The CEA's Mixed Thinking on Labor Market Monopsony, Part IV
By David Henderson
This is the last post in a series of four (here, here, and here) in response to the Council of Economic Advisers’ recent report on monopsony in the labor market. At the end, I’ll sum up. (In yesterday’s Wall Street Journal, CEA chairman Jason Furman and Princeton University professor and former CEA chair Alan Krueger have an op/ed that covers some of these themes, but, as far as I can tell, there is no ground covered there that is not in the CEA report. So I won’t comment on it.)
Recall that their big upset is about monopsony and that a way around the efficiency loss from monopsony is for employers to price discriminate, paying less to workers with low supply prices and more to workers with high supply prices.
Improving Pay Transparency
The CEA writes:
In 2014, the President signed Executive Order 13665 Non-Retaliation for Disclosure of Compensation Information. The EO prohibits Federal contractors from discriminating against employees and applicants “who inquire about, discuss, or disclose their own compensation or the compensation of other employees or applicants.” It represents one step forward in stopping the widespread practice of firing or otherwise punishing employees for talking about their pay. (p. 14)
Do you see a tension here between their lauding this Executive Order and their concern about monopsony? They apparently don’t, but there is one. What would employers who are paying different amounts to the workers want the employers not to do? Discuss their pay with each other. So, ironically, the very Executive Order they like could well undercut the employers’ attempts to reduce the inefficiency that arises from monopsony.
Promoting Equal Pay
The CEA writes:
When firms have wage-setting power, they have an incentive to pay the lowest wage that workers are willing to accept–meaning that individuals who start out facing greater obstacles and fewer opportunities can end up being paid the least. This pattern may be contributing to the gender pay gap.
Women make up nearly half of the U.S. labor force, and are increasingly entering industries and positions traditionally occupied by men. Yet the typical woman working full-time all year earns only 80 percent of what the typical man earns working full-time all year. Despite passage of the Equal Pay Act of 1963, which requires that men and women in the same work place be given equal pay for equal work, the gender wage gap persists. (p. 15)
The CEA does not point out that there is a big difference between the typical woman who works full time and the typical man who works full time: years out of the labor force for women due mainly to child rearing. When economists look at women and men who have never been out of the labor force, and correct for occupation, etc., they find almost total equality of wages.
But put that aside. What if employers did pay systematically less to those with “fewer opportunities,” which typically means those with lower supply prices? Again, that would mean that the employers are finding ways to reduce the efficiency loss from monopsony. But the CEA makes not a peep about that, not even a footnote.
Paid Sick Leave
The CEA writes:
Imperfect competition in the labor market allows firms not only to pay lower wages but also to lower costs through reductions in benefits. Policies that support minimum benefits are therefore an important complement to minimum wage and overtime laws to counter the market power of employers. (p. 15)
Whoa! Let’s assume that there is monopsony (“imperfect competition”). Let’s assume further that employers have not figured out to price discriminate so that they are paying everyone the same. Then it follows that the wage is “too low” relative to the efficient wage. But their claim that minimum benefits are an antidote is false. The employer has no incentive to give a suboptimal mix of wages and benefits. The overall pay will be “too low,” but there is no good reason to think that benefits will be too low relative to wages. Why does this matter? Because their preferred solution would make things worse. Mandating minimum benefits would do nothing to “counter the market power of employers.” Instead, it would cause employers to pay higher than optimal benefits and even lower money wages. Judged how? Judged from the viewpoint of the workers, who find the mandated benefits less attractive than the wages they lost. Thanks, government, for nothing.
It was at this point that I became convinced that this report is not just an economic, but also a political, document. My guess is that Furman, the CEA chairman, is too good an economist not to see this. If so, then it seems that one of the main purposes of the document is to push the Obama agenda even if it doesn’t follow from the economics.
Occupational Licensing and Land Use Regulation
On page 16, the CEA writers emphasize what they wrote earlier in the piece, namely that occupational licensing and land use regulations limit mobility. Good for them.
The CEA writes:
In the absence of an up-to-date standard delineating who is exempt from the overtime protections of the Fair Labor Standards Act, monopsony power can allow firms to demand long hours from workers who are not eligible for overtime but who have relatively low salaries. The salary threshold below which most salaried, white collar workers are entitled to overtime is currently so outdated that it provides automatic overtime protections based on salary to just 7 percent of full-time salaried workers today, compared with 62 percent in 1975. In May, the Department of Labor published a final rule that will automatically extend overtime pay eligibility to 4.2 million workers when it takes effect on December 1st. The rule will entitle most salaried white collar workers earning less than $913 a week ($47,476 a year) to overtime pay. (p. 17)
Sure, monopsony power “can allow firms to demand long hours from workers who are not eligible for overtime but who have relatively low salaries.” But so can zero monopsony power. The CEA doesn’t even consider that there are good reasons to keep flexibility in how workers are paid. I’ve discussed that here. The overtime practices of employers probably have little to do with monopsony. I’m wondering if the CEA writers think that too, given that they don’t even try to make a case for their claim about the effects of monopsony.
The Big Picture
Let’s take a step back and look at the big picture. Where is there likely to be monopsony and how widespread is it likely to be? If workers have many choices of employers, monopsony is likely to be unimportant or non-existent. Which workers would likely be in that position? Most likely, the least-skilled workers. Many employers need unskilled workers. And if that’s not true, it’s likely to be due to the minimum wage pricing them out of jobs. So, far from the minimum wage being an antidote to the lack of opportunities, it’s much more likely to be a cause of this lack.
Where, then, would we find monopsony? It would be much more likely where there are only a few employers. Where would that be? Probably where employees have highly specific skills. Which employees would have highly specific skills? Generally employees who are fairly well paid. And even there, a lot of monopsony power is unlikely.
When I graded for an undergraduate labor economics course taught by the late George Hilton at UCLA, he had a question on an exam based on the famous 1956 Journal of Political Economy article by the late Simon Rottenberg, “The Baseball Players’ Labor Market.” Part (a) of the question asked the student to draw the graph I drew in Part I and explain it. Part (b) asked how effective this monopsony was at holding down baseball players’ wages. The answer George was looking for is that it was effective but not as effective as you might expect because if it were completely effective, it would hold wages down to the players’ supply prices and yet wages were far above their supply prices. Part (c) of the question was, and I remember this almost word for word: “In light of your answer to part (b), how widespread is monopsony likely to be in labor markets in the United States?” What Hilton and I (the grader) were looking for was: “Not very effective because even in a highly specialized labor market with few employers and, moreover, a collusive agreement among employers that is legally enforced, wage rates are well above most workers’ opportunity costs. So when there is no legal collusion, and when workers are less specialized so there are many employers competing for them, monopsony is likely to be unimportant.” The CEA has given very little reason to change that assessment.