‘Redemption’ in an Era of Widespread Criminal Background Checks by Alfred Blumstein and Kiminori Nakamura ne of the stated goals in President The numbers leave no doubt that we have Barack Obama’s crime and law reached a broad penetration of criminal his­ Oenforcement agenda is to break tory records into the fabric of our society: down employment barriers for people who have a prior criminal record, but who have n In 2006, nearly 81 million criminal records stayed clean of further involvement with the were on file in the states, 74 million of criminal justice system. To understand how which were in automated databases.2 many people are affected by some of these n Another 14 million arrests are recorded barriers, we only need look at the wide­ every year.3 spread computerization of criminal history records in the United States. What does this mean for employers? And what does it mean for ex-offenders who According to the Society for Human need a job? Resource Management, more than 80 percent of U.S. employers perform Consider a 40-year-old male who was con­ criminal background checks on prospective victed of burglary when he was 18 years old employees.1 Add two additional factors to and has committed no further crimes. Every that equation — advances in information time he applies for a new job, he tells the technology and growing concerns about potential employer that he was convicted employer liability — and we can begin to of a felony; even if he does not state this understand how complicated the issue of up-front, the employer is likely to do a employing ex-offenders has become. criminal background check. In either case, 10 NIJ JOURNAL / ISSUE NO. 263 he probably will not get the job because many employers are unwilling to hire an There should be some point in time ex-offender.4 after which ex-offenders should not be This situation prompted us to ask the handicapped in finding employment. question: Is it possible to determine empirically when it is no longer necessary The question is when, precisely, for an employer to be concerned about a criminal offense in a prospective should this occur? employee’s past? Most people would probably agree that The National Institute of Justice funded there should be some point in time after our study to “actuarially” estimate a point which ex-offenders should not be handi­ in time when an individual with a criminal capped in finding employment. The ques­ record is at no greater risk of committing tion is when, precisely, should this occur? another crime than other individuals of the In the case of our hypothetical 40-year-old, same age. Although our research is ongoing when should a prospective employer no lon­ — and our findings, discussed in this article, ger consider a burglary that was committed are preliminary — we have created a model more than two decades earlier if the job for providing empirical evidence on when an applicant has stayed clean since then? ex-offender has been clean long enough to be considered, for employment purposes, Currently, employers have no empirical “redeemed.” An in-depth discussion of our guidance on when it might be considered findings and research methods appears in safe to overlook a past criminal record the May 2009 issue of Criminology.6 when hiring an ex-offender for a particular job. Employers generally pick an arbitrary number of years for when the relevance of What We Have Known for Years a criminal record should expire: five or 10 years, for example. It goes without saying It is well known — and widely accepted by that different types of employers will have criminologists and practitioners alike — that recidivism declines steadily with time different sensitivities about the potential 7 employee’s criminal record. Those serving clean. Most detected recidivism occurs within three years of an arrest and almost vulnerable populations like children and the 8 elderly would be particularly sensitive to certainly within five years. But is it possible a prior record involving violence, while a to identify when the risk of recidivism bank hiring a teller would be particularly has declined sufficiently to be considered sensitive to property crimes. A hiring crew irrelevant in hiring decisions? for a construction company might be far less sensitive to most prior records. In our study, we obtained the criminal his­ tory records of 88,000 individuals who were arrested for the first time in New York state The point is that determining when a 9 potential employee’s criminal record may in 1980. First, we determined whether they no longer be relevant has, to date, been an had committed any other crime(s) during arbitrary exercise. Although considerable the ensuing 25 years or if they had stayed research has been done on how to forecast clean. Then we compared this data against possible criminal behavior, no empirical two populations: basis has been found for deciding when (1) People in the general population who a person’s record is stale enough for an 10 employer to consider it no longer useful were the same age. 5 or relevant. (2) People of the same age who had never been arrested. Until now. 11 NIJ JOURNAL / ISSUE NO. 263 We believe that our analysis provides the n Age at the time of the 1980 (first) arrest. n Type of crime. criminal justice community with the first We then compared these hazard rates, as scientific method for estimating how long is they declined over time, to people of the same age in the general population. For “long enough” for someone with a prior record these data, we used the arrest rate (the age-crime curve) from the Uniform Crime to remain arrest-free before he or she Reports, maintained by the Federal Bureau of Investigation. should be considered “redeemed” In the figure on page 13, we show the hazard rate for 18-year-olds when they were by a prospective employer. arrested for a first offense of one of three crimes: robbery, burglary and aggravated assault. The figure shows that for robbery, Our goal was to determine empirically at the hazard rate declined to the same arrest what point in time the risk of recidivism rate for the general population of same- for people in our study group was no greater aged individuals at age 25.7, or 7.7 years than the risk for our two comparison popula- after the 1980 robbery arrest. After that tions.11 To do this, we plotted data curves point, the probability that individuals would to determine when the risk of re-arrest for commit another crime was less than the individuals in our study group: probability of other 26-year-olds in the general population. n Dropped below the risk of arrest for same-aged people in the general The figure also shows our analysis for bur- population. glary and aggravated assault. The hazard n Approached the risk of arrest for people rates of people who committed burglary at who had never been arrested. age 18 declined to the same as the general population somewhat earlier: 3.8 years We believe that our analysis provides post-arrest at age 21.8. For aggravated the criminal justice community with the assault, the hazard rates of our study group first scientific method for estimating how and the general population of same-aged long is “long enough” for someone with individuals occurred 4.3 years post-arrest a prior record to remain arrest-free before or at age 22.3. he or she should be considered “redeemed” by a prospective employer. Individuals who were arrested for robbery at age 18 had to stay clean longer than those Determining the Hazard Rate who were arrested for burglary or aggravat- ed assault to reach the same arrest rate as Our analysis was based on a statistical same-aged people in the general population. concept called the “hazard rate.” The hazard rate is the probability, over time, We also looked at the effect of the that someone who has stayed clean will arrestee’s age at the time of his first be arrested. For a person who has been arrest in 1980. We examined the hazard arrested in the past, the hazard rate rates for three ages of people in our study declines the longer he stays clean. group — 16, 18 and 20 years old — who were arrested for robbery in 1980. Based To determine the hazard rate for our study on the criminal histories of these people, group, we looked at two factors: we found that individuals who were first 12 NIJ JOURNAL / ISSUE NO. 263 Hazard Rate for 18-Year-Olds: First-Time Offenders Compared to General Population The probability of new arrests for offenders declines over the years and eventually becomes as low as the general population. 25% 20% 15% Probability of first arrest for general population or subsequent arrest for previous offenders 4.3 7.7 10% 3.8 General population 5% Robbery Aggravated assault Burglary 0 2 4 6 8 10 12 14 16 18 Years since first arrest arrested when they were 18 years old had were first arrested at age 20 crossed their the same arrest rate 7.7 years later as a curve 4.4 years after their first arrest. same-aged individual in the general popula­ tion. In contrast, those whose first arrest Thus, our analysis showed that the younger occurred at age 16 crossed the curve for a an offender was when he committed same-aged individual in the general popu­ robbery, the longer he had to stay clean lation 8.5 years later, and individuals who to reach the same arrest rate as people 13 NIJ JOURNAL / ISSUE NO.