I recently criticized Don Boudreaux’s two major arguments against legal scholar Eric Posner’s claim that monopsony in U.S. labor markets is widespread. But I do agree with Don that Posner’s case is defective.
Posner writes:
Economic theory says that when a pool of workers has only one potential employer, or a small number of potential employers, those workers will be paid below-market wages.
Actually, economic theory doesn’t say that. I’ll put aside the quibble that wages couldn’t be below-market because whatever we’re observing is the market. It’s clear from context that Posner means below the wage that would be paid in a competitive market, so that’s what I’ll consider.
(By the way, if you follow the link in the Posner quote above, it will take you to a lengthy study on monopsony by former President Obama’s Council of Economic Advisers. I critiqued that study here, here, here, and here.)
Posner is right that economic theory says this is true when there is only one potential employer. But it’s silent on the issue when there are “a small number of potential employers.” There are often industries in which there are only a few firms and these few compete aggressively in the output market. It’s hard to believe that they wouldn’t compete aggressively in the market for inputs, in this case, labor.
In his article “Monopoly” in David R. Henderson, ed., The Concise Encyclopedia of Economics, economist George Stigler referenced a study by fellow University of Chicago economist Reuben Kessel on underwriter spreads in the bond market. Kessel found that when there were 20 bidders, the spread (the price charged by the underwriters) was $10 and when there was one, it was $15.74. That latter price is consistent with Posner’s claim. But, Kessel noted, just adding one extra competitor brought the spread down to $12.64, a major drop. Having 6 competitors brought it down to $10.71, almost all the way to the spread charged by 20 competitors. In short, 6 competitors led to a very competitive result. In most people’s eyes, 6 is a “small number of potential competitors.”
Posner writes:
In one paper, José Azar, Ioana Marinescu, Marshall Steinbaum and Bledi Taska found that more than 60 percent of labor markets exceeded levels of concentration that are regarded as presumptive antitrust problems by the Department of Justice.
Ok, but that’s not enough evidence. That’s simply evidence that the Department of Justice has a presumption. Has the Department of Justice dealt with Stigler-type reasoning and Kessel-type empirical analysis? (The paper he cites does have evidence that concentration is correlated with lower wages, but I haven’t examined it.)
Posner does give some good evidence, writing:
For example, Elena Prager and Matt Schmitt examined hospital mergers and found that when hospitals expand through mergers and gain significant market power, the wage growth of employees declines. Notably, this decline affected skilled health care professionals like nurses — but not administrators and unskilled staff members like cafeteria workers, who could easily find jobs outside hospitals.
I wonder if Posner’s aware of two major factors that lead to concentration in the medical sector. One is regulation. The medical industry is one of the most regulated industries in America and became even more regulated with ObamaCare. When an industry is regulated, there are typically what I call “economies of scale in compliance.” A firm with 10 times the size of another firm bears costs of compliance that are less than 10 times the cost for the other firm. Even without mergers, that knocks out small firms and it also leads to mergers so that firms can take advantage of those economies of scale in compliance.
The other factor is Certificate of Need Regulations that many states have. The acronym is CON, and it’s a great acronym because it is a con. These regulations prevent surgery centers from arising to compete with hospitals and prevent hospitals from expanding to compete with other hospitals. When a surgery center or a hospital goes before a regulatory board to get government permission, guess who often intervenes to object to the new competition? That’s right: the current competitors.
In fact, some of the toughest CON regulations are in Posner’s state of Illinois. And in his 2008 book, Code Red, a health economist just up the street from Posner, David Dranove of Northwestern University, exposed some of the bad effects of CON regulation in Illinois. Prices and wages aren’t the only thing that matter. Dranove writes that due to CON regulations:
Illinois hospitals today are located where Illinoisians lived in the 1950s.
Posner also contradicts himself in his last paragraph, writing:
Labor monopsony affects people at all income levels, but it is a particular problem for lower-income workers and people living in stagnant rural and semirural parts of the country.
But lower-income workers tend to be less specialized. They tend to be the unskilled workers whom Posner says are not affected by monopsony. What gives?
Note: The picture at the top is of George Stigler, whose work is referenced in this post.
READER COMMENTS
Daniel B
Oct 1 2021 at 12:24am
I think another reason to be skeptical of the Justice Department study argument is that the Justice Department has every incentive to see monopoly in as many places as possible. Otherwise there’d be fewer (if any) monopolies to fight, so there’d be less work to do and thus the people in there would lose appropriations and jobs. In short, the fox is in charge of the hen house, and at minimum we have to realize that.
I’m not saying everyone in the Justice Department makes this cynical calculation. But I don’t think anyone can deny the ability of people to rationalize. The well-intentioned in the Justice Department’s anti-trust section (and surely there are plenty of them) are naturally going to see many things as nails, when the main thing they have to “do good” (which is their goal by definition) is the hammer of their anti-trust jobs. And obviously the well-intentioned need money to do them.
I feel there’s still skepticism at my argument so let me reiterate it this way. Money is a factor with the Justice Department, but so are other things. Importance – being the hero that saves the day from monopolists and the psychic satisfaction from that – is one factor. Visibility and influence are involved with importance. I think everybody understands that all these factors have been ample incentives for most people to do things throughout history. Most of us probably realize that when the devil on our left shoulder tells us to do something it can be quite hard to resist its call. It must be even harder to resist when the devil on our left and the angel on our right shoulder tell us to do things. And that’s the situation with the Justice Department I’m pointing out here.
Pete Smoot
Oct 1 2021 at 11:51am
I’m a cynic and even this is too cynical for me. You’ve definitely got a point: whether people at Justice know it or not, they’ve got an incentive to never actually get rid of all monopolies.
However, I think that will play out is people will “solve” one monopoly problem, then go hunting around for new ones to solve, not because they want appropriations but because they believe their job is important so of course there’s more to do. Once the current crises are settled down, I could totally imagine them inventing a hypothetical problem, cherry picking some evidence to prove it could be a problem (even it it’s not yet: look how proactive we are!), then charging away to solve the potential problem.
I doubt you could find anyone at Justice who, even when drunk and in private, would admit they’re inventing problems just so they can solve them, and that they’ve long since passed the point of diminished returns, no more than a typical manager at a typical company will say “we’ve got enough products and are big enough: no need to make things better.”
Daniel B
Oct 1 2021 at 8:00pm
I doubt you could find anyone at Justice who, even when drunk and in private, would admit they’re inventing problems just so they can solve them, and that they’ve long since passed the point of diminished returns
I think I didn’t communicate clearly enough. When I said the JD “has every incentive to see monopoly in as many places as possible” I don’t mean that they are cynically fabricating examples of monopoly. I simply mean it’s in their self-interest to see monopoly – even where it doesn’t exist.
Most people at the JD are sincere and well-intentioned. But as Milton Friedman wisely observed, “sincerity is a much overrated virtue.” I think this paragraph from Free to Choose (pp. 38-39) illustrates my point about self-interest:
Adam Smith pointed to “the interested sophistry of merchants and manufacturers.” They may have been the chief culprits in his day. Today they have much company. Indeed, there is hardly one of us who is not engaged in “interested sophistry” in one area or another. In Pogo’s immortal words, “We have met the enemy and they is us.” We rail against “special interests” except when the “special interest” happens to be our own. Each of us knows that what is good for him is good for the country—so our “special interest” is different. The end result is a maze of restraints and restrictions that makes almost all of us worse off than we would be if they were all eliminated. We lose far more from measures that serve other “special interests” than we gain from measures that serve our “special interest.”
Friedman was right. I recently read a book called Welfare for the Rich that contains telling examples of “special interest” thinking. One of my favorites is in the chapter on zoning, where a Santa Barbara houseowner says “It’s just fine with me if we keep this shortage going” because it means he gets to charge higher rents for the house. Then he says, “We need to get this homeless situation under control” – conveniently overlooking the fact that he is contributing to it with his acceptance of higher rents.
To give a blunt example of a government actor pursuing his own interests, Thomas Sowell’s book Knowledge and Decisions tells this story (pp. 308-309 of the Kindle edition, but it doesn’t match the print version at the time of writing):
Building subsidies in various government housing programs are routinely understated at the outset, even though it will obviously be impossible to conceal them indefinitely, because, as one federal official said (in justification), “if you put these huge capital contributions up front there’s no way any administration would propose it or any Congress would approve it.” In other words, the voters would never stand for it if they knew.
The official is willing to mislead other people for his political goal of subsidies. Thomas Sowell’s story of when he worked at the Labor Department reveals another example of government self-interest. In short, don’t underestimate the ability of people to rationalize (you already understand this though).
zeke5123
Oct 1 2021 at 3:47pm
Another way to describe it is that people who believe heavily in monopolies are likely to come to work for the justice department in anti-trust. That is, the very nature of the job likely results in bias.
Knut P. Heen
Oct 4 2021 at 5:54am
Good point. People tend to select jobs we think are important. The result is that everyone has an inflated view of the importance of our job. We see the same effect with financial analysts. They select to follow stocks they think will do well in the future. Hence, on average, they are too optimistic with their recommendations. You see the same effect in research as well. You don’t become a sociologist if you think that a mathematical description of the world is important.
Mark Z
Oct 1 2021 at 3:03pm
David, I have a more fundamental question that maybe as an economist you can answer. Is anti-trust action really even conducive to increasing competition, even when there is market concentration? If an industry is dominated by two firms with 50% of the market each, and you split those firms up into 4 firms with 25% each, well, now you just have 4 smaller monopolies (since the 4 firms have little incentive to compete with each other, since if any gains more market share, it’ll be targeted for break up under anti-trust law as well). It seems like anti-trust actions just ‘fix’ the preferred measure of ‘monopolism’ – market concentration – rather than fixing the actual problem, lack of competition. Or is there something I’m missing?
If I thought an industry were too monopolistic, I would try to make that industry more competitive, not break up the monopolists. Rather than chopping up Google, I’d liberalize intellectual property law in software so potential competitors could use their code to build their own search engines to compete with it.
David Henderson
Oct 1 2021 at 3:37pm
You write:
It can be. Check the Reuben Kessel data in the Stigler piece I link to.
You write:
I think that’s an overstatement. I think they have more incentive to compete when there are 4 than when there are 2.
The problem is that the discussion takes place without the trustbusters paying careful attention to why there are only 2 firms or 4 firms or whatever. Is it because case of innovation? Is it because of economies of scale? Those are the kinds of questions Harold Demsetz addressed in Steve Globerman’s and my recent book, The Essential UCLA School of Economics.
You write:
So would I, as I note with my discussion of CON regulations.
Comments are closed.