Vigorous debate continues over the issue of “right-to-work” (RTW) laws, and how they affect a state’s economic growth and wage rates. Right-to-work laws make it illegal to require workers in unionized workplaces to pay dues to the unions that represent such workers in collective bargaining. RTW laws are believed to weaken labor unions.  As a result, it could be argued that RTW laws might attract some employers who want to avoid unions, but might also lower wages. The question is, what does economic research show about the strength of such relationships? Can states count on RTW laws as a reliable way to promote state prosperity?

The most recent state debating right to work laws is West Virginia. In West Virginia, recent research by Professor John Deskins and his colleagues, at the Bureau of Business and Economic Research at West Virginia University, has been cited as evidence in favor of RTW laws. For example, West Virginia House Speaker Tim Armstead stated that

“This study appears to support what we’ve been saying for some time – that a right-to-work law would certainly be advantageous to job growth in our state and is something we should take very seriously.”

I previously addressed the RTW issue in a 2013 column for Bridge magazine, and summarized research on the relationship between state RTW laws, state economic growth, and state wages.  I concluded in that column the following:

“Research suggests that the economic development effects of RTW are uncertain, with downside risks. Some studies find RTW boosts job growth, while other studies do not. Some studies find RTW reduces wages, while other studies do not. Precise predictions of RTW’s effects on per-capita earnings, which should be a primary goal for state policy, are not available.”

The question I want to address in this blog post is the following: is there anything in this recent research from West Virginia that would lead to any change in my prior summary of the research literature, that the effects of RTW laws are highly uncertain? The brief answer is No. The latest research does not provide any convincing evidence that a state that adopts RTW laws will as a result experience faster job growth.

The problem with the aspects of the West Virginia study that are cited by Speaker Armstead is that they do not overcome the fundamental problem with RTW research, which is the limited variation over time in which states are RTW states.  Research on the determinants of a state’s economic development has to face the fundamental problem that there are many characteristics of a state, both observed and unobserved, that potentially have major effects on a state’s economic fortunes. Unless a study controls for unobserved characteristics of a state that might affect its economic development, by looking at trends both before and after adoption of a RTW law, it cannot be considered to provide “certain” evidence of RTW effects.

The recent West Virginia results that are claimed to “certainly” show RTW effects on job growth do not look at how a state’s adoption of RTW affects job growth. Rather, the study simply looks at whether a state in a given year is a RTW state, and what its job growth is over the next 3 years. The study includes some controls for some other factors affecting state growth. But economic development research suggests that there are likely to be many other unobserved state characteristics affecting state growth.

Stronger evidence would be provided if the West Virginia study controlled for fixed effects of states on economic growth. The study then would essentially be estimating effects on job growth by looking at whether job growth increased in states that switched from non-RTW to RTW status over the sample period. Instead, the current West Virginia study is largely examining whether states that are persistently RTW states tend to have higher job growth than states that are persistently non-RTW states. But such a comparison does not provide strong evidence, because there can be many reasons why a state’s growth could vary.

For example, most RTW states have been persistently RTW states since before 1950, and are mostly located in the South. Southern states have tended to have faster growth and lower wage rates since 1950. But we wouldn’t want to regard this as strong evidence that RTW laws have caused this growth and lower wage rates. For example, much of the growth of the South has to do with the development of air conditioning, which has made Southern climates more attractive to workers and businesses.  RTW laws did not cause the South’s climate or the development of air conditioning.

I suspect that if the West Virginia study had controlled for state fixed effects, most of the estimated effects of state RTW laws would become statistically insignificant. Why would this be the case? The statistical insignificance of estimated effects of RTW laws in studies that do control for state fixed effects is largely due to the fact that until the last few years, only a few states have switched RTW status .For example, the only states that switched RTW status from 1980 to 2011 were Idaho in 1985 and Oklahoma in 2001.  A sample size of only two states switching RTW status does not provide very much information to determine RTW effects with much certainty.

Recent state policy changes may help allow us to have better estimates of RTW effects, but only after some time. Michigan and Indiana adopted RTW in 2012, and Wisconsin in 2015. After we have observed a full business cycle since these RTW adoptions, we may be able to obtain more precise estimates of RTW effects.

The most rigorous recent study that looks at RTW effects is a study by Eren and Ozbeklik, published in the Journal of Policy Analysis and Management in 2015. This study focuses on Oklahoma. It essentially constructs a counterfactual for what would have happened to Oklahoma if it had NOT adopted RTW in 2001 by matching Oklahoma with a weighted average of non-RTW states that closely match Oklahoma in pre-2001 trends. This study does find that RTW in Oklahoma reduced unionization rates, but finds no significant effects of Oklahoma’s adoption of RTW on job growth or wages.

Actually, the West Virginia study includes information that undermines the claim that adoption of RTW laws “certainly” increases job growth. The study includes some information on trends before and after RTW adoption in 10 states that adopted RTW laws after 1950. In 5 of these states, job growth increased after adoption of RTW, and in 5 states job growth decreased after adoption of RTW. This very mixed and uncertain result is representative of the overall findings of RTW research.  Based on current evidence, it is highly uncertain whether RTW laws have any positive effects on job growth. And some studies find that RTW laws are associated with lower wages, although this result too is highly uncertain.

If the effects of RTW laws are uncertain, are there policies with more evidence of positive economic effects? Probably the state policies with the greatest evidence for long-run effects on state economic growth are policies to increase the skills of a state’s workers. These include policies to increase participation by the children of a state in high-quality preschool, as well as policies to make postsecondary education more accessible. I have outlined the evidence for these economic development policies in numerous posts at this blog.