After the 2022 election, labor unions pushed Michigan Democrats to repeal the state’s right-to-work law. They succeeded, and the Legislature voted in March 2023 to scrap the 10-year-old law. The Mackinac Center attempted to testify before a legislative committee in opposition but was not permitted to do so.

Proponents of the repeal frequently cited a 2015 paper published by the Economic Policy Institute, a Washington, D.C.-based think tank, while the bill was working its way through the Legislature. The paper claimed that average wages were 3% higher in states that did not have right-to-work laws.

The Economic Policy Institute’s analysis was flawed, however. A new Mackinac Center policy brief by economist Christopher Douglas, a professor at the University of Michigan-Flint and an adjunct scholar with the Center, explains why.

Douglas reviews the 2015 paper and finds that it controlled for only two state-level differences: the unemployment rate and the cost of living. Many other factors also influence a state’s average wage. The EPI report failed to control for significant variables, rendering its results unreliable.

Douglas is not the first economist to notice the study’s oversights. Scholars at the Institute for Research on Labor and Employment, a unit of the University of California, Los Angeles, replicated the EPI model but added three additional control variables: state gross domestic product (a measure of economic development), the state’s business tax rate and its minimum wage. Including these very basic economic factors reduced the difference in average wages to just 1.04%, a heady drop from EPI’s 3.2%.

Douglas believes that any analysis of right-to-work’s effects on the average state wage must include more state-level controls. For this he applies a fixed-effects technique to his model, which is explained in his policy brief, “Right-to-Work States Do Not Have Lower Wages.” This technique allows scholars to account for many more variables that could have an influence on the state average, quite apart from a right-to-work law. Those variables might include different historical wages, geography and natural amenities, or differences in industrial makeup, among others.

Douglas cites as one example the fact that Southern states, which are more likely to have a right-to-work law, have on average had lower wages than Northern ones ever since the Civil War. This wage difference existed before Southern states adopted these laws, so this reality needs to be accounted for. The fixed-effects technique helps accomplish that.

Many scholars support this approach, including economist L. Robert Reed. His 2003 Journal of Labor Research article, “How Right-to-Work Laws Effect Wages,” argued that the failure to control for initial economic conditions likely explains why other studies don’t show positive wage impacts from right-to-work laws.

The Economic Policy Institute paper appears to ignore these and other important considerations. When this happens, Douglas says, it “assigns too much weight to the existence of right-to-work laws” by likely “picking up the preexisting condition of lower average wages in Southern states before they passed right-to-work laws

Douglas adopts the fixed-effects approach in his research. The model — which uses the same data as the Economic Policy Institute — finds that workers in right-to-work states have average wages that are 1.9% higher than in states without a right-to-work policy.

The 2015 EPI paper has other limitations. One is that it uses just three years of data. A three-year research window probably isn’t wide enough to capture the full effects of these laws.

This was a concern of Mackinac Center scholars in 2013 when they tried to measure various impacts of right-to-work laws on economic well-being. That is why they chose to include 65 years of data (1947 through 2011) and use a fixed effects technique. They found that the presence of a right-to-work law meant a 0.8 percentage point increase in average annual inflation-adjusted personal income, of which wages are a large component.

It is important to critique the Economic Policy Institute paper. Its methodology has produced questionable results that were used to sell the public and lawmakers on a repeal of Michigan’s right-to-work law. More recently, it was cited in February in an effort to thwart the adoption of right-to-work legislation in New Hampshire.

The Employment Policy Institute was wrong about right-to-work laws and average wages. Right-to-work laws are powerful tools for economic growth and development. Scholarly evidence shows as much. Michigan should re-adopt its own statute as soon as possible.

Permission to reprint this blog post in whole or in part is hereby granted, provided that the author (or authors) and the Mackinac Center for Public Policy are properly cited.