Find New Employees in a Tight Market: Second-Chance Staffing Study Results

Although the role of justice-involved individuals in the current labor market is an important issue, much of the previous academic literature has focused on the entry/exit process itself, with hiring and recidivism rates being the primary variables of interest. In contrast, very few studies have tackled the ex-offenders’ experience on the job itself, perhaps due to the difficulties inherent in the data collection process.

The goal of this study, funded by Allegiance Staffing, the International Warehouse Logistics Association (IWLA), and UniteIndy,, was to examine the relative workplace performance of justice-involved citizens and ex-offenders, as well as identify factors affecting this performance.  It examined the employment records of 856 entry-level employees in five central Indiana logistics and manufacturing organizations of varying size, collecting demographic- and criminal-history-related data, as well as performance metrics.  Due to the variety of performance evaluation systems utilized by the partner organizations, entry-level employees were divided into two categories (“Above Average” and “Below Average”) based on the system in place at each firm, thus allowing us to aggregate across organizations.  Finally, employees at each organization completed a culture survey, known as the Competing Values Framework, in order to characterize the degree to which they perceived their workplace environment as being collaborative, competitive, creative, or control-oriented.

A substantial percentage of entry-level employees in our sample were justice-involved individuals (26-67 percent depending on the partner organization). Those with a criminal record had, on average, two misdemeanors and two felonies, with the average number of years between the most recent felony or misdemeanor and the time of hire being approximately 5-6 years. In terms of the nature of conviction, alcohol and drug-related convictions were the most prevalent type of conviction (44 percent), followed by driving violations (18 percent). 

As a first step, comparing ex-offenders to those without a conviction on a single dimension/variable, it found that those with a conviction were:

more likely to be single males compared to those without conviction, although differences existed across organizations,

employed for 8-9 fewer months at both current and previous employers compared to those without a conviction (i.e. less “attached” to employer),

and experienced a 6-7 month shorter period of unemployment between current and previous job compared to those without a conviction (i.e. more “attached” to employment in general)

More importantly, those with a conviction were more likely to fall in the below-average performance category compared to those without a conviction (56 percent vs 43 percent, respectively), but this result disappeared when we controlled for the nature of conviction and employer size.  In particular, of those with a misdemeanor, 67 percent were in the below-average category compared to 44 percent of those with no misdemeanor conviction.  In contrast, it did not find a statistically significant difference in performance between those with a felony and those with no felony conviction. Turning to employer size, at small employers, it found no statistically significant difference in performance between those with a conviction and those without a conviction.  In contrast, at the large employer those with a misdemeanor conviction performed worse than those without such a conviction.

Secondly, it estimated a more complex model of performance in which multiple predictors were controlled for simultaneously (e.g. gender, age, work experience, criminal record). Similar to our univariate results, it found that those with a conviction had the same odds of falling in the below-average performance category as those with no conviction, but this result was linked to the nature of the conviction. In particular, those with a misdemeanor were 49 percent less likely to be rated as “Above Average” than those without a misdemeanor conviction.  In contrast, there was no statistically significant different in performance-related odds between those with a felony and those without a felony conviction. 

Finally, turning to the workplace environment, it found no statistically significant evidence that the performance of ex-offenders is correlated with an organization’s Cooperate, Create, Control, or Compete scores in the Competing Values Framework.  Note, however, that this result was likely due to the relatively small variation in scores across the employers in this sample. As an alternative approach, it simply controlled for each employer in our sample by incorporating employer-related interaction terms. Using this model specification, it observed differences in the performance of ex-offenders across employers, suggesting that some aspect of workplace environment does matter.  In particular, at small and medium partner organizations, no statistically significant difference in performance-related odds was observed between those with a conviction and those without a conviction, regardless of the severity of the conviction. At the large employer, however, those with a misdemeanor were 45 percent less likely to be rated as “Above Average” than those without a misdemeanor conviction. 

Our study suggests the following key takeaways:

Those with a conviction appear to be more attached to the labor force but less attached to their current employer than their peers. In other words, these individuals exhibit behavior consistent with a strong motivation or need to work but face unique obstacles that may make it relatively more difficult to remain employed.

Helping those with a conviction overcome these obstacles may be particularly difficult, since ex-offenders tend to underreport both the number and nature of their offenses. Coupled with incomplete employment and performance-related records, it is likely that many ex-offenders will not be identified as such and, thus, slip through the cracks.

The severity of the conviction is positively correlated with on-the-job performance. Thus, those who struggle most in the workplace are those with misdemeanors, not those with felonies.

This negative relationship between misdemeanor convictions and workplace performance is mitigated in smaller firms. Thus, there is something about the workplace environment of small firms that allows even those with misdemeanors to succeed.

To become an IWLA member visit, http://31y.93c.myftpupload.com/become-an-iwla-member/