Peter Leasure & Tia Stevens Andersen∞
Upon completion of their sentences and when attempting to ‘reenter’ society, offenders face large barriers, often referred to as the ‘collateral consequences’ of conviction. One of the largest barriers, given the stigma of a criminal record, is finding employment. The problem primarily arises because of increases in the use of background checks by employers and the use of a criminal record to eliminate candidates. Such a practice is partly understandable for employers, as a recent conviction is one of the best predictors of future criminal activity. However, recent evidence suggests that an offender’s risk of reoffending decreases over time and can eventually come “close enough” to that of one who has never offended, even becoming lower than that risk for a random person within the general population. Building off this research, we present our research question, which asks whether there are employment outcome differences between hypothetical applicants with older and more recent criminal records. Results indicate that those possessing older criminal records still face barriers when seeking employment. Based on these findings, we present policy considerations.
III. The Current Study
In January 2000, Douglas El was hired to provide transportation for people with mental and physical disabilities. Within the first few weeks of El’s employment, the hiring company discovered that El had a forty-year-old conviction for second-degree murder; El was convicted when he was fifteen years old. El was terminated and the hiring company stated that the murder conviction was their sole reason. El filed suit, claiming that termination based solely on his conviction record violated the prohibition against employment discrimination under Title VII of the Civil Rights Act of 1964. The hiring company cited its policy of a lifetime ban on employment for those convicted of a violent crime, justifying such policy on the following grounds: (1) it is impossible to predict recidivism with a reasonable degree of accuracy; (2) someone with a conviction for a violent crime has a higher likelihood of subsequent violent offending than those without a criminal record, irrespective of the number of years passed since the criminal activity; and (3) the company’s hiring policy is the most accurate way to screen applicants who present an unacceptable risk.
During the case, the court considered the testimony of the hiring company’s expert witness, Alfred Blumstein. Blumstein stated that although rates of recidivism are relatively high in the first three years after an offense,
[It] is also the case that an individual’s propensity to commit a future crime decreases as that individual’s crime-free duration increases. [A]n individual with a prior . . . conviction who has been crime-free in the community for twenty years is less likely to commit a future crime than one who has been crime-free in the community for only ten years. But neither of these individuals can be judged to be less or equally likely to commit a future… act than comparable individuals who have no prior violent history. It is possible that those differences might be small, but making such predictions of comparable low-probability events is extremely difficult, and the criminological discipline provides no good basis for making such predictions with any assurance that they will be correct.
With this testimony, the court stated:
[W]e have little choice but to [side with the hiring company]. This is not to say that we are convinced that [the hiring company’s] expert reports are ironclad . . . Had El produced evidence rebutting [these] experts, this would be a different case. Had he, for example, hired an expert who testified that there is time at which a former criminal is no longer any more likely to recidivate than the average person, then [his case may have prevailed].
Researchers have sought to address the concerns echoed by the court in El, and recent evidence suggests that an offender’s risk of reoffending decreases over time and can eventually come “close enough” to the risk of offending for those in the general population. However, no study has examined whether the age of a criminal record impacts employment outcomes. The present study seeks to address this gap.
This study will begin with a review of literature on the effects of a criminal record on employment outcomes, how the proliferation in the use of criminal background checks exacerbates these poor employment outcomes, and the importance of an individual’s declining risk of re-offending over time. Such a review is important for two reasons: first, while employment is necessary for successful reentry into society, a criminal record and increased use of background checks makes this necessity difficult to secure for ex-offenders. Second, despite studies showing an individual’s declining risk of re-offending over time, no study has examined whether individuals with old criminal records fare better in employment outcomes. From this review of literature, we present our research question, which asks whether there are employment outcome differences for hypothetical applicants with older and more recent criminal records. Specific methodological approaches and results are then discussed. This study concludes with policy recommendations.
Research has consistently shown that employment is often correlated with an individual’s successful reentry and avoidance of criminal behavior. However, numerous studies have demonstrated that former offenders face tremendous challenges in finding adequate employment because of their criminal record. The earliest of these studies came from Schwartz and Skolnick, who used a correspondence design (submitting fictitious resumes to employers) and found that those possessing various types of criminal records fared worse in employment outcomes than those without a record.
Perhaps the most notable experimental designs come from Devah Pager, who utilized an experimental audit design whereby auditors were randomly assigned a “criminal record” and applied in-person for entry-level job postings advertised in Milwaukee, Wisconsin. Pager’s results showed that for white applicants, approximately half of the employers were unwilling to consider equally qualified applicants on the basis of their criminal record. The disparity was even more substantial for applicants of color. Pager’s findings have been successfully replicated in different locations and with additional racial groups.
More recently, a study conducted by individuals at the University of Minnesota and Purdue University examined the effect of low-level arrests on employment opportunities. While this study found that the effect of arrest history on employment opportunities was not as substantial as found in research examining the effect of prison records, callback rates were about four percentage points lower for testers with an arrest. Such a finding is important as it shows the presence of stigmatization even for very minor criminal records.
As criminal background checks have become easier to access by potential employers, they are increasingly relied upon by employers making hiring decisions, thus exacerbating the negative effect a criminal record has on securing employment post-release. For example, forty state criminal history repositories performed forty-two million background checks in 2008, a 172% increase from only 2006. This estimate does not include checks done for government agencies, civilian checks done by the Federal Bureau of Investigation, or checks done by private companies, which would increase the overall number of checks by tens of millions. Further, the Society for Human Resource Management found that nearly 90% of organizations surveyed reported conducting criminal background checks on at least some job candidates, and nearly 70% reported conducting criminal background checks on all job candidates. Such practices are an important development, given that between sixty-five million and 100 million Americans possess some form of a criminal record.
However, there are some limitations on the use of an individual’s criminal history in making employment decisions via Title VII of the Civil Rights Act of 1964, which is enforced by the Equal Employment Opportunity Commission (EEOC). The 8th Circuit stated that an employer violates Title VII when the potential employee with a criminal record demonstrates that: (1) the employer’s neutral policy or practice has the effect (disparate impact) of disproportionately screening out a protected group; and (2) the employer fails to demonstrate that the policy or practice is job related for the position in question and consistent with business necessity. In Green, three factors were created to determine whether an exclusion based upon a criminal record is job related for the position in question and consistent with business necessity. First, the nature and gravity of the offense or conduct must be considered. Second, the time that has passed since the offense or conduct and/or completion of the sentence must be considered. And third, the nature of the job held or sought must be considered in light of the previous two factors.
The EEOC has specified two circumstances in which employers will consistently meet the “job related and consistent with business necessity” requirement. The first circumstance arises when a criminal conduct screen is carried out by an employer pursuant to the 2010 Uniform Guidelines on Employee Selection Procedures. The second circumstance arises when the employer has developed a targeted criminal conduct screen that takes into consideration the Green factors, and also provides an opportunity for an individualized assessment to determine whether the policy as applied is job related and consistent with business necessity. This means that although an employer cannot generally have an outright policy of not hiring persons with a criminal record, the employer can effectively do so because of the large amount of discretion given to employers by the EEOC’s two circumstances noted above (following Green and the 2010 Uniform Guidelines). In summary, the combination of the increase in the number of Americans with a criminal history, the high prevalence of employer background checks, and lackluster limitations on consideration of a criminal record in hiring decisions results in the exclusion of many individuals from legitimate employment.
One reason for the expansive use of criminal background checks by employers is the well-documented link between recent and future offending. Several studies show that a very strong predictor of future criminal behavior is past criminal behavior. However, focusing on the link between recent and future offending only gives employers a half-painted picture. First, empirical research demonstrates the importance of securing employment for desistance. Second, research also consistently demonstrates that there is a steady decline in criminal activity after an individual’s peak in the late teens and young-adult period. Third, research documents that one’s recidivism risk declines steadily with time after one’s last offense. For example, one study following a group of 962 felons after twenty years showed that although half of the individuals were arrested within 2.2 years, nearly one-third remained arrest-free after the original sentence. Subsequent analysis of the same data showed that, among those who did not reoffend during the first ten years following their conviction, less than 4% were reconvicted within the subsequent decade. Fourth, building off work examining a declining risk of offending, recent studies have specifically focused on those who possess older criminal records and have sought to answer how old a criminal record needs to be before that offender’s risk of reoffending is equal to one who has never committed any crime. These studies address the concerns noted by the court in El and show that an offender’s risk of re-offending declines over time such that it approximates one in the general population or even individuals who have never committed a crime. For example, Kurlychek, Brame, and Bushway (2006) used the longitudinal data from the 1945 Philadelphia birth cohort study and found that differences between non-offenders and recent offenders “weaken quickly and dramatically over time such that the risk of new offenses among those who last offended six or seven years ago begins to approximate (but not match) the risk of new offenses among persons with no criminal record.” However, the risk of offending for ex-offenders never equaled the risk for the general population during the study’s period of observation. Kurlychek, Brame, and Bushway (2007) later conducted a very similar analysis with police contact data from the 1942 Racine birth cohort study. This study found stronger evidence of conversion (offenders with old criminal records equaling risk of offending when compared to non-offenders), likely because it followed participants until age thirty-two.
Continuing this line of research, Blumstein and Nakamura (2009) used data from the New York state criminal-history repository to estimate when offenders’ risk of future offending equals people of the same age in the general population and estimate when an offender’s declining hazard comes “sufficiently close” to the hazard of those who have never been arrested. The study found that an offender’s risk of future offending declines over time (at different rates, depending on age at offense and whether the crime was violent or property) and becomes similar to the general public’s risk of offending between three and nine years. They also found that an offender’s risk of re-offending becomes “sufficiently close” to the hazard rate of those who have never been arrested after about five to eight years, depending again on age at offense and whether the crime was violent or property. Soothill and Francis considered this question using data from England. As in the previous studies, they found that the hazards of offending decline for the groups with prior records and eventually converge within ten to fifteen years with the hazard of those of a comparison non-offending group. Although some groups, strictly speaking, never converge with the hazard of the non-offending group, Soothill and Francis argued that the hazards are essentially non-distinguishable by the ten-to-fifteen-year mark.
Finally, Bushway, Nieuwbeerta, and Blokland used long-term longitudinal data on a Dutch conviction cohort and examined the redemption question (measured by convictions) for youthful offenders (as in previous studies), older offenders, and offenders with multiple convictions. Their study found that it takes approximately ten years for offenders between the ages of twelve and twenty-six to resemble the non-conviction group in terms of re-offense risk; that older offenders begin to look like non-offenders after two to six years; and that offenders with four or more offenses either never resemble non-offenders or only begin to do so after a minimum of twenty-three years.
III. The Current Study
Because the stigma of a criminal record affects an individual’s chances of securing employment, researchers have sought to address the concerns echoed by the court in El and determine when ex-offenders’ risk of re-offending approximates that of the general population. The evidence suggests that an offender’s risk of reoffending does decrease over time and can eventually come “close enough” to one who has never offended and even become lower than a random person within the general population. However, no study has examined employer response to old criminal records. Because Pager and colleagues suggest hiring decisions account for a substantial, yet hidden, proportion of discrimination in the labor market, the current study examines whether there are disparities in the hiring process between applicants with no self-disclosed criminal record, a recent felony drug conviction, and a ten-year-old felony drug conviction. The above literature highlights the importance of the current study through two overall points: First, research has consistently demonstrated the importance of securing employment for successful reentry. Second, the redemption process dictates that ex-offenders who have served their sentence and whose risk of offending is “sufficiently close” to the general population should not face unjustifiable barriers to employment.
The data for this study were collected in the Columbus, Ohio metropolitan area. Ohio’s capital and most populous city, Columbus has an estimated population of more than 850,000. According to the Bureau of Labors Statistics data, Columbus has a moderately strong, primarily service-providing economy. At the time of data collection, unemployment rates were lower than the U.S. national average and ranged from a high of 4.2% in June to a low of 3.6% in August.
Rates of incarceration and reentry in Ohio have mirrored those observed at the national level. Although incarceration rates in Ohio are no longer consistently increasing, they remain extremely high relative to other states and earlier periods. In 2014, Ohio had the third highest correctional supervision rate in the United States, behind only Georgia and Idaho, with more than 325,000 persons under correctional supervision. For every 100,000 adult residents in Ohio, more than 3,600 were incarcerated in state or federal prison, in jail, or on probation or parole. Because almost all incarcerated individuals are eventually released, more than 20,000 individuals are now released from Ohio prisons each year.
To examine the relative influence of old felony records on employment outcomes, we adopted an experimental correspondence approach. Correspondence field experiments rely on fabricated matched resumes submitted to employers. Resumes are created with equal levels of education and experience, and criminal record status (or another characteristic) is conveyed through one or more cues. Discrimination is then examined by randomly assigning resumes cues that signal criminal record status and observing the effect on positive responses by employers. The primary advantage of correspondence experiments is that they require no actual job applicants. This is advantageous for both practical (i.e., logistical ease, trivial cost) and methodological reasons (i.e., greater control over treatment and control conditions, the reduction or elimination of experimenter bias). Although one historical limitation of the correspondence approach was that entry-level, low-wage jobs more often required in-person applications, this is less of a hindrance in the digital age as many employers now require online applications.
For these analyses, we created three sets of matched resumes with identical names (in this case, Matthew O’Brien), educational backgrounds, employment experience, and key skills. Because most state prisoners have no more than a high school diploma, we chose to list each fictitious applicant’s highest level of education as a high school diploma. We also chose to assign favorable and consistent work histories, including experience in manufacturing, sales/customer service, and food service. The only difference between resumes was the type of criminal record. Like Pager and her colleagues, we focused on the effect of a drug-related criminal record on employment opportunities. Sets of resumes were created with the following self-disclosed criminal histories: (1) a one-year-old felony drug conviction; (2) a ten-year-old felony drug conviction; and (3) for the control group, no self-disclosure of a criminal record. The resumes containing these experimental conditions were then randomly assigned to a random sample of potential employers.
Between May and August of 2015, we gathered all entry-level employment ads posted within the prior two weeks for the Columbus metropolitan area from CareerBuilder.com, Craigslist.com, and Indeed.com. Every week during that period, the first author created a population list of the entry-level employment ads. From that weekly population list, approximately thirty employers were randomly drawn for random assignment to one of the resume types. Advertisements that requested applicants to apply in person or explicitly prohibited applicants with criminal records were excluded from the sampling frame. These included jobs in the health care industry, those that work with children and the elderly, those requiring the handling of firearms (e.g., security guards), and those in the public sector. Two employers reported the job had been filled and were therefore excluded from analysis. This resulted in a final sample size of 303 employers. To measure employer response, the first author monitored an email and standard default voicemail account. Responses were recorded as positive when fictional applicants received an interview invitation or an offer of employment, and negative otherwise.
Table 1 presents the percentage of resume applications submitted to each type of job and positive employer response rates. Although customer service positions were the most common entry-level job advertised (30%), most listings advertised “blue-collar” positions, such as general labor (19%), manufacturing (10%) and warehouse shipping (10%). Advertisements for entry-level employment opportunities in service industries were slightly less common and included positions in sales (10%), restaurant or grocery (9%), and clerical work (9%). Overall, nearly one in five applications received an interview invitation or offer of employment. Employer responses were most positive for those applying for driving jobs (33% callback rate) and least positive for those applying for restaurant or grocery occupations (12% callback rate).
Table 1: Frequency of Resumes Submitted, by Treatment Group and Occupational Category (N = 303)
|Type of Job||No Disclosure of a Criminal Record
(n = 107, 35%)
(n = 102, 34%)
(n = 94, 31%)
|Total||Positive Employer Response %|
|Customer service||24||32||36||92 (30%)||17%|
|General labor||19||27||12||58 (19%)||24%|
Note: Positive responses refer to interview invitations or job offers.
Figure 1 presents the percentage of positive responses for equally qualified applicants with no self-disclosed criminal record, a one-year-old felony drug conviction, and a ten-year-old felony drug conviction. Given that identical resumes were submitted to entry-level job listings, the differences between positive employer responses can be attributed to the effect of criminal record. First, comparing applicants with a clean background to those with a recent drug conviction, the self-disclosure of a one-year-old felony drug conviction on job applications substantially reduces the likelihood of receiving an interview invitation or job offer from employers. Although nearly 30% of fictional applicants with clean backgrounds received a positive employer response, less than 10% of applicants with a recent felony conviction received a positive employer response. In other words, the proportion of applicants with a recent felony drug conviction who received an interview invitation or job offer was 66% lower than their equally-qualified counterparts with no self-disclosed criminal records. From these results, we infer that many employers were unwilling to consider qualified applicants largely on the basis of a criminal record.
Figure 1. Positive Employer Responses by Criminal Record Type (N = 303)
Note: Positive responses refer to interview invitations or job offers.
The focus of our analyses involves the potential effect of old criminal records on employment opportunities. In other words, given research that indicates risk of recidivism declines over time, how does the possession of a ten-year old felony drug conviction and no further contact with the criminal justice system affect employment opportunities? As Figure 1 indicates, nearly 20% of applicants with old criminal records received an interview invitation or job offer from prospective employers. Compared to fictional applicants with a one-year-old criminal record, the proportion of applicants with a ten-year-old felony drug conviction were nearly twice as likely to receive an interview invitation or job offer. Although the difference did not reach statistical significance, the proportion of applicants with old felony records who received an interview invitation or job offer was approximately 33% lower than their equally qualified counterparts with no self-disclosed criminal records. Taken together, these results indicate employment outcomes for those with criminal records improve with time free of arrest or conviction.
The above findings have several important policy implications. First, findings here show that applicants with ten-year old criminal records received 33% fewer callbacks than those with no criminal record. This is a crucial finding, as it shows that previous research demonstrating a declining risk of offending over time has not fully influenced employers. Therefore, affirmative steps should be taken, such as providing employers with literature informing them of the diminished value of older criminal records in predicting future criminal activity and thus employability. Second (and relatedly), those in the executive and judicial branches charged with deciding pardons and expungements should also be provided with such information regarding the declining risk of offending and the punitive nature of older criminal records. This is extremely important as the length of a law-abiding period is considered one of the most important factors in pardon and expungement applications; however, it is not clear whether pardon boards have reliable guidelines as to how long a law-abiding period should be for the individual to be deemed appropriate for pardon or expungement.
Third, legislators could create statutes that specifically protect employers from negligent hiring claims when hiring individuals with old criminal records. Interestingly, the immunization of employers from negligent hiring claims is a common fixture of many of the new so-called certificates of recovery/relief. Certificates of recovery/relief are intended to demonstrate that former offenders have been rehabilitated, while not sealing the applicant’s record. Offenders are helped in their employment search by these certificates because such mechanisms remove automatic licensing bars for those with criminal records, offer a stamp of good character from a court, and, as mentioned above, protect employers who hire ex-offenders from negligent hiring claims. Adopting such statutes could significantly aid ex-offenders in their job searches when coupled with statutes such as the already available Work Opportunity Tax Credit and Federal Bonding programs, which aim to aide ex-offenders in securing employment by providing potential employers with financial benefits.
Fourth, state repositories of criminal records could institute policies to not disclose criminal records older than a certain number of years since the last conviction. Similarly, as many employers rely on background-check services provided by commercial vendors of criminal records, a requirement that those old records also be erased from commercial databases should accompany a non-dissemination policy. However, recent case law indicates the limits of such policies for private individuals or organizations as they can be seen as violating First Amendment rights.
These findings indicate that an individual possessing even a ten-year-old criminal record is hindered in his employment search. While the current study offers an important step forward, the low sample size and thus low power of our study necessitates that future research replicate our findings to confirm their significance. Further, because our study did not find redemption after a period of ten years, research is needed to determine at what point in time employment outcomes for former offenders who have stayed crime-free are indistinguishable from their counterparts with clean criminal backgrounds.
∞ Peter Leasure is a Ph.D. candidate in the Department of Criminology and Criminal Justice at the University of South Carolina. Tia Stevens Andersen is an Assistant Professor in the Department of Criminology and Criminal Justice at the University of South Carolina.
1. El v. Se. Pa. Transp. Auth. (SEPTA), 479 F.3d 232 (3d Cir. 2007).
2. Id. at 245
3. Alfred Blumstein, Ph.D. is the J. Erik Jonsson University Professor of Urban Systems and Operations Research at the Heinz College and Department of Engineering and Public Policy at Carnegie Mellon University and has published numerous papers on criminal careers, deterrence, prison population, and drug enforcement policy. Blumstein received the 2007 Stockholm prize in criminology.
4. El, 479 F.3d at 246.
5. Id. at 247.
6. See Alfred Blumstein & Kiminori Nakamura, Redemption in the Presence of Widespread Criminal Background Checks, 47 Criminology 327, 327 (2009); Megan C. Kurlychek, Shawn D. Bushway & Robert Brame, Long-Term Crime Desistance and Recidivism Patterns: Evidence from the Essex County Convicted Felon Study, 50 Criminology 71 (2012); Megan C. Kurlycheck, Robert Brame, & Shawn D. Bushway, Scarlet Letters and Recidivism: Does an Old Criminal Record Predict Future Offending, 5.3 Criminology & Pu. Pol’y 483 (2006); Megan C. Kurlycheck, Robert Brame & Shawn D. Bushway, Enduring Risk? Old Criminal Records and Predictions of Future Criminal Involvement, 53.1 Crime & Delinq 64 (2007).
7. See, e.g., Devah Pager, Evidence Based Policy for Successful Prisoner Reentry, 5 Criminology & Pub. Pol’y 505 (2006); Christopher Uggen, Work as a Turning Point in the Life Course of Criminals: A Duration Model of Age, Employment, and Recidivism, 67 Am. Soc. Rev. 529 (2000).
8. See Richard D. Schwartz & Jerome H. Skolnick, Two Studies of Legal Stigma, 10.2 Soc. problems 133, 13638 (1962) (an early correspondence design showing the negative effects of a criminal record on employment outcomes); Devah Pager, The Mark of a Criminal Record, 108.5 Am. J. soc. 937, 960 (2003); Devah Pager, Bruce Western & Bart Bonikowski, Discrimination In a LowWage Labor Market: A Field Experiment, 74.5 Am. soc. rev. 777 (2009); Scott H. Decker, Natalie Ortiz, Cassia Spohn & Eric Hedberg, Criminal Stigma, Race, and Ethnicity: The Consequences of Imprisonment For Employment, 43.2 J. Crim. Justice 108 (2015); Christopher Uggen, Mike Vuolo, Sarah Lageson, Ebony Ruhland & Hilary K. Whitham, The Edge of Stigma: An Experimental Audit of the Effects of Low‐Level Criminal Records on Employment, 52.4 Criminology 627 (2014).
9. The study included: (1) an applicant who had been convicted and sentenced for assault; (2) an applicant who had been tried for assault but acquitted; (3)an applicant who had been tried for assault, acquitted and had a letter from the judge certifying the applicant’s acquittal and emphasizing the presumption of innocence; and (4) an applicant who had no criminal record.
10. Schwartz & Skolnick, supra note 8, at 136.
11. Pager, supra note 8, at 955, 957.
12. Pager, supra note 8, at 957, 959 .
13. See Decker, Ortiz, Spohn & Hedberg, supra note 8; Pager, Western & Bonikowski, supra note 8.
14. Uggen, Vuolo, Lageson, Ruhland & Whitham, supra note 8.
15. Debbie A. Mukamal & Paul N. Samuels, Statutory Limitations on Civil Rights of People with Criminal Records, 30 Fordham Urb. L.J. 1501, 1510 (2002).
16. Bureau of Justice Statistics, Survey of State Criminal History Information Systems, 2006 (2008); Bureau of Justice Statistics, Survey of State Criminal History Information Systems, 2008 (2009); see also Shawn D. Bushway, Michael A. Stoll, David Weiman , Barriers to Reentry? The Labor Market for Released Prisoners in Post Industrial America 174–200 (2007) (raising concerns about the accuracy of criminal history information provided by private internet based services).
17. SEARCH, The National Consortium for Justice Information and Statistics, Report of the National Task Force on the Criminal Background of America (2005).
18. Justina Victor, Background Checking: The Use of Criminal Background Checks in Hiring Decisions (Society for Human Resource Management) 2 (2012).
19. Bureau of Justice Statistics, Survey of State Criminal History Information Systems, 2014 (U.S. Department of Justice, 2015), https://www.ncjrs.gov/pdffiles1/bjs/grants/249799.pdf; Michelle Natividad Rodriguez and Maurice Emsellem, “65 Million ‘Need Not Apply’: The Case For Reforming Criminal Background Checks For Employment” (New York: National Employment Law Project, 2011), http://www.nelp.org/content/uploads/2015/03/65_Million_Need_Not_Apply.pdf.
20. Civil Rights Act § 2000e–5(a) (1964).
21. Green v. Mo. Pac. R.R., 549 F.2d 1158, 1160 (8th Cir. 1977).
22. U.S. Equal Employment Opportunity Commission, Consideration of Arrest and Conviction Records in Employment Decisions Under Title VII of the Civil Rights Act of 1964 (U.S. Equal Employment Opportunity Commission) (2010), www.eeoc.gov/laws/guidance/arrest
25. See supra note 6.
26. Alfred Blumstein, David P. Farrington & Soumyo Moitra, Delinquency Careers: Innocents, Desisters, and Persisters, 6 Crime and Justice 187 (1985); Robert Brame, Shawn D. Bushway & Raymond Paternoster, Examining the Prevalence of Criminal Desistance, 41 Criminology 423 (2003); David P. Farrington, Predicting Individual Crime Rates, 9 Crime and Justice 53 (1987); Alex R. Piquero, David P. Farrington, & Alfred Blumstein, The Criminal Career Paradigm, 30 Crime and Justice: A Review of Research 359 (2003).
27. Alfred Blumstein & Joel Wallman, The Crime Drop in America (Cambridge University Press) (2006); Robert J. Sampson, John H. Laub & Christopher Wimer, Does Marriage Reduce Crime? A Counterfactual Approach to Within Individual Causal Effects, 44 Criminology 465 (2006); Christopher Uggen, Ex-Offenders and the Conformist Alternative: A Job Quality Model of Work and Crime, 46 Social Problems 127 (1999); Mark Warr, Life Course Transitions and Desistance from Crime, 36 Criminology 183 (1998).
28. David P. Farrington, Age and Crime, 7 Crime and Justice 189 (1986); Travis Hirschi & Michael Gottfredson, Age and the Explanation of Crime, 89 American Journal of Sociology 552, 564 (1983).
29. Allen J. Beck & Bernard E. Shipley, Recidivism of Young Parolees (Bureau of Justice Statistics) (1987); Patrick A. Langan & David J. Levin, Recidivism of prisoners released in 1994 (Bureau of Justice Statistics) (2002); Michael D. Maltz, Recidivism (Academic Press) (1984); Peter Schmidt & Ann Witte, Predicting Recidivism Using Survival Models (Springer Science & Business Media) (2012); Christy A. Visher, Pamela K. Lattimore & Richard L. Linster, Predicting the Recidivism of Serious Youthful Offenders Using Survival Models, 29 Criminology 329 (1991).
30. Don M. Gottfredson, Effects of Judges’ Sentencing Decisions on Criminal Careers (National Institute of Justice) 4 (1999).
31. Analysis of Essex County recidivism data in The Declaration of Jeffrey Fagan, Ph.D. in Nixon v. The Commonwealth of Pa., cited in Blumstein and Nakamura supra note 6, at 331–32.
32. Supra note 6.
33. El vs. Se. Pa. Transp. Auth. (SEPTA), 479 F.3d at 247.
34. Kurlychek, Brame & Bushway (2006), supra note 6, at 483.
35. Kurlychek, Brame & Bushway (2006), supra note 6, at 499.
36. Kurlychek, Bushway & Brame (2007), supra note 6.
37. Kurlychek, Bushway & Brame (2007), supra note 6, at 78.
38. See generally Blumstein & Nakamura, supra note 6.
39. Blumstein & Nakamura, supra note 6, at 344. This study did not account for rearrests that occurred outside of New York state.
40. Blumstein & Nakamura, supra note 6, at 338–39.
41. Blumstein & Nakamura, supra note 6, at 341–44.
42. Keith Soothill & Brian Francis, When Do Ex-Offenders Become like Non-Offenders?, 48 Howard J. of Crim. Just. 373 (2009). This study relied on official data, as did the Blumstein & Nakamura study, supra note 6, but it used conviction as opposed to arrest data. The Soothill & Francis study also uses data for an entire nation, as opposed to one city (cf. Kurlychek, Brame, & Bushway, supra note 6) or one state (cf. Blumstein & Nakamura, supra note 6).
43. Soothill & Francis, supra note 43, at 385.
44. Soothill & Francis, supra note 43, at 38384.
45. See generally Shawn D. Bushway, Paul Nieuwbeerta & Arjan Blokland, The Predictive Value of Criminal Background Checks, 49 Crim. 27 (2011).
46. Id. at 27.
47. See supra note 6 and accompanying text.
48. See Pager, supra note 8, at 948–49 (citing Marc Bendick Jr., Lauren E. Brown & Kennington Wall, No Foot in the Door, 10 J. Aging & Soc.Pol’y 5 (1999)).
50. U.S. Bureau of Labor Statistics, Columbus, OH, Area Economic Summary, BLS.gov (updated Nov. 2, 2016) http://www.bls.gov/regions/midwest/summary/blssummary_columbus_oh.pdf.
51. U.S. Bureau of Labor Statistics, Columbus, OH Economy at a Glance, BLS.gov, http://www.bls.gov/eag/eag.oh_columbus_msa.htm (last visited Nov. 19, 2016).
52. Danielle Kaeble, Lauren Glaze, Anastasios Tsoutis, & Todd Minton, Correctional Populations in the United States, 2014, U.S. Dept. of Just. at fig. 4 (Jan. 2016), https://www.bjs.gov/content/pub/pdf/cpus14.pdf.
53. Id. at app. tbl. 1.
54. Jeremy Travis, But They All Come Back: Rethinking Prisoner Reentry (National Institute of Justice) (2000).
55. E. Ann Carson, Prisoners in 2014 (Bureau of Justice Statistics) at tbl. 7 (2015), https://www.bjs.gov/content/pub/pdf/p14.pdf.
56. See Pager, supra note 8.
57. Joanna N. Lahey & Ryan A. Beasley, Computerizing Audit Studies, 70 J. Econ. Behav. Organ. 508 (2009).
58. Kevork Djansezian, Most Inmates Entering Ohio State Prisons are High School Dropouts, Idea Stream, Apr. 2, 2012; Caroline Wolf Harlow, Education and Correctional Populations (Bureau of Justice Statistics 2003).
59. See Pager, Western & Bonikowski, supra note 8; Pager, supra note 8.
60. In other words, we did not use block randomization which results in a slightly uneven distribution of treatment conditions to employers. See infra note 65 addressing any concerns from this approach.
61. The first author observed reentry practices at several Columbus, Ohio reentry centers for three years. The distribution of jobs in our sample was consistent with the distribution of positions sought by ex-offenders at the Columbus, Ohio reentry facilities. Entry-level positions included jobs in administrative/clerical, customer service, restaurant/grocery, sales, driving, warehouse/shipping, manufacturing, and general labor. These categories are used later for descriptive purposes.
62. We focused on recent postings to avoid applying for positions already filled.
63. Point estimates of the proportion of applicants receiving a positive response are documented by solid black circles. The set of lines surrounding each point estimate represent the 95% confidence interval for the estimate (our uncertainty due to sampling error). Overall Likelihood Ratio χ2 = 12.700, p < .001; No criminal record vs. One-year-old felony Likelihood Ratio χ2 = 12.691, p < .001; Oneyear-old felony vs. Ten-year-old felony Likelihood Ratio χ2 = 3.489, p < .10; No criminal record vs. Ten-year-old felony Likelihood Ratio χ2 = 2.649, n.s.
64. To test the robustness of our results, we also utilized a logistic regression model specified with an inverse probability weights estimator and robust standard errors. This model also controlled for job type. The results from this approach confirm the point estimates and confidence intervals presented in Figure 1. The results were as follows: The potential outcome mean (predicted probability of a positive callback response) for the No criminal record group was 28.93% (with a confidence interval bound of +/ 8.6%). The potential outcome mean (predicted probability of a positive callback response) for the One-year-old felony group was 9.82% (with a confidence interval bound of+/ 5.78%). The potential outcome mean (predicted probability of a positive callback response) for the Ten-year-old felony group was 19.77% with a confidence interval bound of +/ 8.09%).
65. For a description of the importance of effect sizes, rather than statistical significance, see David Weisburd, Cynthia M. Lum & SueMing Yang, When Can We Conclude That Treatments or Programs Don’t Work, 587 Annals Am. Acad. Pol. & Soc. Sci. 31 (2003); Michael D. Maltz, Deviating from the Mean: The Declining Significance of Significance, 31 J. Res. Crime Delinq. 434 (1994); Mark W. Lipsey, Design Sensitivity: Statistical Power for Experimental Research (SAGE, 1990).
66. See Blumstein & Nakamura, supra note 6.
67. But see Margaret Colgate Love, Paying Their Debt to Society: Forgiveness, Redemption, and the Uniform Collateral Consequences of Conviction Act. 54.3 How. L.J. 753, 775-–78 (2011) (noting the lack of uniformity in expungement laws and how the development of private repositories makes the remedy less effective).
68. For examples of certificates of recovery/relief, see Cal. Bus. & Prof. Code § 480(b) (West 2016); 730 Ill. Comp. Stat. Ann. 5/55.525 (West 2016); N.J. Stat. Ann. § 2A:168A7 (West 2016); Ariz. Rev. Stat. Ann. §§ 13904 to 908 (2016); N.Y. Correct. Law §§ 700-06 (McKinney 2016). For an example of a certificate of qualification for employment, see Ohio Rev. Code Ann. § 2953.25 (West 2016).
69. See Peter Leasure & Tia Stevens Andersen, The Effectiveness of Certificates of Relief as Collateral Consequence Relief Mechanisms: An Experimental Study, Yale L. & Pol’y Rev. Inter Alia (11/7/2016), http://ylpr.yale.edu/inter_alia/effectivenesscertificatesreliefcollateralconsequencereliefmechanismsexperimental, for a demonstration of the effectiveness of such certificates.
70. Love, supra note 68.
71. See U.S. Dept. of Labor at https://www.doleta.gov/business/incentives/opptax/.
72. See U.S. Dept. of Labor at http://bonds4jobs.com/.
73. Clay Calvert & Jerry Bruno, When Cleansing Criminal History Clashes with the First Amendment and Online Journalism: Are Expungement Statutes Irrelevant in the Digital Age, 19 CommLaw Conspectus: J. Comm. L. Pol’y 123 (2010); Martin v. Hearst Corp., 777 F.3d 546 (2d Cir. 2015).
74. Lawrence W. Sherman & Heather Strang, Verdicts or Inventions? Interpreting Results from Randomized Controlled Experiments in Criminology, 47 Am. Behav. Scientist 575 (2004).
Scholars discuss the most significant immigration-related cases before the U.S. Supreme Court, their ramifications, and what to expect in 2020.
"It's important to note that scholars have long observed that political discourse and political events can contribute to the frequency of bias incidents. In fact, this phenomenon has a name today. It's called the Trump Effect."
Experts discuss legal developments and related ramifications one year after President Trump declared a national emergency at the U.S. Southern Border with Mexico in order to build a wall.
The discriminatory laws, practices, and policies promised and delivered by President Trump have social, political, and economic ramifications. First, they reinforce misconceptions about Islam as an inherently violent religion. Second, they breed intolerance, fear, and hostility among the general population