Commentary | Economic Growth

A bad answer to the wrong question

With its recent report on The Impact of State Employment Policies on Job Growth, the U.S. Chamber of Commerce (COC) demonstrates yet again that when the only tool somebody has is a hammer, every problem looks like a nail. In this case, the problem is a real one—high unemployment. But the chamber’s hammer—loud condemnation of any public policy that aims to help working families rather than corporations—isn’t going to help solve it.

The chamber has created an employment regulation index (ERI) for each of the 50 states. In the chamber’s view a “perfect” ERI score is a zero, meaning that the state has no regulations stricter than federal minimums. In essence, the chamber is advocating that states surrender all of their policymaking autonomy.

And what are some things that harm (inflate) a state’s ERI? A high minimum wage, more-generous unemployment benefits, requirements that unemployed workers receive their benefits in a timely manner, requirements that employers pay-out employees’ last paychecks in a timely manner, requirements that workers get meal and rest periods during the workday, and prohibitions on employees’ ability to construct work contracts that allow for reasonable collection of union dues (prohibitions that are often misleadingly labeled “right-to-work” laws).

This will strike many as an odd list of things that are bad for a state’s workers, but the chamber claims in their report that they have found a durable statistical relationship between a low ERI and low unemployment. They even go so far as to call the complete surrender of state policymaking autonomy an opportunity for “free economic stimulus.” There are (at least) two problems with these claims over and above the basic one that they ignore the costs to affected workers of removing all state safeguards connected to their employment.

First, the literature cited (often misleadingly) by the chamber examines whether regulations impose costs on employers that lead to higher unemployment. But nobody thinks that today’s high unemployment rates are caused by excessively expensive workers. Economy-wide labor costs were actually lower in 2010 than in 2007, yet the unemployment rate more than doubled between these years.

Second, even if one accepted the chamber’s broad diagnosis that too-expensive workers are the root cause of today’s high unemployment, its claim of a durable statistical relationship between the ERI it has constructed and state unemployment rates is dubious at best. In results reported on in a companion issue brief, I managed to replicate the basic chamber result that a low ERI apparently leads to low unemployment (see the EPI Issue Brief, Why Do Some States Have Higher Unemployment Rates? for more detail). Then, however, I exposed this finding to a few simple questions—does it hold up over most of the time period in the sample? Is the relationship consistent among select group of states? Does the relationship persist if other reasonable controls are added to the model? Each time the answer was no, indicating that the claimed statistical relationship is not robust.

For example, if either 2007 or 2008 is excluded, the chamber’s finding of a relationship between low regulation and low unemployment becomes insignificant. Pull two states—the Dakotas (representing all of 0.48% of the population)—out of the sample and the findings again fall to insignificance. Just adding a control for the share of a state’s workforce engaged in mining yields the same result.

In short, the chamber is asking the wrong questions about today’s high unemployment and is then answering them with poor methods. Its latest report tells us nothing new and is an unneeded distraction from the serious debates we should be having about policies that might actually help to solve the jobs crisis.

See related work on Jobs | Regulation | Economic Growth | Wages, Incomes, and Wealth

See more work by Josh Bivens