Should policy be judged on its intentions or its outcomes?

The economist Milton Friedman once said that “One of the great mistakes is to judge policies and programs by their intentions rather than their results.”

He was right. For example, those who argue in favor of higher minimum wages do not want to see fewer jobs available to low-skill workers. But the balance of empirical evidence shows that that is exactly what happens. This outcome confronts us with a choice. Do we persist with the minimum wage law because its intentions are noble, or do we junk it because its outcomes are harmful? If the employment opportunities and earnings of low skilled workers matter more to us than the hurt feelings of activists, we should follow the latter course.

So called ‘Ban the box’ laws offer another example. In 2013, Minnesota became the third state to pass this legislation. These laws prevent private employers from asking about criminal history information on job applications and during interviews. Since then these laws have spread so that now 34 states and over 150 cities and counties nationwide have adopted them.

The intentions of these laws are noble. As Minn Post reported of Minnesota’s law, it “was created to give ex-offenders a better chance for employment, something that had long been denied to them as a result of the criminal background checkbox on job applications.”

But what have the effects of such laws been? That is the important question.

On initial inspection, we might see some encouraging news. The Urban Institute writes that

Research on ban the box has shown that it increases callback rates for people with criminal records (Agan and Starr 2016). Agan and Starr (2016) find that ban-the-box policies “effectively eliminate” the effect of having a criminal record on receiving a callback. Case studies from specific cities support these results, showing that hiring rates for people with criminal records increased after ban the box was implemented (Atkinson and Lockwood 2014; Berracasa et al. 2016).

But the majority of the consequences of any action are unintended. It turns out that when employers are stopped by law from asking about criminal history, they don’t disregard it as factor in hiring, instead they look for proxies. The Urban Institute writes that

…recent research has concluded that ban the box also reduces the likelihood that employers call back or hire young black and Latino men (Agan and Starr 2016; Doleac and Hansen 2016). These findings suggest that when information about a person’s criminal history is not present, employers may make hiring decisions based on their perception of the likelihood that the applicant has a criminal history.

In other words, the “increase[d] callback rates for people with criminal records” have been bought at the expense of a “reduce[d]…likelihood that employers call back or hire young black and Latino men”.

This situation is very similar to that with minimum wage laws. Examining the empirical literature of their effects, the economists David Neumark, William L. Wascher, and J.M. Ian Salas concluded that “the evidence still shows that minimum wages pose a tradeoff of higher wages for some against job losses for others, and that policymakers need to bear this tradeoff in mind when making decisions about increasing the minimum wage.”

In both cases, as Friedman used to point out, there is no free lunch, as supporters of these policies all too often argue. Instead, there is a tradeoff. Such tradeoffs, not wishful thinking, are, or ought to be, the essence of public policy.

John Phelan is an economist at the Center of the American Experiment.