This post, an exchange between Andrew Hart, Marshall Steinbaum, and Daniel Markovits, continues their debate from our March 2020 series discussing The Meritocracy Trap. You can read all posts in the series here.
Hart & Steinbaum
It seems to us that the issue is not whether one places income in buckets labeled “capital” or “labor,” but rather what those particular buckets signify when it comes to extremely wealthy people. We might all agree to call the $50 million that a healthcare CEO gets for working 80-hour weeks “labor” income, but the fact that the firm or the “economy” has seen fit to allocate $50 million as a proper compensation for a healthcare CEO does not, as far as we can tell, have much to do with the productive value of 4,200 hours of healthcare CEO work over the course of a year. To justify this income by reference to skill is a just-so story—part of the inequality regime of “hyper-capitalism,” as delineated in Thomas Piketty’s recent book Capital and Ideology.
But Markovits seems to accept at least some of the human capital justification for high salaries when he speaks of superordinate workers and their immense skills and training. Put another way, we think Markovits believes the operative question is whether a person needs to work 80-hour weeks to get the $50 million as a healthcare CEO, and if the answer is yes, then the money is labor income. By contrast, we believe that the question should be why a healthcare CEO is “worth” $50 million in the first place, and that the answer to that question may at least cast some doubt on the usefulness of the category “labor income” when a person’s yearly income is high enough.
This ties in with our point about the book lumping together “meritocratic winners,” the high-paid service professionals for whom the analysis is largely persuasive, with “megamillionaires and billionaires,” for whom it isn’t. As far as we can tell, Markovits does not address this in his reply. This creates some confusion because high-paid lawyers, coders, bankers, etc. really do profit from selling their labor hours in a much more meaningful way than CEOs, hedge-fund managers, and the rest of the working ultra-wealthy do, even though, at face value, both seem to be making wages from similarly long and arduous work weeks.
It’s probably best to start at the end. Steinbaum & Hart and I agree that what I call “superordinate workers” neither EARN nor DESERVE their immense incomes. We disagree only, but extremely importantly, on the arguments needed to sustain this conclusion.
One of our disagreements concerns brute facts. I believe that the “high-paid lawyers, coders, bankers, etc. [who] really do profit by selling their labor hours” are much more prominent, and account for a greater share of top incomes overall, than Steinbaum & Hart do, and that “CEOs, hedge-fund managers, and the rest of the working ultra-wealthy” account for a smaller share. By way of illustration, while the average income of an S&P 500 CEO in 2017 was nearly $14 million, there were certainly many more than 500 lawyers and investment bankers who made $14 million that year.
Our deeper disagreement concerns how to characterize the income of the “working ultra-wealthy,” whoever they may be. While I accept that some of their income is not payment for labor in any meaningful sense, I nevertheless believe—and Steinbaum & Hart deny—that most of this income is payment for labor in just the way in which the lawyer’s income is. The reason for this, in my view, is that superordinate workers have induced innovations that transform the technologies of production in ways that make their particular skills in fact incredibly productive, as conventionally measured. I give detailed accounts of these transformations across economic sectors in The Meritocracy Trap and summarize them for the case of management in this article.
This is the essential point: the elite has remade the economy in its own image, so that conventional theoretical measures of productivity cast superordinate workers as “meriting” their enormous incomes. Much of the left, including Steinbaum & Hart, implicitly accept these familiar measures of productivity and therefore can object to elite incomes only by arguing (and I believe mischaracterizing) the facts. A better approach develops new norms to show that merit, so understood, is a sham.
Hart & Steinbaum
Professor Markovits claims that it’s reasonable to adduce high incomes of the meritocratic elite to their labor. Our preferred interpretation focuses on the ability of corporate power and the ultrawealthy to shape political economy through coercion. The concept of “inequality regimes” from Capital and Ideology is a helpful framework. Piketty’s work shows that arguing only over the persuasiveness of elite workers’ justifications for their high incomes misses a crucial question: where that money comes from.
Even the most well-paid lawyer, banker, management consultant, or coder is a wealthy service professional, whose earning potential depends on how much money their bosses or clients have to pay them. The value of these archetypical meritocratic winners is also contingent on the needs of capital, which increasingly shapes political economy to its whims in ways that affect the demand for the service professions such as law and consultancy.
There are also reasons to question Markovits’s theory of increasing returns to management technologies. Such an interpretation was put forward by the labor economist Sherwin Rosen in his 1981 article, “The Economics of Superstars.” But Piketty points out several empirical shortcomings: huge differences in top 1% and 0.1% labor income shares across countries (strongly related to the progressivity of their tax systems), when there’s no reason to believe management technologies are any different between developed economies. Also, that there’s no discernible productivity advantage to the incredibly rich over the merely rich, however slight (and there is one, in Rosen’s model—just not in reality).
Empirical labor economics offers up several more documented phenomena not easy to explain under the superstar interpretation: differential wage increases in firms that obtain valuable patent grants, for example, go to the employees whose names are on the patents—as against similar employees at other firms, and different employees at the inventors’ own firms. That pattern reflects more the importance of retaining control over valuable intellectual property than it does any inherent worker characteristics or accomplishments. There are also very strong gender inequalities in the way these raises are distributed—impossible to explain in Rosen and Markovits’s world, unless you believe women are less likely to benefit from advantages in managerial technologies.
Whether or not high elite incomes are justified is beside the point. The relevant question, which we can only answer if we understand how and why they exist, is whether they come at the expense of the rest of us. Given the evidence cited here and in our review and the interpretation it motivates, we think the answer is a resounding yes.
Steinbaum & Hart make one main point and two subsidiary but still important ones.
First, Steinbaum & Hart re-emphasize their claim that capital concentration drives rising inequality and that superordinate labor remains a second-order phenomenon. I don’t deny that capital matters, but I insist that superordinate labor matters more. In a typical recent year, the five highest-paid employees of S&P 1500 firms (7,500 workers overall) might capture 10 percent of the profits of the S&P 1500; and a typical investment bank disburses roughly half of its revenues (after interest paid) to its professional employees as wages. Labor now dominates even among the very richest: in the 1984, purely inherited fortunes outweighed “self-made” ones in the Forbes 400 by 10:1, but today purely “self-made” fortunes outnumber purely inherited ones; and the share of the 400 top incomes attributable specifically to salaries grew by half between 1961 and 2007. I don’t believe any of these labor incomes is earned, and The Meritocracy Trap develops a detailed argument why not. But this is not the patrimonial capitalism of the ancien régime.
Second, Steinbaum & Hart question the technological mechanism that I say lies behind the explosion of returns to superordinate labor. But their reference to Sherwin Rosen commits a synecdoche: Rosen merely identifies one mechanism by which superordinate incomes expand, and not the most important one. And their international comparisons are mistaken: Europe’s top managers, for example, increasingly capture incomes as great as their U.S. counterparts; Europe’s bankers, lawyers, and consultants are not far behind. Unfortunately, the inequality that The Meritocracy Trap documents in the U.S. is going global.
Third, Steinbaum & Hart suggest that my account of inequality cannot explain why the exploding top incomes have gone disproportionately to men. But The Meritocracy Trap expressly addresses this. When economic inequality is based on human capital, child-rearing—building human capital—becomes one of the central forms of economic activity. In a world that suffers gender subordination, elite women take this on—and lose wages for themselves.
All these disagreements share a theme. Steinbaum & Hart propose that the economic and political arguments that served the left well in fighting against prior inequalities will serve well again today. I believe that today’s inequalities follow unprecedented logics, which render them resistant to these familiar arguments. The Meritocracy Trap aims to develop new arguments, specifically tailored to our new circumstances. That these arguments implicate elite mind-workers, including in particular university professors and other social critics, in the inequalities that they deplore is, to my mind, not a strike against them but a mark in their favor.