Marked: Race, Crime, and Finding Work in an Era of Mass Incarceration

This chapter measures the effects of incarceration on ex-offenders seeking employment. Some researchers argue that the impact of incarceration is negligible and that employment disparities result from preexisting offender traits. Others posit a significant impact on employment and earnings. Pager seeks to resolve this debate through field experiments centering on real-life job searches. This chapter establishes the groundwork for one of the book’s major themes: The Intersection of Race and Criminality in the Milwaukee Labor Market.

An Emerging Field of Research

Scholars have only recently begun to study the links between prison and employment. The four main research methods used are: 1) employer surveys, 2) ethnographies of job seekers, 3) statistical analyses of work trends, 4) experimental studies of hiring behavior. These complementary methods offer insights into employment barriers for ex-offenders. Pager describes and assesses the first three strategies before turning to her primary approach: the experimental audit method.

Non-Experimental Methods for Studying the Labor Market Consequences of Incarceration

Employer attitude surveys are central to studies of employment and incarceration. One of the advantages of surveys is the ease with which employers can be recruited to participate. Large sample sizes allow researchers to identify patterns in employer attitudes. For instance, surveys suggest that attitudes vary by job type, with sales jobs placing greater restrictions of ex-offenders than laborer positions (43). Firm size, the race of the employer and applicant, and the type of crime also influence employer attitudes.

Although employers generally respond truthfully when surveyed, interpreting how their attitudes impact their hiring behavior is challenging. Further, surveys measure the relative preferences of individual employers rather than accurately estimating absolute employer demand. Because race is a sensitive subject, moreover, surveys can also fail to capture racial attitudes that impact hiring practices.

Ethnographies of job seekers are another way of studying the impact of incarceration on employment. This interview-based method provides detailed, firsthand accounts of the struggles ex-offenders face on the job market. Ex-offenders lack the support they need to transition back into homes and jobs. Rich in descriptive details, ethnographic studies not only allow researchers to understand the challenges facing individual ex-offenders, but also to identify commonalities and differences in their experiences. However, variations in subject histories, including drug use, the lack of social support, and unstable housing make it difficult to draw firm conclusions about reentry.

Statistical analyses of employment trajectories present a third approach to the study of reentry. This method relies on largescale surveys that track respondents over many years. Most such surveys reveal a decline in employment after periods of incarceration, which suggests that prison has negative consequences on ex-offenders. However, the dearth of largescale surveys limits the data available to researchers and prevents them from reaching firm conclusions. Unmeasured differences may also skew survey data. In addition, surveys cannot identify specific reasons for rejection, making it difficult to determine which causal explanations are at work (credentialing, legal barriers, institutional trauma, disrupted family ties, and racial discrimination).

Experimental Approaches to Studying Criminal Stigma

Pager used an experimental approach to incarceration and labor market consequences, a method rarely used in reentry studies. The main advantage of her approach is that it allowed her to identify the causal pathways through which imprisonment impacts employment. Pager employed the audit method, which combines traditional experimental approaches with field-based ones. Traditional experiments start by randomly assigning subjects to ‘treatment’ and ‘control’ conditions. All potential environmental influences are controlled. Audit studies relax controls over environmental influences to mimic real-world exchanges. Pager’s audits relied on real contexts: actual job searches. As in traditional experiments, her subjects were not told about the purpose of the experiments to ensure “natural” reactions to experiment conditions.

Scholars have used two types of experiments to study the impact of incarceration on labor markets: correspondence tests and in-person audits. Correspondence tests simulate correspondences between real employers and fake job applicants. They rely on fictitious résumés with equal positive credentials, but with experimental conditions (such as race). Reactions from employers are then measured through employers’ responses, such as letters to staged mailing addresses, or through callbacks to voice mailboxes. Researchers who used correspondence tests, notably, Richard Schwartz and Jerome Skolnick, found that having a criminal record negatively impacts employment opportunities (49).

The audit method simulates the hiring process more elaborately than correspondence tests. Pager describes her approach in detail in an appendix. In brief, she used pairs of testers, or auditors, who posed as applicants for real-life jobs. She paired testers by age, race, physical traits, and personal style. Testers received similar résumés and intensive training to guide their interactions with employers. Unlike correspondence tests, which rely entirely on written communication, the audit method allowed Pager to study entry-level jobs, most of which required in-person applications. Another advantage of in-person audits is that they gather quantitative and qualitative data. For example, the success or failure of an applicant is quantitative in nature, whereas the positive and negative treatment of applicants during interviews is qualitative. By randomly assigning criminal records to applicants, Pager was able to test the impact of criminal status on the decision-making process. Using pairs of Black and white testers allowed her to gauge the impact of race. Importantly, Pager’s method did not assess the discriminatory tendencies or patterns of specific employers because no employer received more than a single audit.

Critiques of the Audit Method

Although the audit method offers important insights into criminal stigma, the approach is not without detractors. Economist James Heckman, for example, argued that employers are influenced by a vast number of applicant characteristics that cannot be effectively controlled. In other words, he questions the ability of researchers to match auditing pairs effectively (53). Pager anticipates this critique and points out that her study focused on the impact of nonembodied characteristics on employment, namely criminal records. Other differences between tester pairs cancelled out over the course of the study. In addition, some testers served as both treatment and control, which ensured that other fixed traits were distributed randomly across conditions.

In contrast to nonembodied traits, race cannot be randomly assigned in audits. Pager addressed this issue by carefully matching testers, taking into account all relevant traits through extensive screening and careful selection. She chose motivated testers with excellent communication skills. Her approach to pairing was equally rigorous; audit partners were matched according to height, weight, age, style, and other factors, including credentials.

A second concern about the audit method is that matching distorts the hiring process. Closely matched partners force employers to rely on minor traits to make their make their decisions. For instance, employers who care little about race may nevertheless take the issue into account if two applicants are otherwise equal, placing greater significance on the trait than they would have under normal circumstances. Pager recognizes the validity of this critique. However, the employers in her study interviewed an average of eight applicants for every job. In cases where race was only a minor concern, both testers made it through the first round of interviews. Conversely, if a Black tester was eliminated early in the hiring process and a white tester was not, Pager concluded that race played a large role in the decision-making process.

A third critique of the audit method relates to experimenter effects, defined as “the possibility that the expectations or behaviors of testers can influence the audit results in nonrandom ways” (56). Experimenter effects present the most serious threat to audit experiments. However, Pager minimized the problem by carefully selecting, training, and supervising her testers. Role playing and video-taped practice interviews were central to this process, as were troubleshooting sessions and daily debriefings. Pager also checked for experimenter effects at the end of her study by comparing audit outcomes for testers who did and did not interact with employers. This allowed her to gauge the degree to which in-person interactions impacted results. Through rigorous care and attention to detail, Pager was able to design and implement a valid audit study to assess the impact of incarceration and race on labor market outcomes. The study results reinforce the theme of race and criminality affecting labor outcomes, not only in Milwaukee, but in comparable cities.