outcomes, and one which appears substantially larger for African Americans. The sequence of interactions preceding hiring decisions suggests that black applicants are less often invited to interview, thereby providing fewer opportunities to establish rapport with the employer. Further, employers' general reluctance to discuss the criminal record of an applicant appears especially harmful for black ex-offenders. Overall, these results point to the importance of rapport-building for finding work, something that the stigmatizing characteristics of minority and criminal status make more difficult to achieve.

Thank you for your consideration of this research. I will look forward to your questions and comments.


Devah Pager
Associate Professor
Department of Sociology
Princeton University

Sequencing Disadvantage:
The Effects of Race and Criminal Background for Low Wage Job Seekers1

Roughly 700,000 inmates are released from prison each year, a fivefold increase from the late 1970s (Bureau of Justice Statistics 2004). Consisting mostly of young men with less than a college education, about two-thirds of ex-prisoners remain out of work a year after prison release, and 60 percent are re-arrested within three years (Bureau of Justice Statistics 2002; Petersilia, 2003; Travis 2005). Those that can find steady work are less likely to return to prison, and are better-equipped to assume the mainstream social roles of spouse and parent (Lopoo and Western 2005; Sampson and Laub 1993; Uggen 2000). Unfortunately, the goal of stable employment remains elusive for a large fraction of ex-offenders. The challenges of reentering society from prison are compounded for many by the racial stigma produced by prejudice and discrimination. Black men are about six times more likely than whites to be sent to prison and are likewise overrepresented among released prisoners (Bureau of Justice Statistics 2004, 2007). Some evidence suggests that blacks may also pay a higher penalty for having a criminal record relative to otherwise similar whites (Pager 2007). Given these patterns, understanding the nature of criminal and racial stigma—and the combination of the two—represents an important goal for research and policy.

We study the effects of race and a prison record on employment with a large-scale field experiment conducted in New York City. In this study, teams of black and white men were matched and sent to apply for hundreds of low-wage jobs throughout the city, presenting equivalent resumés and differing only in their race and criminal background. These results build upon our earlier work (Pager 2003), pointing to a robust interaction between race and criminal background. Further, this research allows us to look with more detail into the interpersonal contact between job seekers and employers for some insight into the process by which race and criminal background translate into labor market disadvantages.

We find a significant negative effect of a criminal record on employment outcomes, and one which appears substantially larger for African Americans. The sequence of interactions preceding hiring decisions suggests that black applicants are less often invited to interview, thereby providing fewer opportunities to establish rapport with the employer. Further, employers' general reluctance to discuss the criminal record of an applicant appears especially harmful for black ex-offenders. Overall, these results point to the importance of rapport-building for finding work, something that the stigmatizing characteristics of minority and criminal status make more difficult to achieve.

Stigma, Rapport, and the Job Matching Process

Little is known about the process by which employers select workers. Economic models of employment often assume that the productivity of prospective workers can be readily assessed, but in reality employers often face acute information shortages in evaluating new hires. Particularly in low wage job markets, where few concrete skills or experience are required, employers typically rely on limited information provided on a resumé or gathered during a cursory interview. Indeed, many employers claim to base hiring decisions on a "gut feeling" about candidates (Moss and Tilly 2001: 209), the source of which remains largely unknown.

Where employers are often looking for applicants with whom they feel an intuitive rapport, applicants with stigmatizing characteristics (such as minority status or a criminal background) may face special barriers to establishing such a rapport, even if possessing otherwise highly appealing characteristics. Indeed, a wealth of social psychological evidence indicates that negative stereotypes compromise interactions and undermine the ability of interaction partners to demonstrate traits that are inconsistent with stereotyped expectations. Experimental evidence suggests that stereotypes lead us to ask fewer questions of stereotyped targets (Trope and Thomson 1997) and to selectively notice and retain information consistent with the stereotype while ignoring information that is inconsistent with initial expectations (Fiske and Neuberg 1990). Though the effects of stereotypes have been shown to weaken as personalizing information becomes available (Allport 1954; Anderson 1999; Fiske and Neuberg 1990), perceivers are less likely to seek out or retain such information when confronted with members of salient social groups.

It is easy to imagine how this process might play out in employment settings. In cases where employers are confronted with stereotyped applicants, they may be more likely to make negative attributions about the individual without probing deeper into the specific characteristics of the applicant in question. Employers may be less likely to grant an interview to such applicants, and during the course of an interview may ask fewer questions or provide less opportunity for the applicant to present his profile in the best light (e.g., Word, Zanna, and Cooper 1974). While in some cases these dynamics may result from overt prejudice, these patterns of behavior can also be the product of simple discomfort or more subtle unconscious biases (Crocker, Major, and Steele 1998).

Most of the research on stereotyping and social interaction focuses on racial differences. In contemporary low-wage urban labor markets, a criminal conviction represents another source of disadvantage that may contribute to the differential treatment of young low-skill men. In fact, one might expect the effect of a criminal conviction to be more disqualifying for job applicants than racial stigma because of its direct association with negative behaviors—like dishonesty, violence, or unreliability—that suggest poor performance on the job. On the other hand, criminal history is a legitimate topic of discussion in a job interview, with job applications commonly asking about criminal backgrounds and employers often discussing criminal convictions with job seekers. These opportunities for candid discussion may provide chances to defuse the effect of a criminal background, a strategy less available in the case of racial stigma where prevailing norms discourage open conversations about race. Further, relatively little is known about how various stigmatizing characteristics may interact to produce new forms of labor market disadvantage. How do employers' assumptions or concerns about black applicants overlap with or intensify their concerns about ex-offenders? In what ways do the barriers facing one applicant type (e.g., a black applicant) contribute to the disadvantages experienced by those with additional stigmatizing characteristics (e.g., a black ex-offender)?

These ideas were previously examined in Pager's (2003) audit study of entry-level jobs in Milwaukee. Using an audit methodology, replicated in this paper, two-person teams of trained testers were assigned resumés with equivalent schooling and work histories. Within each team, one tester was randomly assigned a criminal record. The applicant pairs applied for entry-level jobs, measuring the extent to which race and a criminal background represented barriers to employment. Milwaukee employers strongly disfavored job seekers with criminal records, and the penalty of the criminal record was especially large for blacks. These results suggested that minority status compounds the stigma of a criminal record, though the mechanism through which this stigma is exerted remained unobserved.

We replicate the design of the Milwaukee study in New York City, looking beyond the general patterns of employment to investigate the sequence of events that precede an ultimate hiring decision. In particular, we examine the patterns of interaction (quantity and kind) experienced by black and white job seekers and their relationship to ultimate hiring outcomes. By studying the hiring process through this lens, we can better understand how rapport-building is facilitated or compromised as a function of an applicant's race or criminal background.

Data and Methods

The New York City Hiring Discrimination Study sent matched teams of testers to apply for 250 real entry-level jobs throughout New York City over nine months in 2004. The testers were well-spoken, clean-shaven young men, aged 22 to 26. Most were college-educated, between 5 feet 10 inches and 6 feet in height, recruited in and around New York City. They were matched on the basis of their verbal skills, interactional styles (level of eye-contact, demeanor, and verbosity), and physical attractiveness. Testers were assigned fictitious resumés indicating identical educational attainment, and comparable quality of high school, work experience, and neighborhood of residence. Resumés were prepared in different fonts and formats and randomly varied across testers, with each resumé used by testers from each race group. Testers presented themselves as high school graduates with steady work experience in entry-level jobs. Finally, the testers passed through a common training program to ensure uniform behavior in job interviews. While in the field, the testers dressed similarly and communicated with teammates by cell phone to anticipate unusual interview situations.

To study employers' responses to applicants with criminal records, we fielded two teams of testers. The first team paired two white applicants, one presenting a criminal record and the other a clean record. The second team paired two similar black applicants. None of the testers had real criminal backgrounds, but presented fictitious criminal records to employers. Testers rotated which member of the pair presented criminal background information, allowing for control of within-pair differences that might affect hiring outcomes. The criminal record was typically disclosed in answer to the standard question on job applications: "Have you ever been convicted of a crime? If yes, please explain." When asked, testers were instructed to reveal that they were recently released from prison after serving 18 months for a drug felony (possession with intent to distribute cocaine). Following Pager (2003), the tester's criminal record was also signaled on his resumé by listing work experience at a state prison, and by listing a parole officer as a reference.2

For both teams, employers were sampled from job listings for entry-level positions, defined as jobs requiring no previous experience and no education greater than high school. Job titles included restaurant jobs, retail sales, warehouse work, couriers, telemarketers, customer service positions, clerical workers, stockers, movers, delivery drivers, and a wide range of other low wage positions. Job listings were randomly drawn each week from the classified sections of The New York Times, The Daily News, The New York Post, The Village Voice, and Craigslist. The broad range of job listings allows for comprehensive coverage of the entry-level labor market in New York. From the available population of job listings, we took a simple random sample of advertisements each week. Testers in each team applied to each job within a 24-hour period, randomly varying the order of the applicants. Our dependent variable recorded positive responses in which a tester was either offered a job or called back for a second interview. Callbacks were recorded by voicemail boxes set up for each tester. For more information about the research design and methods, see Pager, Western, and Bonikowski (2008).


Two key findings emerge from the audit results. First, as in earlier research, a criminal record has a significant negative impact on hiring outcomes, even for applicants with otherwise appealing characteristics. Across teams, a criminal record reduces the likelihood of a callback or job offer by nearly 50 percent (28% vs. 15%). Second, the negative effect of a criminal conviction is substantially larger for blacks than for whites. As shown in Figure 1, the magnitude of the criminal record penalty suffered by black applicants (60%) is roughly double the size of the penalty for whites with a record (30%). This interaction between race and criminal record is large and statistically significant indicating that the penalty of a criminal record is more disabling for black job seekers than whites.

N = 250 employers

The intensification of the criminal record effect among blacks is consistent with earlier audit research (Pager 2007), and points to special barriers facing blacks in the transition from prison to work. Employers, already reluctant to hire blacks, appear particularly wary of blacks with known criminal histories. In the remainder of this paper, we examine the sequence of interactions that lead to this ultimate pattern of results. As job applicants pass from the point of application to an interview, and from an interview to an offer, we witness some of the underlying dynamics that may shape employers' decision-making and result in the systematic disadvantage of blacks with criminal records.

The Importance of Personal Contact

One of the ways job applicants can build rapport with employers is through the interview process. Though typically brief for low wage jobs, interviews provide opportunities for applicants to demonstrate communication skills and commitment to work. For employers concerned about soft skills not reflected on a resumé, even a brief interaction can provide important information about the capacity of an applicant. Especially in the case of stigmatized applicants, personal contact may serve an important means of counteracting employers' initial stereotypes. As employers learn more about the person behind the category (e.g., ex-offender, black man), their comfort level with the applicant in question is likely to increase.

The evidence from our audit study indeed confirms that personal contact with an employer has a substantial impact on hiring outcomes. Restricting our sample to cases in which both team partners had the same level of contact with an employer, we find that testers who interact with employers are between 4 and 6 times more likely to receive a callback or job offer than those who do not; and personal contact reduces the effect of a criminal record by roughly 15 percent (see Appendix A).3 Personal contact thus seems to play an important role in mediating the effects of criminal stigma in the hiring process. At the same time, this pathway to rapport-building may not be equally available to all applicant types. Though all testers in the study were instructed to request to speak to the person in charge of hiring and to proceed as far as they could in the interview process, some met with more success than others. In particular, race has a significant effect on the likelihood of personal contact between applicant and employer, with blacks roughly a third less likely to have the opportunity to interact with employers (p<.001). Employers appear to be screening on the basis of race in deciding who to allow to proceed from application to interview (see Figure 2).

N = 250 employers

By contrast, the effect of a criminal record has no discernable impact on the likelihood of interaction. Given that a criminal record is typically unobserved until an employer has spoken with the candidate and/or reviewed his materials, it is not surprising that this is typically not the basis of an employer's decision to interview.

The barriers that emerge in this very early stage of hiring are likely consequential for the disparities observed. With fewer opportunities for face to face contact with employers, black applicants are limited in their ability to demonstrate specific skills and attributes. Particularly in the case of black ex-offenders, for whom employers' concerns are likely particularly strong, limits on interaction reduce opportunities to contextualize a conviction or to demonstrate evidence of successful rehabilitation.

Quality of the Interaction

Being granted an interview, or even a conversation, with the person in charge of hiring provides an important opportunity for establishing rapport. But the nature of that rapport also hinges upon what takes place during the interaction. Particularly for applicants with criminal records, the interview provides a key opportunity to assuage employers' concerns. In this respect, our testers had mixed experiences. The testers had a set script that they were instructed to convey to employers about their prior drug conviction and their commitment to rehabilitation. In some cases employers were extremely receptive to discussing these issues, while in others employers seemed uncomfortable or unwilling to broach the subject. In one interaction, for example, an employer inquired about the most recent job listed on Kevin's resumé, which was at a correctional facility. Kevin reports, "I thought she was asking me what I did to get in there. I said, 'It was for drug possession.' She said, 'No, not that. That's none of my business.'" The employer then quickly moved on to discuss Kevin's previous work experience. It is unclear from this interaction what lingering doubts the employer may have had about Kevin's criminal background, but Kevin does not have the chance to explain further.

In another case, Anthony, an African American tester, reports, "As she looked over [my resumé] I could barely hear her say, 'Oh, I see.' I don't know what it was in response to, but it was pretty quick so I would guess it was my conviction… She then just looked up at me and said, 'I'll give you a call.' It seemed like she ended it a bit abruptly." Some employers seemed uncertain about what was legally or socially appropriate to ask about prior convictions, and others seemed simply uncomfortable with the topic or considered it outside of the realm of employment-related concerns. In these cases, it is difficult to interpret the employers' response to the criminal background, and the applicant typically has less opportunity to account for the stigmatizing record or address employers' underlying concerns.

In other cases, employers' concerns about the criminal record are more transparent. Worries about legal liability, for example, came up in this interview with Chad, an African American tester:

When I finished the application I was interviewed by…a large white man with a thick mustache. He shook my hand, invited me to have a seat, and began to look over my resumé. He said, "First, I need you to explain this...correctional facility and parole officer reference." I told him that I was convicted of a drug charge -- possession with intent to sell. "I can't hire you," he said. He went on to explain that a lot of things can happen and the liability is too great. He said, "Let's say you got into a fight with a guy and you were in the right. The police come and run your background, yes, now you're in the wrong, even though you may have been right. It wouldn't be good for you and it wouldn't be good for us. I couldn't hire you."

Sometimes, employers' negative reactions are less explicit, but their concern over the prospect of hiring an ex-offender is clear nevertheless. For example, Kevin, a white tester, reports his experience at a specialty foods store:

"I noticed a sign on the door which read, 'Help wanted, part time, some experience'.... A few minutes later a man came out.... He told me that he had a great Part Time position [and] there could be some full time positions opening [in] a while. He pointed at my response for reasons for leaving on the application right next to [the correctional facility] and said, 'Why did you write parole?' I said that I was currently on parole. [He] then looked me in the eye and said, 'Did you commit a crime?' I said yes. He then looked down at the sheet and said that he really wasn't hiring right now…."

Kevin's conviction record seemed to catch the employer by surprise. Within seconds, the many signals pointing to a job opportunity (help wanted sign, "great part time positions," etc.) disappeared, as the employer decided he was no longer hiring, or at least not hiring Kevin.

Despite these unpleasant experiences, not all employers reacted negatively upon noticing a criminal record. In fact, on a number of occasions, testers encountered extremely sympathetic employers. For example, Kevin records his experience in applying for a job at a car dealership:

"He saw the correctional facility [on my resumé] and said, 'We're an Equal Opportunity Employer. We don't care about this. About 75% of the people in this business have a record anyway….'" Kevin describes the end of the interview: "He said he was going to call me. Then he said, 'I know what you are thinking. This asshole is never going to call me. I will call you. Not because you have good sales experience but just because you need a chance.'"

This employer appears extremely sympathetic to the plight of an ex-offender looking for work. In fact, the employer seems willing to privilege the desire to give Kevin a second chance over his need for workers with relevant experience. Indeed, this employer called Kevin about the job two days later.

To be sure, many of the sympathetic responses received by testers in the criminal record condition simply reflect the preexisting attitudes of employers, independent of the interaction. Employers who feel sympathetic toward ex-offenders are likely to express such sympathies in conversations with ex-offender applicants. But above and beyond employers' predispositions, we observe some evidence that the interaction itself can work to clarify and shape the employers' interpretation of the criminal record. For employers who have ambivalent feelings about hiring ex-offenders, or who have anxieties about particular kinds of ex-offenders, interaction with the candidate allows the employer to interrogate these concerns directly. In one case, for example, Keith, a white tester, has an extended conversation with the manager at a furniture rental store:

After finishing the application I brought it back to [the employer], along with my resumé. He invited me to take off my backpack and have a seat. He began looking over the res./app. and his first question was, 'Were you selling or using?' I told him, 'Using. It was a minor thing. A stupid mistake and I'm now clean.' He was sympathetic saying, 'I gotcha. It was a question, not a criticism.' [The employer then asked him a few questions about his driver's license and driving history.] He invited me to sit with him out at the door while he smoked. When we got there he informally sat me down and lit up. He turned to me and said, 'So why should I hire you instead of one of the 47 other guys I got coming to me?' I told him, 'I'll work hard for you, bust my ass. It's a condition of my parole that I work.' He said, 'But do you want to work?' I answered, 'Yeah, I'm looking to get back into society...' He interrupted, 'You want to get your shit together.' 'Yeah,' I said. …The conversation ended with him saying, 'My inclination is to say yes,' (regarding hiring me). He added, 'My bosses, the owners, are a little more close-minded than me.... Look, you paid, you don't have to pay for it the rest of your fucking life. People make mistakes. I'll get you my card.'

This employer expresses some open-mindedness about Keith's criminal background from the start ("It was a question, not a criticism"), but he also wants evidence that taking a chance on Keith would be warranted amidst the large pool of other candidates. The conversation appears to provide important reassurances to the employer, who ends the interview with an encouraging note.

In another case, Anthony, an African American tester, discusses his criminal background with the manager of a healthcare company:

"[The employer] said, 'I'm sure people must take double-takes [when they see the correctional facility].' I replied, 'Yeah, that does happen sometimes.' He told me that he knows the law, says I have to provide that information to a possible employer, but not to worry because he has had other employees who have 'fucked up in the past.' He said, 'I feel safer knowing you're telling me up front than me having to wonder if you're gonna come here and tear shit up. Let's face it, interviews are bullshit. You can't know someone from a 5-minute interview. So let's cut to the chase.'

- 'How long were you in?' (18 months)

- 'When'd you get out?' (A few weeks ago)

- 'Ok, so you're fresh out and trying to get back on track?' (Yes, I am)

- 'What'd you do?' (I had a small amount of drugs on me)

- 'So you were guilty?' (Yes, I was young and made some mistakes but I learned from them and am completely drug-free)

- (jokingly) 'So you weren't innocent?' (No, it was my fault.)

- 'Don't worry, I find that those that messed up and want to set things right are better workers.'

In this interaction the employer does acknowledge certain concerns about hiring someone with a record, but seems to respond favorably to Anthony's honesty, and after learning more of the details of Anthony's background, offers him an encouraging response.

Thus, while a criminal record does have a significant negative impact on the employment prospects of job seekers, some employers are willing to look beyond the conviction, or to downplay its significance in the context of other information they acquire during the interview. In these cases, a kind of empathy seems to develop between employer and job seeker, with goodwill often translating into a substantial improvement in employment prospects. Of course the types of individualizing information employers look to are not always in the applicants' control. Race and ethnicity, in particular, appear to affect some employers' interpretations of the seriousness of the criminal background and the depth of empathy generated by the interaction. For example, Keith, a white tester, reports on his interaction with the manager of a restaurant supply company:

[The employer] sat me down and went over my application and resumé. He first saw [the correctional facility] and asked about my working there. I told him that I had been incarcerated. In a lower voice he said, "What did you do?" I told him about my being caught with cocaine, my time served, my current sobriety, which my parole officer could verify, the fact that it was a mistake and I had learned my lesson. At this point he said, "Zarriello...what is that, Italian?" I said, "Yes," and could immediately tell he was now on my side. He asked more questions like, "What happened exactly?" I told him I was in a car with some ex-friends that was pulled over and we were all searched and they found 6 grams in my possession. He asked, "You come from a nice Italian family? What do your parents think about this?" I told him they were disappointed but thought I had learned my lesson. He told me that "people make mistakes." He went on to say, "It would make me feel good to help a guy like you, more than just the rewards of doing my job, but good as a person if I can help someone."

This employer's emphasis on the value of second chances and the desire to help a young ex-offender get back on his feet are similar to sympathetic reactions we heard from other employers in interactions with both black and white testers. Here, though, we see the employer explicitly invoke Keith's ethnic background as the basis for solidarity, and as a key turning point in the employer's reaction to the criminal background information, as Keith moves from being viewed as an ex-offender to someone from a "nice Italian family."

In a similar case, Kevin applied for a job with a staffing agency and was asked a number of detailed questions about his criminal history. Toward the end of the interview Kevin reports,

"[The employer] said, 'Do you have Irish in you?' I said, 'On my mother's side I do.' He asked what else I have. I said, 'French.' He was delighted! He said he has Irish and French in him, too. He said we could be related because we are both from FrIreland (France & Ireland). I said yes and laughed with him." Kevin concluded his report by offering his impression of the interaction: "He really wanted to help me out and seemed to be going to great lengths to find me a job."

Once again, the ethnic solidarity expressed in this interaction appears to help establish a positive rapport between candidate and employer. Conversations with employers thus simultaneously offer the opportunity to present personalizing information about the applicant's work ethic and commitment to rehabilitation, but may also generate new bases for categorical distinctions.

Employers thus appear to offer a range of reactions to ex-offender applicants, varying in terms of employers' comfort level in discussing criminal backgrounds and their evaluative assessments of this information. To examine employers' responses more systematically, we coded tester interactions with employers according to the nature of their response to the criminal record information, based on narrative data provided by the testers. Focusing on testers with criminal records who had personal contact with employers (roughly 50% of all tests), we code responses as: ambiguous or no response, negative response, and sympathetic response.

Looking to the results in Figure 3, we see that overall employers are most likely to avoid talking about the conviction altogether. Between two-thirds and three-quarters of employers either avoided the subject of the criminal record altogether or gave little indication of how they viewed the information. By contrast, less than 10 percent of employers made explicitly negative comments. Though the hiring outcomes from the audit study indicate a large negative impact of a criminal record on employers' evaluations, we see little of this reflected in their explicit comments to job applicants. If expressing a clear valence, employers are more likely to offer sympathetic reactions, with roughly 35 percent of employers coded as sympathetic toward the ex-offender applicant.4 We see some evidence that blacks are more likely to receive a negative response (6% vs 3%) and less likely to receive a sympathetic response (30 vs 36%), though these differences are not statistically significant.

Note: Includes those testers in the criminal record condition who interacted with the employer (n=144)

Overall, these results point to a reluctance among employers to address the criminal record issue head on, or to reveal their reaction to the record to the applicant in question. Our final question, then, considers the extent to which these differential responses in interaction correspond to differences in hiring outcomes.

Relationship between Type of Interaction and Employment Outcomes

The nature of interaction between employer and applicant is significant primarily to the extent that it proves consequential for hiring. Matching interaction experiences with employment outcomes provides some leverage on the pathways through which ex-offenders find opportunity. Figure 4 presents the percent of applicants with criminal records who received a callback or job offer, by race of the applicant and type of employer response. Not surprisingly, employment outcomes are most favorable among those who received a sympathetic response from employers. These employers are not simply paying lip service to the value of second chances, but demonstrate an actual willingness to hire ex-offenders. Among those who receive sympathetic responses from employers, whites are more likely to receive an actual callback or job offer (42 vs 32%), though this difference is not statistically significant.

Note: Includes those testers in the criminal record condition who interacted with the employer (n=144)

Showing even more consistency between interactions and outcomes, employers who express negative reactions to applicants with criminal records in no cases made offers or callbacks to these applicants. The group with less consistent results includes those employers who offer no reaction or ambiguous reactions to the criminal background. Among these employers, we see a large racial difference in employment outcomes with white applicants roughly three times more likely to receive a callback or job offer relative to blacks who have similar encounters (29 vs 8%). Relative to those who receive a sympathetic response, the penalty associated with limited or no discussion about the criminal record is roughly 30 percent for whites; for blacks, this limited interaction appears far more consequential, resulting in 75 percent fewer callbacks or job offers relative to those who received a sympathetic reaction. Though we cannot directly interpret employers' underlying reactions, this evidence is consistent with the role of stereotypes inhibiting the acquisition and impact of personalizing information. If employers who are concerned about the record among black applicants choose to remain silent about the issue, the applicant then has little opportunity to anticipate or address the employer's concerns. Where for white ex-offenders this reduced communication does not appear overly consequential, black ex-offenders seem to face substantially lower employment prospects as a result.


The results of this study show a strong reluctance among employers to hire applicants with criminal records, especially when considering black ex-offenders. Despite the many appealing personal characteristics of our testers, employers often appear to base their decisions on the more salient markers of race and criminal background. What is perhaps more noteworthy in these data relates to the cases in which testers with criminal backgrounds are given a chance at employment. Employment prospects improve significantly for applicants who have a chance to interact with the hiring manager, and more so among those who elicit sympathetic responses in the course of those interactions. Surely, some of this variation is attributable to pre-existing characteristics and preferences of the employers, with little or no effect of personal contact. Employers who are eager to hire will be more likely to meet with applicants on the spot, and those who are sympathetic toward ex-offenders will be more likely to express such sentiments in the course of interaction. Still, we suspect that the interaction itself plays a non-trivial role in this hiring process. Employers have many reasons to be concerned about applicants with prior histories of incarceration. Concerns about theft, violence, and drug use are each relevant, not to mention the more mundane concerns over worker reliability and performance. Personal contact with an applicant cannot reveal all of these issues, but can help to provide some signals as to the disposition of the applicant, and can help the employer develop a "gut feeling" about whether or not this individual is likely to diverge from the stereotype of the ex-con.

Unfortunately, the ability to have such a hearing does not appear available to all applicants. Blacks are significantly less likely to be invited to interview by employers, offering them fewer opportunities to present indicators at odds with their stigmatized group membership. Further, though the distribution of reactions from employers is roughly similar among black and white applicants with criminal backgrounds, actual employment outcomes differ for those who have little opportunity to discuss their criminal record: Among whites, these limited interactions are not overly consequential, whereas for blacks, job opportunities appear substantially reduced.

These findings must be contextualized in light of the sampling design of the study, which focused exclusively on jobs obtained through formal classified listings. Given that many job seekers find employment through social networks and other informal channels, our analysis may understate opportunities for personal contact made possible through mediated contacts. However, evidence on social networks in employment suggests racialized consequences of these pathways as well, with blacks less likely to obtain quality leads to employment from their networks relative to similarly situated whites (cf. Royster 2003). Racial disparities in access to social networks have also been shown in the case of ex-offenders (Sullivan 1989). These informal methods of job search behavior, therefore, may result in greater evidence of racial disparities in employment following incarceration than what is reported here.

Overall these findings point to the importance of rapport-building in the employment process, particularly for applicants with stigmatizing characteristics. In light of these findings, policy intervention should aim to defuse stigma and provide employers with more information about their prospective workers. Initiatives that facilitate the matching of workers with employers in ways that help to overcome these initial barriers may have a substantial impact. Job referral services that act as labor market intermediaries who vouch for job applicants represent one important policy approach to bridging this divide. Certificates of rehabilitation and public education campaigns might also weaken the effects of stereotyping.

As incarceration rates have increased over the last few decades, official criminality compounds the stigma of race and deepens the economic disadvantage of young African American men. Instead of merely adding to the deficits of low-skill black men, a criminal record modifies the effect of racial discrimination, raising the bar to employment higher for blacks than similarly-situated whites. In this context, we can understand the growth of incarceration rates, and the racial disparities that characterize them, as producing a new form of institutional racism with wide-reaching economic effects.


Allport, Gordon. 1954. The nature of prejudice. New York: Doubleday Anchor Books.

Anderson, Elijah. 1999. Code of the streets: Decency, violence, and the moral life of the inner city. New York: W. W. Norton.

Bureau of Justice Statistics. 2002. Recidivism of prisoners released in 1994. By Patrick Langan and David Levin. Washington, DC: U.S. Department of Justice, June.

Bureau of Justice Statistics. 2004. Reentry trends in the U.S. Washington, DC: U.S. Department of Justice

Bureau of Justice Statistics. 2007. Prisoners in 2006. Washington DC: U.S. Department of Justice.

Crocker, Jennifer, Brenda Major, and Claude Steele. 1998. "Social stigma." In The handbook of social psychology, edited by D. Gilbert, S. Fiske, and G. Lindzey, 504-553. Boston: McGraw-Hill.

Fiske, Susan, and S. L. Neuberg. 1990. "A continuum model of impression formation: from category-based to individuating processes: Influence of information and motivation on attention and interpretation." In Advances in experimental psychology, edited by M. P. Zanna, 1-74. New York: Academic Press.

Holzer, Harry. 1996. What employers want: Job prospects for less educated workers. New York: Russell Sage Foundation.

Lopoo, Leonard M. and Bruce Western. 2005. "Incarceration and the formation and stability of marital unions." Journal of Marriage and Family 67:721–734.

Moss, Philip and Chris Tilly. 2001. Stories employers tell: Race, skill, and hiring in America. New York: Russell Sage Foundation.

Pager, Devah. 2003. "The mark of a criminal record." American Journal of Sociology 108:937-75.

Pager, Devah. 2007. Marked: Race, crime, and finding work in an era of mass incarceration. Chicago: University of Chicago Press.

Pager, Devah, Bruce Western, and Bart Bonikowski. 2008. "Race at work: Results from a field experiment of discrimination in low wage labor markets." Princeton University.

Petersilia, Joan. 2003. When prisoners come home: Parole and prisoner reentry. New York: Oxford University Press.

Royster, Deirdre. 2003. Race and the invisible hand: How white networks exclude black men from blue-collar jobs. California: University of California Press.

Sampson, Robert J., and John H. Laub. 1993. Crime in the making: Pathways and turning points through life. Cambridge, MA: Harvard University Press.

Sullivan, Mercer L. 1989. "Getting paid": Youth crime and work in the inner city. Ithaca, NY: Cornell University Press.

Travis, Jeremy. 2005. But they all come black: Facing the challenges of prisoner reentry. Washington, D.C.: The Urban Institute Press.

Trope, Yaacov, and Erik P. Thomson. 1997. Looking for truth in all the wrong places? Asymmetric search of individuating information about stereotyped group members. Journal of Personality and Social Psychology 73:229–241.

Uggen, Christopher. 2000. Work as a turning point in the life course of criminals: A duration model of age, employment, and recidivism. American Sociological Review 65:529–546.

Word, Carl O., Mark P. Zanna, and Joel Cooper. 1974.  "The nonverbal mediation of self-fulfilling prophecies in interracial interactions."  Journal of Experimental Social Psychology, 10:109-l20.

Appendix A. The Effect of Personal Contact on the Likelihood of a Callback/Job Offer and the Magnitude of Criminal Stigma

No interaction with employer
Both testers interacted with employer

no record (NR) criminal record (CR) ratio 1 (NR/CR)
no record (NR) criminal record (CR) ratio 2 (NR/CR)
% change in CR effect
White 8.82 5.88 1.50 (n = 34) 44.44 35.19 1.26 (n = 54) -0.16
Black 10.00 2.82 3.55 (n = 70) 41.86 13.95 3.00 (n = 43) -0.15

Note: First two columns of each section represent rates of positive response (callback/job offer) for testers by race and criminal status. Percent change in the effect of a criminal record (final column) is calculated as: (ratio 1 – ratio 2)/ratio


1 Testimony prepared for the EEOC hearing on employment of individuals with criminal records, November 20th, 2008. Research for this report was supported by grants from the National Science Foundation and the JEHT Foundation. We gratefully acknowledge the generous assistance of Commissioner Patricia Gatling and the New York Commission on Human Rights, and the contributions of Glenn Martin of the Fortune Society.

2 Results from Pager (2003) suggest that providing information about a criminal record to employers who do not request the information does little to affect hiring decisions. Those employers who request the information are those most likely to use it.

3 We restrict our sample here to cases in which both testers on a team received the same level of personal contact in order to better control for compositional differences between those employers more or less likely to interview candidates on the spot. By comparing the effect of a criminal record within teams where either both or neither tester interacts with the employer, we can better understand the ways in which personal contact may mediate the effects of stigma. This sample restriction has little effect on the substantive conclusions of the analysis.

4 This proportion corresponds closely with the 33 percent of urban employers surveyed by Holzer (1996:59) who report that they would "probably accept" or "definitely accept" an applicant with a criminal background.