The Agonizing Trade-Offs That Must Be Made to Reduce Bias in the Hiring Process

Like many economists, I try to form opinions on the economy based on research and data, rather than anecdotes. And like many economists, I worry that occupational licenses do more harm than good. Obtaining a license requires people to take part in training and pay fees in order to work in certain jobs. In the 1950s, only 5% of American workers were licensed; in 2008, 29% were. Licensing now largely serves the purpose that unions once played among American workers.

Some jobs should require special certification on the grounds of public safety: doctors, teachers, or electricians. But paying $22,000 for a license to braid hair is harder to justify. For some jobs, licensing is needlessly expensive and time consuming. Evidence suggests that occupational licenses boost wages for those that get them, but depress overall job growth by creating an artificial barrier to work. Licenses also bind workers to low-pay regions because moving states often requires getting a new license. The data also suggest excessive licensing regulations do not improve service quality, but instead raise prices for consumers and depress economic output.

I started rethinking my reflexive skepticism of licenses while interviewing former prisoners for a research project. One man proudly listed a few different licenses he earned since his last stint in prison, more than 10 years ago. Time in prison devastates employment prospects because many employers are wary of hiring ex-cons. Studies suggest that this lack of opportunity may contribute to high recidivism rates, setting off a vicious cycle.

The man I met had more success finding work than other ex-cons because he sought jobs that required a license. It seemed that having a license provided a signal to employers that, despite time in prison, he was hard working, reliable, and committed to staying out of prison in the future. The license also meant he faced less competition for jobs with people who hadn’t been in prison.

Thus, licensing appears to address a market failure: Employers won’t hire a productive, able, willing former prisoners because they don’t trust them; Occupational licenses serve as a signal that they should.

Mind the gap

Economic researchers are taught to not read too much into one person’s story. But a recent research paper based on census data reached a similar conclusion. Peter Blair and Bobby Chung of Clemson University noticed that in the US, black men’s wages increase 12.5% in jobs that require a license versus those that don’t. White men’s wages increase too, but by a lesser amount, 7.5%. The result is a 43% reduction in the wage gap between black and white men in licensed occupations when compared with unlicensed jobs. For women, licenses reduced the gender pay gap by 40% for white women and 36% for black women.

The researchers speculate that licenses help equate pay because some employers make discriminatory assumptions about the quality of a worker based on race, gender, or criminal history. A license provides a signal of a high-quality worker.

Some licenses aren’t available to convicted felons. Blair and Chung estimate that the black-versus-white wage gap is even narrower in occupations that require a license that excludes ex-cons. White men’s wages go up by 3.2% if the job requires a non-felony license, while black men’s wages rise by 19.6%. Research suggests that employers don’t hire black men because they assume they are more likely to have spent time in prison. A license that is unavailable to ex-cons serves as a signal that a job candidate hasn’t served time.

If occupational licenses reduce discrimination in this way, they may be more valuable than the conventional economic wisdom suggests. That said, the license system can still be improved. The federal government can establish best practices that states follow in determining what occupations require a license, and what the standards should be. Right now, the rules vary widely across states. There should be a common procedure to transfer occupational licenses across states, similar to drivers licenses.

Blair and Chung’s research also raises the contentious issue of whether restricting licenses to felons is a good idea. On one hand, more restrictions means black men without past convictions face less discrimination when looking for jobs. Alternatively, occupational licenses serve useful role for former prisoners after they have served their punishment and face the daunting task of finding work after incarceration. Further license restrictions would make an already difficult labor market even harsher for them.

Since we don’t live in that world without discrimination, occupational licenses serve to foster a more just labor market. It is absurd that workers who face ingrained discrimination must invest extra time and money to get a more fair chance at a good job. It is even more absurd that we must ask ourselves whether it worth subjecting former inmates to even more hardship in the job market so non-convicts don’t face the same.