Credit Screenings Lead to Unfair Hiring

Checking up on a job applicant’s financial history has become a common practice in hiring — even for service industry jobs like serving frozen yogurt or driving a delivery truck.

Employers might assume that a job candidate’s financial history provides a real-world measure of their trustworthiness and reliability. However, new research contends that screening candidates based on credit checks does little to ensure quality hires, and instead often leads to discrimination.

The analysis, from University of New Mexico psychological scientist Sabrina Volpone and her colleagues, suggests that basing job selection decisions off of credit scores may unfairly—and perhaps even illegally—disadvantage black applicants.

Credit reports are designed to allow banks and other financial lenders to assess someone’s history of paying their debts—student loans, credit card bills, mortgage payments—before offering terms for a loan. Even when income, education, and marital status are controlled for, research has shown that racial minorities tend to have lower credit scores than whites, suggesting that racial bias is at play.

Use of this kind of financial background check has become increasingly pervasive in hiring at all levels, and not just for high-level positions. According to a 2012 survey by Society for Human Resource Management (SHRM), nearly half of employers (47%) use these types of reports to make hiring decisions. Several states have already passed laws prohibiting employers from denying jobs to applicants based on their credit history.

To examine effects of credit scores on employment decisions, Volpone along with APS fellow Scott Tonidandel (Davidson College), Derek Avery (Temple University), and Safiya Castel (University of California, Los Angeles) conducted a simulation study. They used existing records to construct a hypothetical (i.e., simulated) sample of 10,000 individuals, some black and some white. The research team then conducted a Monte Carlo simulation—a form of sophisticated statistical analysis used to forecast a range of potential outcomes.

For the simulation, they included several factors typically used in hiring (i.e., educational background, personality). Reflecting real averages, the credit scores of black individuals in the simulation were lower on average than the credit score of white individuals. They also simulated different sample sizes, tried different combinations of hiring criteria, and used different measures to estimate the influence of credit scores on hiring outcomes.

Regardless of all other hiring factors, the researchers found that making decisions based on credit scores disproportionately affected black job applicants over white ones.

The authors recommend against the use of credit scores, stating, “Including credit scores as a predictor, even when utilized during a final stage of the hiring process, produces adverse impact in a majority of cases.”

Another study, led by Laura Koppes Bryan (University of West Florida) and Jerry Palmer (Eastern Kentucky University), analyzed credit reports from 178 real employees working at a large financial services organization. They found that an employee’s credit history had no relationship to their performance ratings or likelihood of misconduct citations at work.

“While intuitively a credit history is measuring responsibility and the ability to meet deadlines, in practice this measure is probably contaminated by many other factors outside an employee’s control, as well as inaccuracies,” Bryan and Palmer write. “Thus, our data indicate there is no benefit from using credit history to predict employee performance or turnover.”

References 

Koppes Bryan, L., & Palmer, J. K. (2012). Do Job Applicant Credit Histories Predict Performance Appraisal Ratings or Termination Decisions? The Psychologist-Manager Journal, 15(2), 106-127. doi: 10.1080/10887156.2012.676883

Volpone, S. D., Tonidandel, S., Avery, D. R., & Castel, S. (2015). Exploring the use of credit scores in selection processes: Beware of adverse impact. Journal of Business and Psychology, 30, 357-372. doi: 10.1007/s10869-014-9366-5