Employer aversion to criminal records: An experimental study of mechanisms

• DOI: http://doi.org/10.1111/1745-9125.12228
• Date: 01 February 2020
• Author: Dallas Augustine,Naomi F. Sugie,Noah D. Zatz
• Published date: 01 February 2020

Received: 31 October 2018 Revised: 15 July 2019 Accepted: 20 July 2019

DOI: 10.1111/1745-9125.12228

ARTICLE

Employer aversion to criminal records:

An experimental study of mechanisms∗

Naomi F. Sugie1Noah D. Zatz2Dallas Augustine1

1Department of Criminology, Law and

Society, University of California, Irvine

2School of Law, University of California,

Los Angeles

Correspondence

NaomiF. Sugie, University of California,

Irvine, 3319 Social Ecology 2, Irvine,

CA92697.

Email:nsugie@uci.edu

Fundinginformation

UCLAAcademic Senate Council on Research;

UCLAInstitute for Research on Labor and

Employment;UC Irv ine AcademicSen-

ateCouncil on Research, Computing, and

Libraries; Universityof California Consortium

onSocial Science and Law

∗Theaut hors wouldlike to dedicate this article

inmemor y of DevahPager. The authors grate-

fullyacknowledge the helpful comments on

previousdrafts received from William Bielby,

DevahPager, and Roger Waldinger.This study

receivedfunding from the UCLA Institute for

Researchon Labor & Employment, UCLA Aca-

demicSenate Council on Research, University

ofCalifornia Consor tium on Social Science and

Law,and the UC Irvine Academic Senate Coun-

cilon Research, Computing, and Libraries.

Abstract

The mark of a criminal record is clearly harmful for

employment. The reasons for employer aversion, however,

are not well established even though legal, policy, and

scholarly responses rely on particular explanations. We

propose that explanations for aversion often fit under a

repetition risk framework in which employers use records

as neutral sources of information about prior illegal activity

and make decisions to minimize risk of similar future con-

duct. A second explanation is stigma, in which the records

themselves, independent of conduct, trigger stereotypes,

status loss, and discrimination. Using an experimental

employer survey, we find that employers evaluate appli-

cants with records more negatively than they do applicants

with similar behavior signaled through non-criminal-

justice sources (e.g., social media); this effect remains

after accounting for predictions about future conduct. It is

also most apparent among higher status jobs rather than

among manual labor jobs, and it persists after adjusting for

firm-level and legal constraints. We conclude that aversion

reflects not only repetition risk but also the stigma of crim-

inal justice contact. Insofar as criminal record screening

is not exclusively a form of rational risk management, this

finding may lead to altered assessments of the benefits of

screening relative to the costs of perpetuating inequality

produced by the criminal justice system.

KEYWORDS

criminal records, employment, stigma

Criminology. 2020;58:5–34. wileyonlinelibrary.com/journal/crim © 2019 American Society of Criminology 5

6SUGIE ET AL.

The mark of a criminal record has harmful consequences for hiring, and these penalties follow indi-

viduals long after formal criminal justice sanctions end (e.g., Holzer, Raphael, & Stoll, 2007; Pager,

2003; Pager, Western, & Bonikowski, 2009; Uggen, Vuolo, Lageson, Ruhland, & Whitham, 2014;

Western, 2002). These well-known implications—particularly for less-skilled, low-wage jobs—are

profoundly important given the prevalence of records in the United States and their disproportionate

concentration among racial/ethnic minorities. As of 2014, states held more than 100 million arrest and

conviction records (Bureau of Justice Statistics, 2015), which corresponds to approximately25 percent

of the U.S. adult population (Jacobs, 2015a), and it is estimated that 13 percent of adult males (and 33

percent of African American males) have felony conviction records (Shannon et al., 2017). Changes

in information technology also have made it easier for employers to conduct background checks. The

results of a recent survey indicated that 73 percent of human resource professionals in the Society for

Human Resource Management conduct background checksfor new hires (Society for Human Resource

Management, 2018).

Employer aversion to records—combined with the concentration of criminal justice contact among

low-income, less-skilled, racial/ethnic minorities—exacerbates racial/ethnic and class disparities in

unemployment and related domains (e.g., housing, addiction, and reoffending; Wakefield & Uggen,

2010). In response, social justice advocates and legislators have promoted policies such as “Ban the

Box,” in which record screening is restricted during initial hiring stages, as well as stronger “Fair

Chance” laws, in which restrictions are placed on how employersultimately use records. Currently, 35

states (and more than 150 localities) have enacted some form of Ban the Box (Avery, 2019). Concern

about the disparate impact of criminal record screening on African American and Hispanic men also

prompted the U.S. Equal Employment Opportunity Commission (EEOC) to issue guidance that federal

antidiscrimination law limits consideration of records in ways similar to Fair Chance policies (2012).

These legal and policy responses are premised on the notion that unregulated criminal record screen-

ing can expose people with records to unfair treatment and racial discrimination. Emerging scholar-

ship, however, also indicates that restricting such screening might unintentionally exacerbate overall

inequality. These research findings show evidence of statistical discrimination, or the use of easy-to-

discern characteristics such as race/ethnicity, age, and gender, to make assumptions about criminal

behavior in the absence of record information (e.g., Agan & Starr, 2016; Doleac & Hansen, 2016;

Vuolo, Lageson, & Uggen, 2017). Some criticisms of screening restrictions, such as Ban the Box, are

premised on this possibility that employers will engage in illegal discrimination, to the detriment of

racial/ethnic minorities without records, if employers cannot access relevant record information (Agan

& Starr, 2016; Doleac, 2016; Doleac & Hansen, 2016; Jacobs, 2015b; Strahilevitz, 2008).

A central question in these emergent legal, policy,and scholarly debates is why decision makers are

averse to records and whether these reasons are relevant to evaluation. Many common explanations

are focused on the risk of future reoffending, and we put forward the term repetition risk to describe

these explanations. According to this logic, instrumentally rational decision makers seek information

about the risk that applicants will, after hiring, engage in harmful conduct, including criminal conduct

such as violence, drug use, or theft; they use criminal records as a source of relevant information

about applicants’ past wrongful conduct on the theory that past conduct is predictive of similar future

conduct. That is, they are averse to the risk that the conduct indicated by the record will be repeated.

In a second prominent explanation, aversion is treated as manifesting and reproducing stigma that

comes with being marked by the criminal justice system as morally suspect and socially other. We

draw on Link and Phelan’s (2001) articulation of stigma as labeling, stereotyping, separation, status

loss, and discrimination to describe how records are used to mark individuals deserving of exclusion;

the criminal record itself, independent of behavior, therefore, influences aversion.