Breadcrumb

  1. Home
  2. Federal Sector
  3. Federal Sector Reports
  4. Second Chances Part I - Federal Employment for Workers with Past Arrests or Convictions

Second Chances Part I - Federal Employment for Workers with Past Arrests or Convictions

Table of Contents

Executive Summary

Background on the Employment of Formerly Incarcerated Workers

Rules for Hiring Workers with Arrest or Conviction Records

Appropriate Use of Arrest and Conviction Records Under Title VII

Suitability Decisions, Arrest/Conviction Records, and Federal Hiring

Results

Available Data Suggests That People With Prior Arrests or Convictions Are Unlikely to Become Federal Workers

Ban-the-box policies enabled more and more meritorious EEOC complaints

Conclusions, Recommendations, and Next Steps

Conclusions

Recommendations

Next Steps

References

 

Executive Summary

In June 2021, President Biden signed Executive Order 14035 to advance diversity, equity, inclusion, and accessibility in the Federal workforce. This order called for an evaluation and expansion of Federal employment opportunities for formerly incarcerated persons.

Before this executive order, in fiscal year (FY) 2020, the U.S. Equal Employment Opportunity Commission (EEOC) formed a task force to identify vulnerable workers and determine ways to better serve them. The EEOC identified formerly incarcerated persons as one category of vulnerable workers due to the challenges they face in securing employment after their incarceration. In the FY 2017-2021 Strategic Enforcement Plan, the EEOC identified the use of background checks related to arrest and conviction records as among its national substantive area priorities because African Americans and Latinos are disproportionately incarcerated.

This report comes from the EEOC’s Reports and Evaluations Division at the Office of Federal Operations (OFO). It asks two related questions:

1)    How likely are people with prior arrests or convictions to work in the Federal sector?

2)    Could regulating the timing of background checks during the recruitment process (e.g., ban-the-box policies) protect those applicants with a prior arrest or conviction from discrimination?

How Likely Are People With Prior Arrests or Convictions to Work in the Federal Sector?

In answering the first question, this report first notes the lack of systematic data collection tracking employment outcomes for workers with prior records of incarceration. Using a nationally representative survey with data from 2003 through 2017, this report finds that people with prior records were nearly half as likely to be Federally employed compared to those without such records. This gap suggests that there were roughly 300,000 fewer previously formerly incarcerated workers in the Federal workforce than expected.

The current study does not have sufficient data to fully assess the causes of lower rates of Federal employment for individuals with a history of incarceration during the 2003 to 2017 time period. For instance, it is possible that people with incarceration records erroneously believe they are barred from Federal employment and self-select out of the applicant pool. It is also possible that hiring managers are less likely to hire applicants with any kind of record -incarceration, arrest, or conviction. Additional research directly examining the hiring practices of Federal agencies towards workers with a record is necessary to better understand how employment for these community members could be increased.

The lack of available data constrains the ability to measure Federal employment for workers with conviction, arrest, or incarceration records. Therefore, one high-value policy change would be to track the number of Federal workers with arrest or conviction records over time. Existing data could be leveraged to generate such a measure. For example, researchers could use data from the Defense Counterintelligence and Security Agency (DCSA)—the central processor of background investigations for the Federal government—to determine the proportion of Federal employees that had a background check return an arrest or conviction record.[1] Tracking this statistic over time would allow the EEOC and other Federal agencies to assess how successful different policies and procedures are in expanding Federal employment opportunities to formerly incarcerated persons. The EEOC should coordinate with the Office of Personnel Management (OPM) and the DCSA to gather data on suitability adjudications, with a focus on the characteristics of applicants whom employers deem to be appropriate to hire for a given position.

Could Ban-the-Box Policies Protect Applicants With a Prior History of Arrest or Conviction From Discrimination?

To answer the second question, this study looks to lessons learned from outside the Federal sector. Researchers examined data on complaints filed with the EEOC where complainants alleged that their past exposure to the criminal legal system was improperly adjudicated. The results show that so-called ban-the-box policies, which generally prohibit criminal background checks until after a conditional job offer is made, led to significantly more and more meritorious complaints filed per month. Specifically, ban-the-box policies made it more likely that investigators were able to find enough evidence to proceed further in the complaint process. Without additional data and analysis, it is difficult to determine the exact cause behind these results. It is possible that, after ban-the-box timing restrictions were imposed, employers engaged in more discrimination based on arrest and conviction records. However, it may also be the case that applicants were more aware of their protections after timing restrictions were implemented.

Title 5 of the Code of Federal Regulations regarding Recruitment, Selection, and Placement (General) and Suitability (5 CFR parts 330 and 731) prohibit Federal employers from collecting criminal history information, unless an exception has been granted until after until after a conditional job offer has been made and accepted.[2] The analysis presented in this study suggests that such a policy may be helpful in enforcing Title VII and should be continued. These recruitment rules were expanded to Federal contractors in December 2021, which may improve Title VII monitoring and enforcement even more.[3]

To ensure adherence to best practices, recruiters should follow these rules and regulations, and the EEOC should conduct follow-up research to ensure Federal employers are following best practices. Applicants should also be made aware of their rights regarding the accuracy of background checks and the appropriate uses of arrest and conviction records in employment decisions.

Background on the Employment of Formerly Incarcerated Workers

In June 2021, President Biden signed Executive Order (EO) 14035 to advance diversity, equity, inclusion, and accessibility in the Federal workforce. This order called for an evaluation and expansion of Federal employment opportunities for formerly incarcerated persons (Executive Order No. 14035, 2021).

Even before EO 14035 was issued, in Fall 2019, the U.S. Equal Employment Opportunity Commission (EEOC) formed a task force to identify vulnerable workers and determine ways to better serve them. Formerly incarcerated persons were among those identified as vulnerable workers due to their difficulty in securing employment after their incarceration, which includes pre-trial detention and post-conviction incarceration. The task force built upon 2012 guidance on the use of arrest and conviction records in employment decisions. Furthermore, because Black and Latino people are disproportionately incarcerated, the EEOC identified the use of criminal background checks as among its national substantive area priorities in the FY 2017-2021 Strategic Enforcement Plan. The analysis contained throughout this report assesses Federal employment for formerly incarcerated people and the policies that promote or frustrate Federal work for these community members.

Prior incarceration impacts many U.S. citizens’ abilities to find employment. When formerly incarcerated community members find stable jobs, they promote public safety and break cycles of economic hardship for future generations of Americans. Yet in 2017, 622,400 people were released from state and Federal prisons (Bronson, 2019), and many of them faced challenges reentering the workforce (Couloute & Kopf, 2018). Additionally, the costs imposed by the judicial system are not borne equally by all. Research estimates that 8 percent of all adults and one-third (33 percent) of African American men had a felony record as of 2010 (Shannon et al., 2017). Another study found that African American adults were nearly 6 times more likely to be incarcerated than White adults in 2018, while Hispanic adults were about 3 times as likely to be incarcerated compared to White adults (The Sentencing Project, 2018). These data likely represent a conservative estimate of racial disparities in the criminal legal system given that the studies described above do not include individuals who were arrested or convicted but did not serve time in jail or prison for a felony.

Substantial evidence suggests that applicants with arrest or conviction records face many barriers to employment. For example, individuals with arrest or conviction records have much lower employment rates than those without these records.[4] Mueller-Smith and Schnepel (2021) found that, for first-time felony defendants, avoiding a felony conviction through diversion (i.e., wherein public officials choose to pause, terminate, or divert someone’s progression through the justice system) halved recidivism (re-offending) rates, increased quarterly employment by 53 percent (or 18 percentage points), and increased quarterly earnings by 64 percent. In other words, a person who avoids a felony conviction works almost two more years and earns about $60,000 more than if they had been convicted. While some employers worry about the performance of workers with arrest or conviction records, a five-year study involving almost 500 ex-offenders found lower turnover rates among ex-offenders compared to non-offenders (Paulk, 2016).

Research has also shown that a lack of employment for the formerly incarcerated community members harms national productivity and increases recidivism, racial income inequality, and crime rates (Abraham & Kearney, 2020; Schnepel, 2018). Research has shown that people released from prison into counties during better economic conditions returned to prison at significantly lower rates than individuals released into counties experiencing economic downturns (Yang, 2017). Employment is a key part of returning to and reintegrating into communities—decreasing not only the likelihood of reoffending but also overreliance on public assistance programs.[5] Prior research has focused almost exclusively on private and state employers, while employment in local and Federal agencies for formerly incarcerated persons remains understudied.

Another area of interest to the EEOC is credit history, which may be a factor in suitability determinations, depending upon the risk and sensitivity of the position in question. Credit histories are governed by many of the same processes and rules as criminal histories, especially in the Federal sector. However, criminal histories and credit histories may play different roles in the Federal hiring process than in the private sector, due to a more uniform process for criminal background checks and different potential liabilities, including national security considerations. Since EEOC researchers could not obtain data on credit history, this report focuses solely on the impact of past exposure to the criminal legal system. Despite the lack of data to study the impact of credit histories on the Federal hiring process, many of the patterns and policies considered in the criminal history context have parallels to credit history, as shown in previous research (Bartik & Nelson, 2018; Dobbie et al., 2020).

This report presents evidence that, from 2003 to 2017, the Federal sector may have had higher barriers to employment for formerly incarcerated workers than elsewhere in the economy. However, this report also shows that delaying arrest and conviction record inquiries in the hiring process may provide the EEOC more opportunity to protect these applicants.

A related READD report, Second Chances Part II: History of Criminal Conduct and Suitability for Federal Employment, served to take up the “Next Steps” offered at the end of this present report.

Rules for Hiring Workers with Arrest or Conviction Records

Several laws and regulations govern the use of past arrests or convictions in Federal employment decisions. Federal agencies collect the information to conduct background investigations on their job candidates. The hiring agency enters information concerning the processing of candidates’ investigations is into a centralized database, which is managed by the Defense Counterintelligence and Security Agency (DCSA). The hiring agency, following the adjudication of information received from the background investigation, reports determinations to Office of Personal Management (OPM). The information collected regarding the prior arrests or conviction is the OPM Form OF-306, which specifically asks: “During the last 7 years, have you been convicted, been imprisoned, been on probation, or been on parole?”

Appropriate Use of Arrest and Conviction Records Under Title VII

The EEOC has a long history of combating discrimination caused by employers’ use of arrest or conviction records in employment decisions—with decisions dating back to the 1970s.[6] While having a prior arrest or conviction is not a protected category under Title VII, discrimination based on race, color, religion, sex, and national origin are prohibited. The EEOC has issued multiple policy statements on the appropriate consideration of arrest or conviction records in employment decisions.[7] The most recent and comprehensive of these guidance documents was EEOC Enforcement Guidance, Consideration of Arrest and Conviction Guidance in Employment Decisions, Apr. 25, 2012 (Arrest/Conviction Guidance). The Arrest/Conviction Guidance clarifies how an employer’s use of an individual’s criminal history in making employment decisions may violate the prohibition against employment discrimination prohibited by Title VII.[8] The Arrest/Conviction Guidance superseded previous EEOC policy statements on using arrest or conviction records in employment decisions and is intended for use “by employers considering the use of criminal records in their selection and retention processes; by individuals who suspect that they have been denied jobs or promotions, or have been discharged because of their criminal records; and by EEOC staff who are investigating discrimination charges involving the use of criminal records in employment decisions” Section II.

The Arrest/Conviction Guidance describes two types of discrimination that can arise from an employer’s consideration of arrest and conviction records: disparate treatment discrimination and disparate impact discrimination. For example, disparate treatment discrimination occurs when an employer rejects a Black applicant based on a prior arrest or conviction but hires a similarly situated White applicant with a comparable arrest or conviction. Title VII requires employers to treat arrest and conviction records the same across protected classes. Disparate impact discrimination occurs when an employer uses policies or practices that do not intentionally discriminate on a protected basis on its face but, as applied, the policy or practice disproportionately screens out people on a protected basis and the policy is not job related or consistent with a business necessity.

Consideration of arrest and conviction records in employment decisions is a hiring policy or practice. For criminal conduct exclusions, relevant information includes the text of the policy or practice, associated documentation, and information about how the policy or practice was implemented. Additional relevant information to consider is “which offenses or classes of offenses were reported to the employer (e.g., all felonies, all drug offenses); whether convictions (including sealed or expunged convictions), arrests, charges, or other criminal incidents were reported; how far back in time the reports reached (e.g., the last five, ten, or twenty years); and the jobs for which the criminal background screening was conducted.” Section V(A)(1).

Black men, in particular, are arrested and incarcerated at rates disproportionate to their numbers in the national population. These national disparities provide a basis for the EEOC to investigate a Title VII disparate impact charge; however, some courts have found that local data can undercut national disproportionalities and make a legal finding of disparate impact less likely at a local level.[9] For example, an employer might present regional data that shows Black men are not arrested/convicted at disproportionally higher rates than White men within the locality in question.

Additionally, employers have an opportunity to show why using a specific policy related to arrest and conviction records is job related and consistent with business necessity. This means that the policy must “bear a demonstrable relationship to successful performance of the jobs for which it was used” and “measures the person for the job and not the person in the abstract.”[10] As detailed in the Arrest/Conviction Guidance, the business necessity defense may be met in two ways:

  • The employer validates the criminal conduct screen for the position in question using the Uniform Guidelines on Employee Selection Procedures (29 C.F.R. § 1607.5).
  • The employer uses targeted screening that considers at least the nature of the crime, the time elapsed, and the nature of the job. The employer then provides an opportunity for an individualized assessment for people excluded by the screen, to determine whether the policy as applied is job related and consistent with business necessity.
  • As described in the Arrest/Conviction Guidance, an individualized assessment should include:
  • A notice to the applicant that he has been rejected because of past criminal conduct.
  • An opportunity for the applicant to show that they should not be rejected because of their particular circumstances.
  • An employer reevaluation as to whether the applicant’s additional information warrants an exception to the exclusion and shows that the policy as applied to the applicant is not job related and consistent with business necessity.

Best hiring practices should include an individualized assessment of applicants excluded because of past criminal conduct. As the Arrest/Conviction Guidance explains, the following factors are relevant:

  • The facts or circumstances surrounding the offense or conduct.
  • The number of offenses for which the individual was convicted.
  • Older age at the time of conviction or release from prison.
  • Evidence that the individual performed the same type of work, post-conviction, with the same or a different employer, with no known incidents of re-offending.
  • The length and consistency of employment history before and after the offense or conduct.
  • Rehabilitation efforts (e.g., education and training).
  • Employment or character references and any other information regarding fitness for the particular position.
  • Whether the individual is insured under a Federal, state, or local bonding program.[11]

Suitability Decisions, Arrest/Conviction Records, and Federal Hiring

Federal employment requires that the employee be reliable, trustworthy, of good conduct and character, and loyal to the United States. In other words, to be hired for a Federal job, an applicant must be deemed suitable. This suitability decision is separate from whether the applicant has the requisite skills and qualifications to perform the job’s tasks. Rather, suitability inquiries assess when the jobseeker’s identifiable character or contacts may have an impact on the integrity or efficiency of their service. This includes questions about an applicant’s honesty, sound judgment, reliability, responsibility, and ability to follow rules. All competitive service positions require a suitability determination whereas excepted service and contractor positions may require a suitability-like determination, referred to as fitness.

There are different levels of investigation based upon the risk and sensitivity level of the position. Low risk positions require less investigation than do public trust (moderate or high risk) and/or national security sensitive positions. The risk and sensitivity level of the position may be relevant to the suitability determination, suitability decisions are handled by the hiring agency. A smaller number of suitability decisions are under OPM’s purview when certain conditions are met or at OPM’s discretion. For instance, OPM may make a suitability decision when there is evidence of an intentional false statement or fraud.

For hiring within the Federal government, the depth of a background investigation is influenced by the tier of the position, with higher tiers getting more scrutiny. Investigative forms have branching questions that only have to be answered in the event of affirmative answers to certain questions. While background investigations at the higher tier include a credit check, the investigation at the lowest tier will not in every instance, it may forgo a credit check for Tier 1 applications that have no other flags. In most cases, the criminal history is checked during the course of the background investigation which is conducted by an authorized investigative service provider. The investigation is not initiated under the individual accepts the conditional offer. After conditional offer, agency may ask applicants about their criminal history, such as via the OF 306. Other agencies might differ slightly in their procedures, but further research, which could be accomplished by surveying those adjudicating hiring within Federal agencies, should be conducted to better understand equity and fairness outcomes of the Federal processes on a systemic level.

Evidence of “criminal or dishonest conduct” is one of eight bases for which an agency can find an applicant unsuitable (5 CFR 731.202(b)(2)). The Declaration for Federal Employment, Optional Form (OF) 306 typically collects this information via two questions:

  • Questions 9, which asks, “During the last 7 years, have you been convicted, been imprisoned, been on probation, or been on parole?”
  • Question 13, which asks, “Are you delinquent on any Federal debt?”[12]

Additionally, completion of this form requires the applicant to consent to the release of arrest, conviction, and credit information. If an agency makes an unfavorable suitability determination, it may also take a suitability action, following the procedures specified in 5 CFR part 731. Suitability actions by agencies can include cancelation of eligibility, cancelation of reinstatement eligibility, removal, and/or debarment.

Changes to the agency that performs background investigations

The agency in charge of performing background checks has changed in recent years. The National Defense Authorization Act of 2018 mandated that the Department of Defense (DoD) take over the background investigation for its own personnel, in part due to concerns about the accuracy, wait time, and cost of security investigations (Government Accountability Office, 2018; Bur, 2019). OPM delegated to DoD authority to conduct investigations for other agencies for the purpose of determining suitability, fitness, or eligibility to hold a personal identity verification credential to keep background investigation services consolidated with one primary investigative service provider.[13]

Rescission of a conditional offer of employment due to prior arrest or convictions is not generally appealable

Applicants may not appeal an agency’s decision under 5 CFR 332 of “non-selection” (i.e., rescission of a conditional job offer) to the Merit Systems Protection Board. With respect to prior arrest or conviction records, agencies can change an applicants’ determination of eligibility after adjudication (i.e., completing an individualized assessment of the arrest and/or conviction) if there is a foreseeable potential risk in completing the hire. Moreover, the cancellation of a tentative job offer based on an objection under 5 CFR 332.406 is not appealable even if it is based on the criteria for making suitability determinations set forth in 5 CFR 731.202.

Other major rules governing the use of criminal history

The Fair Credit Report Act (FCRA) provides protections to applicants when credit or criminal reports are considered as a basis for denying employment. Under the FCRA, when information from a background report is used in an adverse action against the applicant, the applicant is entitled to know what was in the background report. Applicants also have the right to dispute incomplete or inaccurate information in the background report.

In the Federal process, a number of systems may be queried to assess an applicant’s suitability, including the State Criminal History Repository Check, Credit Check, and the National Crime Information Center/Interstate Identification Index Check. These checks are not foolproof. For instance, studies have shown that 50 percent of arrest records in the FBI’s Interstate Identification Index were associated with arrests that did not lead to convictions—substantially delaying the hiring process and potentially resulting in otherwise qualified candidates missing out on employment opportunities (Nancy et al., 2017). In addition, in any of these queries, there is the possibility of a false positive, or mistakenly flagging someone as having an arrest or conviction record.

When background checks can start in the Federal hiring process (“ban-the-box”)

In 2016, OPM issued a final rule revising when a hiring agency can request information typically collected during a background investigation for Federal employment (Office of Personnel Management, 2016). The rule became effective in January 2017 with compliance required by March 31, 2017. Like many “ban-the-box” rules implemented by state and local governments, OPM’s rule requires federal agencies to make a conditional offer of employment before it reviews or requests the content of OF-306, the declaration for federal employment form that collects applicants’ background information.[14]

This rule is not absolute. In certain situations, agencies may have a business need to obtain information about the background of applicants earlier in the hiring process to determine if they meet the qualifications or suitability requirements. If so, agencies must request an exception from OPM. OPM will only grant exceptions when the agency shows specific job-related reasons why the agency needs to evaluate an applicant’s arrest, conviction or adverse credit history earlier in the process or consider the disqualification of candidates with prior arrests, convictions or other conduct issues from particular types of positions.[15] 

The Fair Chance to Compete for Jobs Act of 2019 (Fair Chance Act) codified OPM’s ban-the-box rule in federal statute and extended the rule’s application to Federal contractors. (Public Law 116–92, 2019). The Fair Chance Act explicitly covers almost all executive agencies (including cabinet agencies and the U.S. Postal Service, but not the armed forces), the legislative branch, and the judicial branch of the Federal Government (other than judges, justices, and magistrates). It also applies to civilian agency and defense contractors. Finally, it requires each agency to establish a complaint and appeal procedure for violations of the Fair Chance Act.[16]

Overview of other (non-Federal) ban-the-box policies

Ban-the-box started as a grassroots policy supported by nonprofit organizations like All of Us or None and the National Employment Law Project (NELP). These policies aim to encourage employers to hire formerly incarcerated workers by removing arrest and conviction inquiries from the job application. Ban-the-box proponents hoped to increase applications from formerly incarcerated job seekers and encourage employers to make individualized assessments rather than categorical exclusions of these community members. For example, employers would evaluate each prior arrest or conviction by considering the age of the offense and its relevance to the job at hand. This also allows applicants to present evidence of rehabilitation and review background checks, giving them a chance to “get a foot in the door” and not be prejudged by a prior arrest or conviction. By humanizing the applicant and encouraging the employer to assess work-readiness, ban-the-box policies aim to minimize discrimination against applicants with an arrest or conviction record as a class (Avery & Lu, 2020).

Ban-the-box refers to a wide range of policies, so defining terms is helpful. All ban-the-box policies require the affected employer to delay inquiring about previous arrests or convictions, although the length of the delay varies by specific policy. Some policies may require a conditional job offer, while others just require removal from the initial application. The nature of these policies depends, in part, upon the employer enacting it. Ban-the-box policies have been implemented at the state level by state executives and legislatures. Similarly, many cities and counties have chosen to implement similar policies.

Employer coverage also varies with different ban-the-box policies. Ban-the-box has been applied to Federal, state, and local government employers, contractors, and private employers. In September 2020, NELP found that “a total of 35 states have adopted statewide laws or policies applicable to public-sector employment” and estimated that over three-fourths of the U.S. population lived in a jurisdiction that has some version of ban-the-box, illustrating the broad scope of the policy.

However, there is considerable variation across policies. For instance, Arizona’s ban-the-box executive order (Arizona Executive Order 2017-07) applies just to employment in the executive branch, whereas Delaware passed legislation (Delaware House Bill 167 (2014)) that applies to all public employees including the state agencies and political subdivisions, such as cities and counties. Many states have also implemented ban-the-box policies for contractors, as have 36 non-state jurisdictions (including the District of Columbia). In addition, 14 states and 20 localities have applied ban-the-box laws to private employment. The degree to which certain private employers might be exempt varies by state, but policy implementation is broad.

Critics of ban-the-box policies, such as the National Federation of Independent Business (NFIB), see such laws as unnecessarily masking important information that might impact the safety of the business, work projects, and customers. Critics also argue that ban-the-box laws generally make the hiring process more difficult to navigate for both applicants and employers. Some employers, for example, report that they would rather avoid making offers that they will later have to rescind. And employers who hire in multiple jurisdictions may have to navigate many varying ordinances. NFIB’s chief executive has argued that “ban-the-box laws make it harder for employers to talk about a criminal record at a time that is convenient for them, this means that a small-business owner may spend hours, days, or even weeks going through the hiring process only to find a worker is unqualified” and that employers have legitimate reasons for preferring to inquire about arrest and conviction records early in the hiring process (Maurer 2018).

Another critique of ban-the-box policies is that they may encourage more direct racial and ethnic discrimination against Black and Latino job seekers. If employers would rather avoid extending offers to workers with past arrests or convictions but are unable to directly learn this information, then employers may turn to observable proxies for criminal history. While ban-the-box policies may combat racial animus or inaccurate prejudices by encouraging more interpersonal connection with a given applicant, it is unlikely to eliminate other reasons for excluding formerly incarcerated applicants from jobs (e.g., statistical discrimination, legal requirements, or negligent liability concerns). More specifically, if certain traits, such as race, gender, age, education, work experience, residency, or other factors correlate with rates of arrests or convictions, employers may begin to favor groups within those metrics that have lower average criminal justice exposure (Raphael, 2021). Thus, critics argue that while ban-the-box is well-intended, the statistical discrimination that results harms young Black men with limited educational attainment—including those who do not have any arrest or conviction histories (Doleac, 2019). Critics also argue that the harm to this group outweighs the potential benefits from ban-the-box policies. However, given the hiring structures implemented in the Federal sector, such as a formal adjudication process, it is possible that some of these concerns may be mitigated in practice.[17]

Interaction between ban-the-box and Title VII

The Federal sector has implemented ban-the-box policies in different forms over time. Given current data availability, it is difficult to assess how successful these policies have been in increasing Federal employment opportunities for people with arrest or conviction records. As such, it is also too early to evaluate the potential impact of the Fair Chance Act’s expansion of ban-the-box protections to Federal contract workers in December 2021.

While not perfectly analogous, some lessons can be drawn from the EEOC’s experience enforcing Title VII’s protections against the discriminatory use of arrest and conviction records in private and state and local government settings. Ban-the-box policies may influence Title VII enforcement in a variety of ways:

  1. Hypothesis 1: When applicants are not asked about arrest or conviction records until later in the hiring process, they may be better able to identify whether this information prevented them from being hired. In contrast, when employers receive applicants’ arrest or conviction history at the same time as the rest of relevant background information and qualifications, applicants may find it more challenging to identify if some other qualification or the disclosure of a past arrest or conviction prevented them from being hired. If this hypothesis is true, there would likely be more Title VII complaints alleging improper use of arrest and conviction records in jurisdictions after enacting ban-the-box laws.[18] The subsequent analysis supports this hypothesis, as complaints increased in jurisdictions after enactment of ban-the-box policies.
  2. Hypothesis 2: Previous research suggests that after implementing a ban-the-box policy, employers may screen out applicants with characteristics that they believe correlate with having an arrest or conviction record (e.g., race or ethnicity), since they are less able to ask about these records directly at earlier stages in the hiring process. This pattern is sometimes referred to as “statistical discrimination.” Statistical discrimination would result in fewer complaints alleging discrimination based on the improper use of arrest or conviction records, while there may be more complaints of discrimination based directly on protected classes. The data is not fully consistent with this hypothesis, as arrest and conviction record complaints increased in jurisdictions after the enactment of ban-the-box laws. This does not imply that this hypothesis is not also at play, but it does suggest that it may be of a smaller magnitude than hypothesis 1. This report does not test for increased hiring complaints overall, but this would be an important area for further study.

Results

Available Data Suggests That People With Prior Arrests or Convictions Are Unlikely to Become Federal Workers

The purpose of this report is to determine the impact of prior arrests and/or convictions on an applicant’s ability to obtain employment with the Federal government.

Methodology

This report uses nationally representative surveys to estimate the number of Federal workers with arrest or conviction records. Specifically, this report used a sample from the Panel Study of Income Dynamics (PSID), which collects information about people’s past behavior over time. This sample consisted of almost 40,000 people and covered the years 2003-2017.[19] The PSID is funded by the National Science Foundation and collects data continuously on employment, income, wealth, expenditures, health, marriage, childbearing, child development, philanthropy, education, and numerous other topics, including exposure to the criminal legal system. A related data source, the Transition to Adulthood Supplement (TAS), was used to expand analysis. The TAS follows all PSID sample children who are entering early adulthood, and who comprise the future focal sample members of Core PSID.

There are some drawbacks to this methodology. For one, the number of people in the PSID with past exposure to the criminal legal system is relatively small (885 people). As such, this report’s estimates may not be representative of the population as a whole. Also, the type of data collected by the PSID makes it hard to accurately count the number of individuals in the survey who may have been incarcerated.[20] Indeed, the present analysis may understate the challenges faced by individuals with exposure to the criminal legal system because data linking them to specific labor market outcomes are lacking and can be unreliable. Efforts are underway to improve data availability, which may increase the accuracy of future research. (See “Policy recommendations to facilitate further research” on page 17).

Findings

Data from the PSID suggests that, from 2003 to 2017, people who were previously incarcerated were less likely to be Federally employed than people without such records. According to the PSID, about 3.3 percent of all respondents reported having been (or currently being) Federally employed. However, only 1.8 percent of respondents who had been previously incarcerated reported being Federally employed. In other words, between 2003 and 2017, respondents with who has been previously incarcerated were about half as likely to be Federally employed compared to those without records (Figure 1). This gap suggests that there were roughly 300,000 fewer Federal applicants with workers than expected during this time period.[21]

Figure 1: Probability of Federal Employment

Source: U.S. Equal Employment Opportunity Commission using data from the University of Michigan, Panel Study of Income Dynamics (PSID).

Respondents with prior records of incarceration may differ from those without records across many dimensions. For a variety of reasons, including systemic disparities in the criminal legal system, PSID respondents with prior records are more likely to be male, Black, and have lower educational attainment. If the probability of Federal employment varies across these factors, it is possible that these factors (rather than differences in incarceration records) explain the different hiring rates between the two groups (a phenomena sometimes referred to as omitted variable bias). This study uses linear regression to isolate the correlation between having a record of incarceration and Federal employment. A regression calculates the association between the outcome (dependent) variable and each explanatory (independent) variable, when all other explanatory variables are held constant. The results of a regression include coefficients for each explanatory variable that quantify the magnitude and direction of the relationship between the outcome variable and the corresponding explanatory variable. The results suggest that workers with records of incarceration are less likely to be Federally employed. These results hold regardless of the particular statistical specification.

Column 1 of Table 1 displays the coefficients from a regression that sets Federal employment within the past two years as the dependent variable and an indicator for prior incarceration as the only explanatory variable. Results show that 3.3 percent of surveyed workers without arrest or conviction records were Federally employed. Federal employment for workers with arrest or conviction records was 1.5 percentage points lower. The coefficient on arrest and conviction history was statistically significant with a p-value less than .001, meaning that it is very unlikely (at the 0.1 percent significance level) to be purely by chance. This means that PSID respondents with prior records of incarceration were almost 50 percent less likely to be Federally employed compared to similarly situated respondents without such records.

Column 2 accounts for whether or not the respondent is currently working. Currently employed respondents were 1.4 percentage points more likely to have Federal jobs within the past two years. Workers with a record of incarceration were less likely to be working overall (and working at all is a precondition for working Federally). Including whether the respondent is currently working allows for a comparison between respondents who are similarly attached to the labor force. After controlling for whether or not a respondent was currently working, the respondents with records were still 1.3 percentage points less likely to be Federally employed. This figure is statistically significant. This means that PSID respondents with records of incarceration were almost 30 percent less likely to be Federally employed compared to similarly situated respondents without such records.

Column 3 controls for an array of additional factors measured in the PSID. Column 3 effectively compares individuals who are similar in their current work status; years of education; age; residency in states with similar unemployment and wage rates; survey response year; and are of the same race, sex, and state of residence. After controlling for all these factors, respondents with records of incarceration were 1.5 percentage points less likely to have been Federally employed. This means that PSID respondents with records of incarceration were almost 50 percent less likely to be Federally employed compared to similarly situated respondents without such records.

Column 4 looks at changes in individuals over time and at Federal employment before and after a given individual was incarcerated. After a respondent developed an record of incarceration, they were found to be 1.9 percentage points less likely to be Federally employed.

Table 1: Probability of Federal Employment in the PSID
 

Column 1: Past 2 Years

Column 2: Working

Column 3: Comparators

Column 4: Before and After Contact

Incarceration Record

-0.0148***

-0.0132***

-0.0150***

-0.0191***

Currently working

 

0.0142***

0.00537***

0.00654***

Constant

0.0331***

0.0232***

-0.798*

-0.189*

Person FE

No

No

No

Yes

State FE

No

No

Yes

No

Year FE

No

No

Yes

No

Other controls

No

No

Yes

Yes

* p <0.05, **p<0.01, ***p<0.001.

Notes: FE indicates fixed effects were included. State fixed effects control for the average Federal employment within a state (e.g., if a higher proportion of workers are Federally employed in Virginia than Montana, state fixed effects would account for this). Year fixed effects control for the average Federal employment in a given year. The constant shows the average probability of Federal employment in the sample after netting out all other controls. The p-value measures the probability of seeing a result as extreme as the one observed from a random sample in which the variable had no effect. When the p-value is less than .05, a result is generally labelled statistically significant.

Source: U.S. Equal Employment Opportunity Commission using data from the University of Michigan, Panel Study of Income Dynamics (PSID).

This analysis suggests that, from 2003 to 2017, workers with records faced significant barriers to Federal employment (both in absolute terms and relative to other forms of employment). What this analysis cannot assess is the underlying mechanisms that give rise to these barriers. For instance, it is possible that people with records erroneously believe they are barred from Federal employment and self-select out of the applicant pool. It is also possible that hiring managers are less likely to hire applicants with any kind of incarceration, conviction, or arrest record. More research directly examining the hiring practices of Federal agencies towards workers with records is necessary to better understand how their employment could be increased.

Policy recommendations to facilitate further research

A high-value policy change would be to track the number of Federal workers with criminal histories over time. One potential way to measure this would be to examine the share of Federal employees that had a background check return arrest or conviction information as a proportion of the current Federal workforce. Tracking this statistic over time would allow the EEOC and other Federal agencies to assess how successful different policies are in expanding employment opportunities to formerly incarcerated workers and those with prior arrests or convictions.

Better data, collected and aggregated from existing Federal sources, should be studied in the future. More data on applicants and new hires would allow for a more accurate assessment of the barriers facing formerly incarcerated people and those with prior arrests or convictions. A larger sample size would allow for additional study of what Federal jobs are correlated with the highest barriers, whether the workers are being screened out after applying or are not applying at all, and how a prior arrest or conviction might impact different protected classes.

Ban-the-box policies enabled more and more meritorious EEOC complaints.

It is still too early to measure policies aimed at boosting Federal employment for formerly incarcerated workers and those with prior arrests or convictions. However, by analyzing data from jurisdictions that have enacted ban-the-box policies, this report found that such policies correlate to more and more meritorious EEOC complaints.

Methodology

To evaluate the impact of ban-the-box policies, employment of formerly incarcerated persons and those with prior arrests or convictions, and Title VII compliance, this report assesses the number and composition of Title VII complaints filed with the EEOC from 2008 to 2021. This report also compares jurisdictions that adopted ban-the-box policies to those that did not. The data includes 2,181 unique complaints. Data also include all complaints from individuals with an arrest or conviction recorded in the Strategic Enforcement Plan (#1 Barriers to Recruitment and Hiring) and other complaints that flag the improper use of arrest or conviction records as a factor in the complaint. Due to the structure of the enforcement mechanisms, complaints of this nature are not generally recorded in the Federal sector. However, complaints in other settings are informative. EEOC researchers combined this complaint data with data on ban-the-box rules compiled by the National Employment Law Project (NELP) for analysis (Avery & Lu, 2020).

To understand the impact of ban-the-box policies on the number of EEOC complaints filed, this report compares metropolitan statistical areas (metro areas) that adopted a ban-the-box rule to areas that have not yet adopted such a rule. For this report, a metro area is deemed covered by a ban-the-box rule if any area within the metro area has implemented a ban-the-box rule. Since simple correlation analysis is unlikely to estimate the causal effect of ban-the-box policies, this report uses a differences-in-differences statistical analysis. Difference-in-difference relies on the assumption that areas are following similar patterns of complaints before a ban-the-box policy is implemented and this pattern would continue in the counterfactual scenario in which no ban-the-box policy had been implemented. Areas without ban-the-box laws are used to estimate the number of complaints that would have been filed in the counterfactual scenario that no ban-the-box policy was implemented. That is, this assumes that the ban-the-box metro areas would have had a similar evolution of filed complaints to non-ban-the-box metro areas, except for the passage of ban-the-box rules. While this assumption cannot be formally tested, if treated and non-treated areas had similar complaint trends prior to implementation of ban-the-box, there is evidence that difference-in-difference analysis should yield the causal impact of ban-the-box rules on complaints alleging improper use of arrest or conviction records.[22] In this case, metro areas with and without ban-the-box policies were on similar paths prior to implementation, suggesting the causal identification assumption is reasonable in these situations.

Findings

The number of complaints increased beginning in 2012, the same year when the most recent EEOC policy guidance on the consideration of arrest and conviction records in employment decisions was issued. As shown by the area shaded in red in Figure 1, the number of complaints peaked in 2014, and has since been in decline. The area shaded in blue plots the number of complaints filed in a given year that had been deemed “for cause” by July 2021. There were fewer complaints deemed for cause, with the total correlating highly with the number of complaints filed. Only one year (2014) had for cause complaints exceeding 50.

Figure 2: EEOC Complaints Filed Alleging Misuse of Criminal History in State/Local Ban-the-Box Jurisdictions, 2008-21

Figure 1 displays the number of complaints broken out between for cause and all complaints from 2008 to 2021. These complaints increase after 2011, peaking over 300 complaints per year and declining since 2014.

Note: “For cause complaints” shows the number of complaints filed in a given year that were deemed for cause, meaning the EEOC found reasonable cause to believe that unlawful discrimination occurred.

Source: U.S. Equal Employment Opportunity Commission (EEOC) using EEOC private sector complaint data and ban-the-box rules compiled by the National Employment Law Project (NELP). No public sector data was available at the time this study was completed.

Table 2 below shows that, following the implementation of ban-the-box rules for state and local public employers, more workers filed complaints, the EEOC deemed more of these complaints “for cause,” and this number represented a greater percentage of for cause complaints. Ban-the-box rules aimed at private employers were rarer and do not have a statistically significant impact on the number of complaints (although there is some weak evidence that more complaints are filed after these laws are implemented). Each regression in the table has the same set of independent variables: whether or not a metro area has adopted a ban-the-box rule for public employers, whether the metro area has a ban-the-box policy for private employers, an indicator for each state, and an indicator for the year and month (e.g., July 2021).[23]

Column 1 of Table 2 suggests that ban-the-box rules focused on public employers resulted in seven more complaints for every 100 (i.e., 0.07, p<.01) complaints filed in a metro area per month. By comparison, ban-the-box rules focused on private employers led to four more complaints (i.e., 0.04) per month for every 100 complaints, which is not a significant increase. Column 2 repeats the analysis with for cause complaints. Here, ban-the-box laws focused on public employers generate one more complaint for every 100 (i.e., 0.01, p<.01) additional for cause complaints per month. There is additional evidence that three percent more complaints were deemed for cause after a ban-the-box policy is implemented, as Column 3 shows.[24]

Table 2: Impact of Ban-the-Box (BTB) Laws on Complaints Against State/Local Employers Alleging Misuse of Criminal History
 

Column 1: Complaints

Column 2: For Cause

Column 3: % For Cause

Public BTB rule

0.072**

0.013***

0.028*

Private BTB rule

0.035

0.001

-0.0002

State FE

Yes

Yes

Yes

Year-month FE

Yes

Yes

Yes

* p <0.05, **p<0.01, ***p<0.001.

Notes: FE indicates fixed effects were included. Fixed effects control for the average of the dependent variable within the state and/or year-month. The p-value measures the probability of seeing a result as extreme as the one observed from a random sample in which the variable had no effect. When the p-value is less than .05, a result is generally labelled statistically significant.

Source: U.S. Equal Employment Opportunity Commission.

EEOC researchers found similar results with more sophisticated statistical tools. For instance, using a more robust statistical estimator proposed by Callaway and Sant’Anna (2020), the results are qualitatively unchanged with public employer ban-the-box laws generating 64 more complaints per month for every 1000 complaints (i.e., 0.064, p<.01; SE = 0.012), which is statistically significant.[25] This estimator can also be used to trace out the impact of ban-the-box rules over time. Figure 2 shows the estimated impact in time relative to the implementation of a ban-the-box rule. The x-axis shows the number of months prior to or after ban-the-box enactment, while the y-axis shows the impact of ban-the-box on complaints related to the use of arrest and conviction records. Red dots indicate the estimates prior to the implementation of a ban-the-box policy, while the blue dots trace out the differences after a ban-the-box policy is in place. At the time of our analyses, areas where ban-the-box legislation had been passed but had not yet gone into effect and areas that had not passed a ban-the-box law have a similar number of complaints prior to ban-the-box implementation. After implementing ban-the-box, the number of complaints increases in covered jurisdictions relative to uncovered jurisdictions, steadily increasing up to 10 years after ban-the-box enactment, when the effect starts to taper out (although relatively few areas have had a ban-the-box policy in place for more than 10 years).

Figure 3: Average Impact of State/Local Ban-the-Box (BTB) Laws on Complaints Alleging Misuse of Criminal History Before or After BTB Implementation

This figure plots the average effect of BTB on EEOC criminal history complaints by time since BTB was implemented. The estimated effect prior to BTB is scattered around 0, while after it increases to .15 at about 75 months after BTB adoption, before declining somewhat.

Notes: Red dots indicate the estimates prior to the implementation of a ban-the-box policy. Blue dots trace out the differences after ban-the-box is in place.

Source: U.S. Equal Employment Opportunity Commission.

Without additional data and analysis, it is difficult to say for certain the exact mechanism driving these results. It is possible that, after ban-the-box, employers engaged in more discrimination based on arrest or conviction records. However, it is also possible that the applicants were more aware of their protections after ban-the-box was implemented.

Conclusions, Recommendations, and Next Steps

Conclusions

This report examined Federal employment for people with arrest and conviction records and assessed state and local ban-the-box policies. These policies generally control when in the hiring process an employer can ask an applicant about prior arrests or conviction. This may promote or frustrate the ability of formerly incarcerated people and those with prior arrests or convictions to work in public sector jobs. Using data from a nationally representative survey and EEOC complaint data, EEOC researchers found the following main results:

  • Between 2003 and 2017, there was a significant shortfall in the number of workers with arrest or conviction records in Federal employment.
  • Delaying inquiry into arrest and conviction records until later in the recruiting process enables more active Title VII enforcement.
  • Additional research and data are necessary to assess policies that facilitate Federal employment for formerly incarcerated workers and those with prior arrests and convictions.

Recommendations

A lack of available data constrains the ability to measure the relative rate of Federal employment for formerly incarcerated workers and those with prior arrests and convictions. As such, one high-value policy would be to track the number and performance of Federal workers with arrest/conviction records and flagged credit issues (e.g. the intentional disregard of a just debt) over time. At present, DCSA does not collect or maintain adjudication statistics on the total Federal workforce; instead, it collects data on applicants and employees agencies initiate for investigations. Specifically, if DCSA could generate the proportion of Federal employees that had a background check return arrest or conviction information compared to the total current Federal workforce, these statistics would be helpful to the EEOC. In order to accomplish the goal of increasing Federal employment opportunities for individuals with an arrest or conviction record, it is important to measure trends in the number of Federal employees with arrest or conviction records. Unfortunately, this study was unable to assess this directly due to data limitations.

Future research should also collect, aggregate and study better data from existing Federal sources to update the initial results presented in this study. For example, using data on applicants and new hires allows for more accurate assessment of the magnitude of barriers people with prior arrests or convictions face. A larger sample size would allow for a more accurate examination of which Federal jobs correlate with the highest barriers, whether individuals with an arrest or conviction record are being screened out after applying or are not applying at all, and how arrest and conviction records might impact different protected classes of applicants.

The answers to these questions have important policy implications. Tracking these measures over time would allow the EEOC and other Federal agencies to assess how successful different policies are in expanding Federal employment opportunities for individuals with an arrest or conviction record. For instance, it is possible that people with arrest or conviction records erroneously believe they are barred from Federal employment and self-select out of the applicant pool. If this is the case, Federal agencies could conduct additional outreach to correct this perception. It is also possible that hiring managers are less likely to hire applicants with arrest or conviction records. If this is the case, Federal agencies may need to provide training and additional guidance for hiring managers. Agencies need additional research directly examining the hiring practices of Federal agencies regarding individuals with arrest or conviction records to better understand how employment rates could be improved for these community members.

Current rules and regulations restrict when Federal employers can inquire into arrest or conviction history until after a conditional offer has been made. The analysis presented in this study suggests that such a policy can be helpful in enforcing Title VII and should be continued. The expansion of these recruitment rules to Federal contractors at the end of 2021 may improve Title VII monitoring and enforcement. It is important that recruiters follow these rules and regulations, and follow-up research should be conducted to guarantee Federal employers are following best practices. To ensure robust adherence to best practices, stakeholders should also make applicants aware of not only their right to protect the accuracy of background checks, but also the improper and appropriate use of arrest and conviction records in employment decisions.

Further research should use the techniques proposed in this study to analyze race-based hiring complaints around the adoption of ban-the-box policies to assess whether there is a risk of hiring managers engaging in statistical discrimination when access to arrest or conviction information is restricted. Further research should also survey hiring officials at a sample of Federal agencies to better understand the particular processes that evaluate applicants with arrest/conviction records or flagged credit scores. This survey data could be combined with the hiring statistics derived from NBIB and OPM data to evaluate best practices in lowering barriers to Federal employment for formerly incarcerated persons, in addition to persons with prior arrests or convictions.

Next Steps

OFO should examine of the experiences of Federal employees with past arrest or conviction records will help identify and eliminate barriers to employment for formerly incarcerated persons and for people with arrest or conviction records. OFO should:

  1. Extend the current analysis of how ban-the-box policies impact filed complaints to include an analysis of all race-based hiring complaints to address concerns about statistical discrimination.
  2. Generate and analyze data from the DCSA to measure the number of currently employed Federal workers with arrest/conviction records and flagged credit histories.
  3. Assess what data is available to analyze a similar statistic for formerly incarcerated people.
  4. Evaluate whether workers flagged in the suitability process are more likely to be hired into particular jobs or agencies, and measure job tenure and reasons for job separation.
  5. Conduct a survey of Federal agencies to evaluate what procedures are practiced for applicants and workers with arrest/conviction records or flagged credit histories.
  6. Conduct a focus group to understand how Federal agencies and contractors are planning to implement the Fair Chance Act’s ban-the-box requirements.
 

References

Abraham, K. G., & Kearney, M. S. (2020). Explaining the decline in the US employment-to-population ratio: A review of the evidence. Journal of Economic Literature, 58(3), 585-643.

Agan, A., & Starr, S. (2018). Ban the box, criminal records, and racial discrimination: A field experiment. The Quarterly Journal of Economics, 133(1), 191-235.

Archuleta, K. (2014, Oct. 14). Governmentwide Guidance to Ensure Fair Employment Opportunities for Applicants Who Are Unemployed or Facing Financial Difficulty Through No Fault of Their Own [Memorandum]. Office of Personnel Management. https://www.chcoc.gov/content/governmentwide-guidance-ensure-fair-employment-opportunities-applicants-who-are-unemployed

Avery B. & Lu H. (2020, September 30). Ban the Box: U.S. Cities, Counties, and States Adopt Fair Hiring Policies. National Employment Law Project. https://www.nelp.org/publication/ban-the-box-fair-chance-hiring-state-and-local-guide

Bartik, A. W., & Nelson, S. T. (2018). Credit reports as resumes: The incidence of pre-employment credit screening. Massachusetts Institute of Technology, Department of Economics.

Bronson, J. Bureau of Justice Statistics. (2019). Prisoners in 2017 (Report No. NCJ 252156). Office of Justice Programs. https://bjs.ojp.gov/content/pub/pdf/p17.pdf

Bur, J. (2019, February 4). The government is whittling down its security clearance backlog. Federal Times. https://www.Federaltimes.com/management/2019/02/04/nbib-cuts-down-clearance-backlog-as-dod-transfer-nears

Callaway, B., & Sant’Anna, P. H. (2020). Difference-in-differences with multiple time periods. Journal of Econometrics.

Couloute, L. & Kopf, D. (2018, July). Out of prison & out of work: Unemployment among formerly incarcerated people. Prison Policy Initiative. Retrieved from https://www.prisonpolicy.org/reports/outofwork.html.

Decker, S. H., Ortiz, N., Spohn, C., & Hedberg, E. (2015). Criminal stigma, race, and ethnicity: The consequences of imprisonment for employment. Journal of Criminal Justice, 43(2), 108-121.

Doleac, J. L. (2019). Empirical Evidence on the effects of Ban the Box policies: The state of the literature in 2019. Testimony prepared for the US House Committee on Oversight and Government Reform. Working paper. http://jenniferdoleac.com/wpcontent/uploads/2019/03/Doleac_testimony_BTB_2019.pdf

Dobbie, W., Goldsmith‐Pinkham, P., Mahoney, N., & Song, J. (2020). Bad credit, no problem? Credit and labor market consequences of bad credit reports. The Journal of Finance, 75(5), 2377-2419.

Equal Employment Opportunity Commission (2012, April 25). Enforcement Guidance on the Consideration of Arrest and Conviction Records in Employment Decisions under Title VII of the Civil Rights Act. Title VII, 29 CFR Part 1601, 29 CFR Part 1606, 29 CFR Part 1607. https://www.eeoc.gov/laws/guidance/enforcement-guidance-consideration-arrest-and-conviction-records-employment-decisions

Equal Employment Opportunity Commission (2017). Strategic enforcement plan: Fiscal years 2017-2021. https://www.eeoc.gov/us-equal-employment-opportunity-commission-strategic-enforcement-plan-fiscal-years-2017-2021

Executive Order No. 14035, 86 Fed. Reg. 34593 (2021, June 5). Executive Order on Diversity, Equity, Inclusion, and Accessibility in the Federal Workforce. The White House.

Government Accountability Office (2018, January 5). GAO Adds Government-wide Personnel Security Clearance Process to “High Risk List.” https://www.gao.gov/press-release/gao-adds-government-wide-personnel-security-clearance-process-high-risk-list

Holzer, H. J., Raphael, S., & Stoll, M. A. (2004). How willing are employers to hire ex-offenders. Focus, 23(2), 40-43.

Holzer, H. J., Raphael, S., & Stoll, M. A. (2007). The effect of an applicant’s criminal history on employer hiring decisions and screening practices: Evidence from Los Angeles. Barriers to reentry, 4(15), 117-150.

Kling, J. R. (2006). Incarceration length, employment, and earnings. American Economic Review, 96(3), 863-876.

Leasure, P. and Stevens Andersen, T.. (2017). Recognizing redemption: Old criminal records and employment outcomes. NYU Review of Law and Social Change, The Harbinger, 41: 271-286.

Leasure, P. (2019). Misdemeanor records and employment outcomes: An experimental study. Crime & Delinquency, 65(13), 1850-1872.

Office of Personnel Management (2016, December 1). Recruitment, Selection, and Placement (General) and Suitability. 81 Fed. Reg. 86555. https://www.Federalregister.gov/
documents/2016/12/01/2016-28782/recruitment-selection-and-placement-general-and-suitability

Office of Personnel Management (2019, October). Fillable Optional Form 306 with Instructions. https://www.opm.gov/forms/pdf_fill/of0306.pdf

‌Maurer, R. (2018). ‘Ban the Box’ Turns 20: What Employers Need to Know. SHRM, SHRM, November 12. www.shrm.org/resourcesandtools/hr-topics/talent-acquisition/pages/ban-the-box-turns-20-what-employers-need-to-know.aspx

Mueller-Smith, M., & T. Schnepel, K. (2021). Diversion in the criminal justice system. The Review of Economic Studies, 88(2), 883-936.

Nancy, M., Vigne, L., Lynch, M., & Reimal, E. (2017). Criminal Background Checks Impact on Employment and Recidivism. http://www.urban.org/sites/default/files/publication/88621/
criminal-background-checks-impact-on-employment-and-recidivism.pdf

NELP News. (2015, Nov. 2). On the President’s Announcement on ‘Ban the Box’ Hiring, https://www.nelp.org/news-releases/on-the-presidents-announcement-on-ban-the-box-hiring

Pager, D. (2003). The mark of a criminal record. American journal of sociology, 108(5), 937-975.

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

Pager, D., Bonikowski, B., & Western, B. (2009). Discrimination in a low-wage labor market: A field experiment. American sociological review, 74(5), 777-799.

Paulk, P. (2016). The Johns Hopkins Hospital Success in Hiring Ex-Offenders. Johns Hopkins Medicine. https://www.bgcheckinfo.org/sites/default/files/public/5thMtg_1-0c-Plenary_Pamela_Paulk_Presentation.pdf

Raphael, S., & Weiman, D. F. (2007). The impact of local labor market conditions on the likelihood that parolees are returned to custody. Barriers to reentry? The labor market for released prisoners in post-industrial America, 304-332.

Raphael, S. (2021). The intended and unintended consequences of Ban the Box. Annual Review of Criminology, 4, 191-207.‌

Redcross, C., Millenky, M., Rudd, T., & Levshin, V. (2011). More than a job: Final results from the evaluation of the Center for Employment Opportunities (CEO) transitional jobs program. OPRE Report, 18.

Schnepel, K. T. (2018). Good jobs and recidivism. The Economic Journal, 128(608), 447-469.

Shannon, S. K., Uggen, C., Schnittker, J., Thompson, M., Wakefield, S., & Massoglia, M. (2017). The growth, scope, and spatial distribution of people with felony records in the United States, 1948–2010. Demography, 54(5), 1795-1818.

Society for Human Resource Management (SHRM) (2021, May). Getting Talent back to Work: A Workplace Survey on Hiring and Working with People with Criminal Records.

The Sentencing Project. (2018, April 19). Report to the United Nations on Racial Disparities in the U.S. Criminal Justice System. https://www.sentencingproject.org/publications/un-report-on-racial-disparities

Yang, C. S. (2017). Local labor markets and criminal recidivism. Journal of Public Economics, 147, 16-29.

 

[1] The central reciprocity system managed by DCSA does not capture data on the status of every individual investigated (i.e., applicant, Federal employee, contractor). It also does not currently capture affiliation across all populations (e.g., is the applicant hired, when a Federal employee leaves service and in some cases, why they leave service). It's possible this type of data may be available in a successor system.

[2] OPM is in the process of implementing the Fair Chance to Compete for Jobs Act which will change placement of this requirement with respect to the timing of collection of criminal history information within the regulations. See Notice of Proposed Rulemaking: https://www.federalregister.gov/documents/2022/04/27/2022-08975/fair-chance-to-compete-for-jobs.

[3] The Fair Chance to Compete for Jobs Act of 2019 expanded the populations covered, including contractors. However, it still be implemented through rule changes in the Code of Federal Regulations. https://www.govinfo.gov/link/plaw/116/public/92.

[4] Some examples of this evidence include Pager (2003, 2007), Holzer et al. (2007), Pager et al. (2009), Decker (2015), Agan & Starr (2017), Leasure & Anderson (2017), and Leasure (2019).

[5] For additional details and estimates see Kling (2006), Raphael & Weiman (2007), and Redcross et al. (2011).

[6] EEOC Decision No. 72-1497 (1972) (challenged an arrest or conviction record exclusion policy based on “serious crimes”); EEOC Decision No. 74-89 (1974) (challenged a policy where a felony conviction was considered an adverse factor that would lead to disqualification); EEOC Decision No. 78-03 (1977) (challenged an exclusion policy based on felony or misdemeanor convictions involving moral turpitude or the use of drugs); EEOC Decision No. 78-35 (1978) (concluded that an employee’s discharge was reasonable given his pattern of criminal behavior and the severity and recentness of his arrest and conviction).

[7] EEOC Policy Statement on the Issue of Conviction Records Under Title VII of the Civil Rights Act of 1964 (Feb. 4, 1987; EEOC Policy Statement on the Use of Statistics in Charges Involving the Exclusion of Individuals with Conviction Records from Employment (July 29, 1987; EEOC Policy Guidance on the Consideration of Arrest Records in Employment Decisions Under Title VII (Sept. 7, 1990); EEOC Compliance Manual Section 15: Race & Color Discrimination § 15-VI.B.2 (April 19, 2006), https://www.eeoc.gov/policy/docs/race-color.pdf.

[8] Consistent with Texas v. EEOC, 933 F.3d 433, 451 (5th Cir. 2019), the EEOC may not treat its 2012 Arrest/Conviction Guidance on the consideration of arrest or conviction records in employment decisions as binding in any respect against the State of Texas.

[9] Note that some courts have required plaintiffs in these cases must have a disparate impact on the relevant hiring pool beyond national statistics. Mandala v. NTT Data, Inc., No. 19-2308 (2d Cir. 2020). https://law.justia.com/cases/federal/appellate-courts/ca2/19-2308/19-2308-2020-09-21.html.

[10] 401 U.S. at 431, 436.

[11] See also, e.g., 514.37 of USPS’s Handbook EL-312, Employment and Placement: https://about.usps.com/handbooks/el312.pdf.

[12] Additional details regarding the suitability adjudication process are available at this OPM explainer: https://www.youtube.com/watch?v=dOsLSw1AMB8. The discussion on debt begins at about 13:30.

[13] For additional information about the transfer of background investigations from the Office of Personnel Management to the Department of Defense, see https://www.opm.gov/news/releases/2019/10/opm-finalizes-transfer-of-background-investigations-to-defense-department.

[14] See Subpart M-Timing of Background Investigations § 330.130.

[15] OPM will consider factors such as the nature of the position being filled and whether a clean arrest and conviction record is essential to the ability to perform one of the duties of the position effectively. OPM may also consider positions for which the expense of completing the examination makes it appropriate to adjudicate suitability at the outset of the process (e.g., a position that requires that an applicant complete a rigorous training regimen and pass an examination based upon the training before selection can be finalized). A hiring agency must request and receive an OPM-approved exception before publicly posting the position in question. 

[16] The Fair Chance Act took effect on December 20, 2021. As of the completion of this report, regulations implementing the Fair Chance to Compete for Jobs Act of 2019 had not been issued. However, on August 31, 2023, OPM issued a final rule to the Act. https://www.opm.gov/news/releases/2023/08/release-office-of-personnel-management-issues-final-regulations-to-prohibit-federal-agencies-from-requesting-criminal-history-during-hiring.

[17] Note that statistical discrimination may well be a violation of Title VII as well, as Title VII prohibits “not only decisions driven by racial [or ethnic] animosity, but also decisions infected by stereotyped thinking…Thus, an employer’s decision to reject a job applicant based on racial or ethnic stereotypes about criminality—rather than qualifications and suitability for the position—is unlawful disparate treatment that violates Title VII.” https://www.eeoc.gov/laws/guidance/section-15-race-and-color-discrimination

[18] For an example of facts underlying one recent complaint, see Ramos v. Walmart, Inc., No. 21-cv-13827 (D.C.N.J., July 19, 2021).

[19] More information concerning the PSID and related collections may be found at https://psidonline.isr.umich.edu/.

[20] Constructing a measure of individuals in the PSID with past exposure to the criminal legal system is problematic. This study uses the following construction for generating prior arrest or conviction: If the reason an individual is a non-respondent in a given year is that they are in jail or prison, they are marked as having an arrest/conviction starting in that year. This data was combined with survey questions from 1995 and the Transition to Adulthood Survey that asked respondents about arrest or conviction records directly. Because this data is self-reported, it is possible that the data did not fully capture all individuals with an arrest or conviction. It is most likely that measurement error in this setting will result in an analysis that understates the challenges formerly incarcerated persons face in becoming Federally employed.

[21] This calculation assumes that there are 20 million people with arrest or conviction records. Research suggests that this might be an underestimate given estimates from Shannon et al. (2017) that there were about 20 million people with felony convictions in 2010.

[22] The following results are estimated using a two-way fixed effects regression: yit=β*BTBpublic.i.t+BTBprivate.i.t+θi+ϕt+εi.t. The dependent variable of interest is y, i indexes metro areas, and t indexes time at the year-month frequency.

[23] Robustness checks were also implemented. Including metro-area fixed effects instead of state fixed effects yields a qualitatively similar result, but with smaller effect sizes. Controlling for time also yielded similar results.

[24] Analysis on final outcomes of the complaint (e.g., settlement amount) are not included as there are relatively few observations in the studied time frame for which data is fully available. In addition, not all complaints had been fully resolved.

[25] This estimate is the average treatment on the treated using a calendar period aggregation. An estimate of 0.052 more complaints per month (standard error of 0.0136) is obtained using treatment time aggregation.

Enabled In-page Navigation