The Not-So-Simple Economics of Right-to-Work Laws
There was much hand wringing and an equal amount of triumphant cheering last month, when Michigan became the twenty-fourth state to adopt a right to work (RTW) law. It joined Indiana, which went RTW last February, as only the second major industrial state with such a policy. Much of the commentary portrayed the spread of RTW as a victory for corporate power over working America, but do we really know what the effects will be? The rhetoric surrounding the passage of Michigan’s law has shed more heat than light on the matter. A less impassioned review of the issues suggests that the effects of RTW on labor markets are not as simple as supporters and opponents make them out to be.
What are right-to-work laws?
The term “right to work,” with its misleading libertarian connotations, is itself part of the problem. In both RTW and non-RTW states, contracts between unions and employers are extensively regulated. RTW laws make only a few changes in the rules.
The story starts with the Taft-Hartley Act of 1947, which outlawed closed shop contracts. Previously legal, such contracts stipulated that the employer could hire only workers who were already union members. Since Taft-Hartley, closed shops are illegal even if both workers and employers unanimously agree to them. The act originally did allow union shop agreements, which require workers to join a union after hiring, and agency shop agreements, under which non-members must pay dues or equivalent fees to a union that bargains collectively for all employees.
Since that time, U.S. Courts have chipped away at union shop and agency shop agreements, as well, further weakening the sway of unions over nonmembers. The courts have ruled that no one can be compelled to join a union, all but eliminating union shops. Even in agency shops, where nonmembers are not required to join the union, a rule established in Communications Workers v. Beck (1988) limits the fees they must make in lieu of union dues. Those agency fees can only reflect the actual costs of collective bargaining, and not, for example, union political activities.
At the same time, even the RTW laws that states have passed in response to Taft-Hartley leave the influence of unions over nonmember workers intact in some important respects. True, RTW does grant the right to work in a unionized workplace without paying dues or agency fees. However, in RTW states as in non-RTW states, collective bargaining remains subject to many restrictive rules. In RTW states as elsewhere, a union must bargain for all workers in a relevant category if it bargains for any of them. Even if both workers and unions wanted to, they could not enter into a contract that set wages and benefits of union members collectively while leaving otherwise similar nonmembers to the mercy of one-on-one bargaining.
In short, workers in non-RTW states are not subject to “compulsory unionism,” as the overheated rhetoric of RTW supporters sometimes claims. Similarly, the “rights” granted by “right to work” laws fall well short of complete freedom of contract.
Measuring the effects of RTW
If we could assign RTW or non-RTW status randomly among the fifty states, it would probably be possible to measure the economic effects using standard statistical methods. In practice, though, RTW states are by no means a random selection. Those that have adopted RTW laws are largely concentrated in the South, the Great Plains, and the Mountain West. Most of them went RTW soon after passage of Taft-Hartley. In the thirty-five years from 1976 to 2011, just two small states (Idaho in 1985 and Oklahoma in 2001) passed RTW laws. The result is that the RTW states differ from non-RTW states in systematic ways, including both quantifiable attributes, such as demographics, and less quantifiable traits, such as social attitudes and business environment.
The nonrandom distribution of RTW laws creates problems for statistical studies based on cross-sectional data. Most such studies have focused on three major differences between RTW and non-RTW states:
- The percentage of union workers is higher in non-RTW states. Before 2012, only one of the 25 states with the highest percentage of union members had RTW laws, while all but one of the 20 least unionized were RTW. A recent study by Elise Gould and Heidi Shierholz for the Economic Policy Institute found that 18.6 percent of workers in non-RTW states were union members or covered by union contracts, compared to just 7.6 percent in RTW states. Even though non-RTW states are more heavily unionized, it is worth noting that the great majority of workplaces are non-union in all states. That fact makes it harder to tease out the statistical effects of the laws than it would be if RTW states were union-free and non-RTW states were predominately unionized.
- A number of studies, often cited by RTW opponents, show wages to be higher in non-RTW states. For example, Gould and Shierholtz concluded that wages averaged 13.7 percent higher in non-RTW states. When they used multiple regression methods to adjust the data for differences in age, education, ethnicity, cost of living, and other variables, the difference decreased to a less dramatic but still statistically significant 3.2 percent. Both union and nonunion workers earned higher wages in non-RTW states.
At the same time, other research, regularly cited by RTW advocates, maintains that employment and output have grown more rapidly in states that have adopted RTW laws. For example, a study by Richard Vedder published in Cato Journal, estimated that if two states were otherwise similar, the one with a RTW law would have a 23 percent higher rate of economic growth by 23 percent.
Not all studies agree that RTW lowers wages and stimulates job growth. However, for a theorist, there is nothing surprising about either result. They are exactly what we would expect from the simple textbook model of labor unions. In that model, unions use collective action, including strikes, to raise wages. That pushes employers up and to the left along their labor demand curves. Workers end up with higher wages but fewer jobs, which is great for workers who retain their union jobs. At the same time, employers who do not want to deal with unions, and workers who are not lucky enough to get high-wage union jobs, have an incentive to migrate to RTW states.
It is worth noting, in passing, that an inverse relationship between labor demand and wage rates undermines the “free rider” argument often invoked by RTW opponents. For example, an AFL-CIO “primer” on RTW says “Because everyone benefits [from collective bargaining], everyone should pay their fair share. But RTW laws allow cheaters to take unfair advantage of the system.” However, if union activities mean better jobs for some at any given unionized company, but fewer total jobs, it would not necessarily be true that everyone benefits. In particular, unionization of a workplace or a contract that increases unit labor costs would clearly not benefit workers who lack seniority or key skills, because they would be the ones at greatest risk of losing their jobs. That is presumably one reason why NLRB elections and ratification votes are not always unanimous. Workers don’t join unions or who vote against the recommendations of union leaders on ratification ballots may simply be rational actors, not “cheaters.”
The question of causation
Even if issues regarding the statistical associations of RTW laws with wages and job growth could be fully resolved, a fundamental ambiguity of causation would remain: Are RTW laws themselves the source of observed differences among state labor markets, or are they only markers for more fundamental differences?
One hypothesis is that most RTW states were those that were less industrially developed at the time Taft-Hartley was passed. Their lower level of industrialization meant that labor unions were politically weaker, making passage of RTW laws easier. If that were the case, then even if Taft-Hartley had not come along to allow passage of RTW laws, we would still have expected lower wages and less unionization in those states, and, as a result, lower wages and faster job growth.
A variant on this hypothesis is that RTW states were, from the beginning, culturally less hospitable to unions and politically more hospitable to employers. Accordingly, such states have enacted not just RTW laws, but whole packages of pro-business measures, including lower business taxes, less regulation, favorable zoning, and so on. In that case, it may be impossible to isolate the specific contribution of RTW to the effects of the larger pro-business package.
The bottom line
The bottom line is that the economic effects of RTW laws are not nearly as clear-cut as their advocates and opponents make them out to be. Correlations of RTW laws with wages and employment are economically small even when they are statistically significant. Most problematic of all is the question of causation—does RTW cause observed differences between states, or do pre-existing differences cause the passage of RTW laws?
What we need to get a clearer picture of the effects of RTW on labor markets is a controlled experiment. If all the world were a laboratory, we could do something that has never been done before. We could switch a couple of strongly unionized states in the industrial Midwest from non-RTW to RTW. Michigan (fifth most unionized) and Indiana (twenty-first most unionized) would be good choices. By watching what happened over the next five or ten years, we might finally learn whether RTW laws really make a difference.
So, on behalf of all experimental economists, thank you, Governors Snyder and Daniels! We don’t know yet if what you have done will be good for workers or businesses in your states, but we may finally get the answers to some longstanding questions about Right to Work laws.
12 Responses to “The Not-So-Simple Economics of Right-to-Work Laws”
Corporations clearly benefit from right to work laws. It would be interesting to examine the corporate profitability, corporate tax rates and corporate campaign contributions in RTW vs. Non-RTW environments.
Liked your piece today.
You should read book: The Southern Advantage. About the various advantages of the American South in inbound recruitment of regionally-footloose investment. Since RTW is concentrated in 17 Southern States, some relevance.
My thought: As more and more states go RTW, the advantage is dilluted. Also, there are other sources of competitive advantage, RTW being only one.
"At the same time, employers who do not want to deal with unions, and workers who are not lucky enough to get high-wage union jobs, have an incentive to migrate to RTW states."
This is only partially correct. Employers also have an incentive to move work overseas and former employees have the right to lose their jobs and not recover them. Proper analysis of RTW vs non RTW should not be viewed as a closed system within the confines of the USA alone.
It shouldn't take an economic genius to figure out that all other things being equal, if manual labor can be had at a lower cost than elsewhere then companies will "eventually" move their operations to the lower labor cost provider. If Unions guarantee union shop employees have higher labor cost than non union shop employees (after all that is the service that unions are supposed to provide their members – in more ways than one, e.g. cost of strikes, cost of demarcation, additional income to offset membership dues) then inevitably companies move away from union shops (all other things being equal).
That has been the trend over the last 40 years, so do we really need more research reports to conclude the obvious? During that same period of time Unions have been pricing their members out of the market – which in case they haven't realized is global. The main reason the shift has been slower than expected is the cost of companies moving operations, government trade protectionism, monopolistic powers certain large unions have such as the UAW and legacy skills/knowledge/know how of workers vs that of new green site employees, but that last advantage is all but eroded away in most industries.
You say "This is only partially correct. Employers also have an incentive to move work overseas and former employees have the right to lose their jobs and not recover them. Proper analysis of RTW vs non RTW should not be viewed as a closed system within the confines of the USA alone."
Yes, I agree. I can't make a full extension of the discussion to a global scale in this comment box, but I would note two points that fit your suggestion:
1. Viewed in a global context, the RTW states act as a sort of safety valve that reduces the amount of off-shoring that would take place if all parts of the country were equally unionized up to the level of the most unionized states.
2. However, it is not all about off-shoring, but about on-shoring, too. For example, for Japanese and German car makers, the US is a low-cost manufacturing site, and most, if not all, foreign auto manufacturing operations in the US are in RTW states.
With respect to New Reality's comments, obviously more research is needed as causality hasn't even been established, ie did business friendly states enact RTW as a result of them being business friendly ordid RTW result in them being more business friendly.
Secondly, one of the commentator's points is that the change has taken place slower than expected due to the high cost of moving. However, US states subsidize interstate business relocation at arguably the highest levels in the world yet the transfer of industry is still slower than expected.
I think the bigger issue with unions is lost productivity and the commentator identifies some of these issues. Additional issues with unionization include inflexibility within the workplace that restricts a company's ability to respond to changing economic conditions. By this I don't mean by laying of workers but rather to reassign them. Interesting to note that most of the issues surrounding unionization are for non-professional unions, ie low skill mfg unions and not professional unions like electrical ones or higher skill mfg unions. I've always said that if unionize workers were the most highly skilled and productive workers then most reasonable companies would pay the higher rates easily.
Ed, thanks for the reply.
On item 2, you are correct the facts clearly show that there has been some on shoring back to the US by foreign companies that sell product within the US. Now I'm not sure how much of that is driven by the low cost of skilled/productive labor within this country and how much of that is driven by state sponsored incentives (low/no taxes, lower utility costs, 0% land leasing, etc – there are many carrots that states offer). Having said that though what is also clear from foreign companies specially in the auto industry is the fact that they have moved to RTW states and avoided like the plague states such as Michigan(Detroit). You would think that Michigan(Detroit) of all places would have the most skilled/productive auto work force in this country that they (foreign companies) would have been eager to tap into that skill/productive labor pool not to mention the surrounding auto ecosystem when they first moved here. I still hold that Unions are pricing their members out of the market and the consequences i.e. trend has been pretty clear and continues to be so, massive bailouts of the auto companies aside.
"Secondly, one of the commentator's points is that the change has taken place slower than expected due to the high cost of moving."
C Panschow, I assume you are referring to my comments, here is what I said:
"The main reason the shift has been slower than expected is the cost of companies moving operations, government trade protectionism, monopolistic powers certain large unions have such as the UAW and legacy skills/knowledge/know how of workers vs that of new green site employees, but that last advantage is all but eroded away in most industries."
i.e. I didn't just say it was cost alone. However having said that, cost can take on many forms and sometimes not easily quantifiable For example if a company that is heavily unionized in non RTW states say one of the auto companies and wishes to setup a new plant in say a RTW state, what cost will the company incur when inevitable strikes occur in their existing plants? These strikes have a cost and ends up being indirect cost associated with setting up the new plant. I referred to this type of scenario as "monopolistic powers certain large unions have such as the UAW" in my original comment, but as you can see ultimately they translate into some form of cost in their cost analysis when making decisions.
This also dovetails with your statement:
"Additional issues with unionization include inflexibility within the workplace that restricts a company's ability to respond to changing economic conditions."
E.g. Union inflexibility in allowing a new plant to be setup in a RTW state may force the company to decide not to setup the plant when direct and indirect costs are all added and factored in.
You also said:
"I've always said that if unionize workers were the most highly skilled and productive workers then most reasonable companies would pay the higher rates easily."
This would be great, but: Are Unions responsible for the skill and productivity of their members? If the answer is yes, then I would argue that a Union's self interest has traditionally been to lower the bar and not raise it relative to the cost being negotiated for the labor.
Great article and the great comments on very complicated & emotional issue.
"Labor markets" – is the key word. Labor is a commodity. Capital may be a commodity also, but it's also surplus value created by labor. Workers, as a commodity, are deluded if they think they are rational actors – their "rational" actions are circumscribed and determined by market conditions. Collective bargaining is the leverage that workers have to get a human result from a mechanical situation. Management doesn't need this because they feather their nests with the surplus value created by labor, i.e. luxuries, amenities, the trappings of affluence. Conservatives have, with some appeal to "freedom", manipulated workers into thinking they are entrepreneurs – which of course, is false. Marx was correct.
The post is interesting and can be extended to labor unions and labor market rigidities, such as the Italian law that does not allow workers to be dismissed for "economic reasons". These rigidities are very important in the EU today. Also investment in the U.S. auto industry is primarily in the right to work law states and is heavily by foreign nameplates.
Ed, please double check this. Both sides have agreed with this statement over the years, until recently, when spokesman from both sides have acknowledged that under most recent court decisions, this is no longer the case:
"In RTW states as elsewhere, a union must bargain for all workers in a relevant category if it bargains for any of them. Even if both workers and unions wanted to, they could not enter into a contract that set wages and benefits of union members collectively while leaving otherwise similar nonmembers to the mercy of one-on-one bargaining."
Both sides have made statements recently that indicate that a union is free under current federal law to negotiate a members-only contract. And of course, both state and federal law could be amended to require members-only contracts.
Hello New Reality, Good comments back. My only additional one is that in Ontario some construction unions have actually built their own trades training centres (Cdn spelling of center) and use these to train their members. That shows leadership, which is certainly something other unions could learn from. The goal should be having the best workers in the world as members.