Assuming Your Result: Another Great Moment in Economic Silliness
-- Posted by Neil H. Buchanan
Are executives overpaid? If you are inquisitive and open to thinking about evidence and logic, that is a difficult and nuanced question. If, instead, you are a mainstream economist, then this -- like all questions -- is easy. Although an awful lot of work goes into dressing up the analysis, the answer boils down to this: Executives cannot be overpaid, because what they are being paid must be the amount that they should be paid. Think I'm joking? If only.
The latest example of this nonsense showed up in a fawning news article by a New York Times sports reporter last week. Two economists (one of whom teaches at a prominent law school) wrote a typical economics article, in which they claimed to show that college football coaches' salaries are justified by the coaches' jobs being CEO-like. That is, they noted that big-time college football coaches run large organizations, in which they manage hundreds of people, and they are asked to deal with the media, politicians, the public, and so on. Moreover, they are supposedly held to account by the rigorous and unyielding performance standards of winning and losing.
(Note: Although I have written at length recently about college sports, this post is not concerned with any of those issues. I am using the Times article and the paper that it discusses to make a point about current practices and norms in academic economics. The football aspect is incidental.)
Once one has made the CEO comparison, however, it might occur to most people that the right way to think about this is through the well-established body of research that shows that executive pay is anything but market-driven. This research goes back decades, and it is based on both objective evidence and the testimony of many people who helped rig the executive pay game in the first place. Salary committees (at least those that are trying not to look too obviously like the CEO's puppets) set up nominally objective measures that, by design, push compensation inexorably upward.
The standard approach, widely discussed by everyone who follows such things, is to set up a game of leap-frog, comparing Company A's CEO's pay to that of Companies B, C, and D. If D's CEO gets a raise (based on, say, an increase in the share price for one quarter), then A's CEO can say, "But wait, I was the highest-paid CEO in this group, and now I'm not, even though I've done nothing wrong." B and C say something similar, and because it is all an inside job, "no" is not an acceptable (or even thinkable) answer. There is an upper limit, because there is a finite amount of money, but there is no logical connection between the quasi-equilibrium and "market fundamentals."
Moreover, there has been a large body of research indicating that the measures of "performance" that are used to measure CEOs' worthiness are deeply flawed. In organizations that sometimes employ tens or hundreds of thousands of people, all results (or at least all positive results, but sometimes negative results as well) are attributed to the top executive. "Creation of shareholder value" is said to be the doing of the dynamic chief executive. In the analogy to college football, wins and losses happen entirely because of the head coach.
These objections to the idea that CEOs are paid "what they are worth" are so well-known, in fact, that even NYT op-ed columnist Joe Nocera knows about them. In his column last Friday, he ran through most of the common objections to the conservative economic mantra on CEO pay. Of course, being a slave to the conventional wisdom, he indulged in a bit of academy bashing: "It’s amazing the things academics can find worthy of study, isn’t it?" Yes, the problem is woolly-headed professors, not the takeover of economics and related fields by a cult-like group of apologists for power. Right? But I digress.
When I say "even ... Joe Nocera knows" these things, I am referring to the obvious fact that Nocera frequently does not appear to know what he is talking about, even in his original field of journalistic expertise: business. He is an enthusiast, letting his emotions get ahead of facts and logic. That is not to say that he is always wrong. (He is no David Brooks or Ross Douthat.) For example, his newest column provides a reasonable discussion of "corporate inversions," which are a major tax policy issue. My point is that, if even someone as insistently unimaginative as Nocera knows about the arguments regarding executive pay, then those arguments must be reliably mainstream. (Nocera also spins out of control at the end of the column, failing to understand the cash flows in university budgets. But that is par for the course.)
That, however, does not stop our intrepid economics professors. Rather than acknowledge that there are competing theories within labor economics, they pepper their paper with references to a monolithic thing called "economic theory," which provides unambiguous answers that can be tested. From the abstract: "[Coaches] contracts’ features are consistent with what economic theory would predict." And from the second paragraph of the paper: "Yet economic theory tells us that lucrative compensation is not necessarily a problem if the structure of the employment contract is aligned with value creation." And in the conclusion: "In sum, we find that a close comparison of CEO employment contracts and college football coach contracts shows that both sets of agreements conform to the predictions of economic theory."
What is this unvarying economic theory to which the authors refer? They never quite say, because they apparently believe it to be obvious. The idea is that rational actors would not pay people more than they bring into the organization, because that would be irrational. (Circularity is a big part of modern economics.) Indeed, the final sentence of the article all but gives away the game: "Thus, if one believes that CEO compensation is set by the market at an appropriate level, and that employment contracts reflect this equilibrium, then one should reach the same conclusion about football coaches." (This sentence was reprinted in the NYT article that I noted at the beginning of this post.)
And should we believe that CEO compensation is appropriately set by the market? The authors clearly view this, too, as a trivial question. Indeed, they could more honestly have written: "Thus, because one should believe that CEO compensation is set by the market at an appropriate level ... ." For true believers, there is no "if" about it.
What is especially interesting here is that, as bad as the logical starting point might be, the analogy between CEO compensation and college coaches' pay is worse. At least in for-profit businesses, the notion of "maximizing shareholder value" can be defined with some logical coherence. Again, that does not mean that CEOs are actually paid according to a sensible definition of such value, but one can at least start from there to analyze the important questions. But other than inventing the term "program value creation," the authors have nothing useful to say about how to measure "value" in the college sports context. Maybe they are assuming that the money coming in from TV contracts is somehow tied to coaches' win-loss records (a truly bizarre idea); but how is the increased "value" of a big-time sports program meaningfully compared to what a coach does?
Indeed, the paper cannot (and does not try to) deal with the fact that most of the programs in question are drags on the parent universities' finances. It would be more sensible financially to shut down many of these programs and turn them into club sports or Division III programs without athletic scholarships or TV money. Yet, we can be sure that alumni and, in the case of public universities, legislators would quickly tell us that the true "value" of the program is much more than the TV, ticket, and memorabilia revenue. It is all about creating warm glow, building loyalty, encouraging applications, and so on.
What would "economic theory" tell us about that? Well, if we assume that CEO pay is set by the market, and if we assume that the college football market rationally and efficiently aligns coaches' pay with whatever people "value," then the answer is that the coaches are being paid what they should be paid. Assuming one's result is easy. What takes some skill, as this paper demonstrates, is making it appear as if one is doing anything more than merely justifying the status quo.
Are executives overpaid? If you are inquisitive and open to thinking about evidence and logic, that is a difficult and nuanced question. If, instead, you are a mainstream economist, then this -- like all questions -- is easy. Although an awful lot of work goes into dressing up the analysis, the answer boils down to this: Executives cannot be overpaid, because what they are being paid must be the amount that they should be paid. Think I'm joking? If only.
The latest example of this nonsense showed up in a fawning news article by a New York Times sports reporter last week. Two economists (one of whom teaches at a prominent law school) wrote a typical economics article, in which they claimed to show that college football coaches' salaries are justified by the coaches' jobs being CEO-like. That is, they noted that big-time college football coaches run large organizations, in which they manage hundreds of people, and they are asked to deal with the media, politicians, the public, and so on. Moreover, they are supposedly held to account by the rigorous and unyielding performance standards of winning and losing.
(Note: Although I have written at length recently about college sports, this post is not concerned with any of those issues. I am using the Times article and the paper that it discusses to make a point about current practices and norms in academic economics. The football aspect is incidental.)
Once one has made the CEO comparison, however, it might occur to most people that the right way to think about this is through the well-established body of research that shows that executive pay is anything but market-driven. This research goes back decades, and it is based on both objective evidence and the testimony of many people who helped rig the executive pay game in the first place. Salary committees (at least those that are trying not to look too obviously like the CEO's puppets) set up nominally objective measures that, by design, push compensation inexorably upward.
The standard approach, widely discussed by everyone who follows such things, is to set up a game of leap-frog, comparing Company A's CEO's pay to that of Companies B, C, and D. If D's CEO gets a raise (based on, say, an increase in the share price for one quarter), then A's CEO can say, "But wait, I was the highest-paid CEO in this group, and now I'm not, even though I've done nothing wrong." B and C say something similar, and because it is all an inside job, "no" is not an acceptable (or even thinkable) answer. There is an upper limit, because there is a finite amount of money, but there is no logical connection between the quasi-equilibrium and "market fundamentals."
Moreover, there has been a large body of research indicating that the measures of "performance" that are used to measure CEOs' worthiness are deeply flawed. In organizations that sometimes employ tens or hundreds of thousands of people, all results (or at least all positive results, but sometimes negative results as well) are attributed to the top executive. "Creation of shareholder value" is said to be the doing of the dynamic chief executive. In the analogy to college football, wins and losses happen entirely because of the head coach.
These objections to the idea that CEOs are paid "what they are worth" are so well-known, in fact, that even NYT op-ed columnist Joe Nocera knows about them. In his column last Friday, he ran through most of the common objections to the conservative economic mantra on CEO pay. Of course, being a slave to the conventional wisdom, he indulged in a bit of academy bashing: "It’s amazing the things academics can find worthy of study, isn’t it?" Yes, the problem is woolly-headed professors, not the takeover of economics and related fields by a cult-like group of apologists for power. Right? But I digress.
When I say "even ... Joe Nocera knows" these things, I am referring to the obvious fact that Nocera frequently does not appear to know what he is talking about, even in his original field of journalistic expertise: business. He is an enthusiast, letting his emotions get ahead of facts and logic. That is not to say that he is always wrong. (He is no David Brooks or Ross Douthat.) For example, his newest column provides a reasonable discussion of "corporate inversions," which are a major tax policy issue. My point is that, if even someone as insistently unimaginative as Nocera knows about the arguments regarding executive pay, then those arguments must be reliably mainstream. (Nocera also spins out of control at the end of the column, failing to understand the cash flows in university budgets. But that is par for the course.)
That, however, does not stop our intrepid economics professors. Rather than acknowledge that there are competing theories within labor economics, they pepper their paper with references to a monolithic thing called "economic theory," which provides unambiguous answers that can be tested. From the abstract: "[Coaches] contracts’ features are consistent with what economic theory would predict." And from the second paragraph of the paper: "Yet economic theory tells us that lucrative compensation is not necessarily a problem if the structure of the employment contract is aligned with value creation." And in the conclusion: "In sum, we find that a close comparison of CEO employment contracts and college football coach contracts shows that both sets of agreements conform to the predictions of economic theory."
What is this unvarying economic theory to which the authors refer? They never quite say, because they apparently believe it to be obvious. The idea is that rational actors would not pay people more than they bring into the organization, because that would be irrational. (Circularity is a big part of modern economics.) Indeed, the final sentence of the article all but gives away the game: "Thus, if one believes that CEO compensation is set by the market at an appropriate level, and that employment contracts reflect this equilibrium, then one should reach the same conclusion about football coaches." (This sentence was reprinted in the NYT article that I noted at the beginning of this post.)
And should we believe that CEO compensation is appropriately set by the market? The authors clearly view this, too, as a trivial question. Indeed, they could more honestly have written: "Thus, because one should believe that CEO compensation is set by the market at an appropriate level ... ." For true believers, there is no "if" about it.
What is especially interesting here is that, as bad as the logical starting point might be, the analogy between CEO compensation and college coaches' pay is worse. At least in for-profit businesses, the notion of "maximizing shareholder value" can be defined with some logical coherence. Again, that does not mean that CEOs are actually paid according to a sensible definition of such value, but one can at least start from there to analyze the important questions. But other than inventing the term "program value creation," the authors have nothing useful to say about how to measure "value" in the college sports context. Maybe they are assuming that the money coming in from TV contracts is somehow tied to coaches' win-loss records (a truly bizarre idea); but how is the increased "value" of a big-time sports program meaningfully compared to what a coach does?
Indeed, the paper cannot (and does not try to) deal with the fact that most of the programs in question are drags on the parent universities' finances. It would be more sensible financially to shut down many of these programs and turn them into club sports or Division III programs without athletic scholarships or TV money. Yet, we can be sure that alumni and, in the case of public universities, legislators would quickly tell us that the true "value" of the program is much more than the TV, ticket, and memorabilia revenue. It is all about creating warm glow, building loyalty, encouraging applications, and so on.
What would "economic theory" tell us about that? Well, if we assume that CEO pay is set by the market, and if we assume that the college football market rationally and efficiently aligns coaches' pay with whatever people "value," then the answer is that the coaches are being paid what they should be paid. Assuming one's result is easy. What takes some skill, as this paper demonstrates, is making it appear as if one is doing anything more than merely justifying the status quo.