Contrary to what many economic models suggest, salaries aren’t constantly recalibrated based on skills or technology. They follow the economy and politics—and common sense: hire when needed, promote from within, and slow hiring when budgets tighten.
Professor Lawrence Katz, of Harvard, is co-author with Professor Claudia Goldin, also of Harvard, of a book that defines the modern mainstream theory of labor markets. Entitled The Race Between Education and Technology, the Goldin-Katz book tells how supply and demand determine the relative wage of skilled labor. Put simply, when science and technology surge ahead, skilled wages and salaries rise, increasing inequality. When education catches up, wage differentials decline. This is a paradox of education: it is necessary to get ahead, but if too many pursue it, the advantage fades away.
The Goldin-Katz theory has been widely (if not always explicitly) used to justify the pay of economists, as mathematical theory and data analysis have become de rigueur. According to The New York Times of July 28, 2025, starting academic “base” salaries for economists now run about $150,000 per year, and $200,000 in consulting. For academics these figures do not include summer pay, research funding or other perks, nor the potential to supplement their incomes with outside work. New technologies, the theory tells, have relentlessly driven up the demand for economists’ skills, and because the supply of “top” PhD economists is necessarily restricted, the upward pressure on wages is what balances supply and demand.
But now, according to the headline, “the bull market in economists is over”. This is not, apparently, due to a surge in supply. Professor Katz reports that he graduated only seven new PhDs this past year. He did not say whether this was an increase over previous years, let alone a large one, but for a scholar at the top of his field it does not seem excessive. More generally, the Times specifies that the issue is not growing supply but declining demand, stating that “universities and nonprofits have scaled back hiring amid declining state budgets and federal funding cuts.” Of his seven new PhDs, Professor Katz was able to place only three in tenure-track positions.
The Times does not explain just why the hiring slump has hit “top” economists so very hard. It does not suggest any loss of confidence in the quality of their work or the viability of their theories. Indeed there is no hint, in the case of Professor Katz specifically, that the insight embodied in his theory of labor markets has fallen under suspicion. The Times merely refers to him, accurately, as a “prominent labor economist.” This if anything is understatement.
Another prominent labor economist, Professor Betsey Stevenson of the University of Michigan, did suggest to the Times that artificial intelligence may be playing a role. “The advent of AI,” she wrote, “is also impacting the market for high-skilled labor.” This is interesting because it would suggest a twist in the Goldin-Katz theory. AI is a massive advance in technology. If Goldin-Katz applies, that should increase the relative demand for highly-skilled labor, including PhD economists from top departments, assuring their full employment at even higher rates of pay. Yet, though new technologies supposedly raise demand for high-skilled labor, technologies also cut demand for low-skilled labor – and have been doing so at least since the invention of the wheel. Thus, if demand is falling, then there is the possibility that economists’ labor should no longer be classed as “highly-skilled.”
There is a deeper challenge in this story – to the basic theory of labor markets advanced by Goldin, Katz and their colleagues. The theory, like all neoclassical market theories, is a theory of price (and wage) adjustment. In this case, wages went up – to very high levels by academic standards – because demand outstripped supply. And now demand is falling. Should not wages and salaries fall too?
Why, in short, should the problem be one of unemployment, which hardly figures in the labor market model? Why shouldn’t new PhDs simply secure the jobs they want, but at modestly (or even, severely) lower rates of pay? These pay rates are negotiated individually – PhDs in economics do not usually belong to unions. Is there something wrong with the standard theory of labor market adjustment?
The standard theory acknowledges that for various unfortunate reasons – “institutions” is the catch-all term, covering unions, labor standards, the minimum wage, and the force of habit – wages and salaries can be “sticky,” generating unemployment that is sometimes termed “frictional.” But this is a transient phenomenon. The ultimate remedy remains a lower real wage. Perhaps after a year or two “on the market” Professor Katz’s unemployed students will be offered, and be willing to accept, salaries more customary in anthropology or English literature. But surely economists, of all people the greatest champions of free and flexible markets, are the in the best position to test the advantages of a rapid adjustment to the new intersection of supply and demand?
Of course, the option of lower pay does exist and will be taken. PhDs can drive Ubers, like anyone else. But this is not what the Goldin-Katz theory specifies. Theirs is a theory of wage adjustment in each class of labor, so that efficient markets assure full employment and so that valuable “human capital” does not run to waste. Yet it does not seem to occur to top economics departments to negotiate down their pay offers. Nor does it seem to occur to Professor Katz’s students, steeped though they are in his theory, to secure the jobs they covet by offering to do them for less. Still less does it occur to Harvard to cut the pay of a professor whose falling productivity is in evidence, for instance by a failure to place even half of his graduating students in the jobs they were led to expect when they enrolled.
I raise these points not to poke fun at Professors Katz and Goldin, nor at the eminent colleagues whose endorsements grace their book. Economics departments are not unusual. Practically no regular employer recontracts professional pay as suggested in the Goldin-Katz theory. They all set salaries, fill jobs as needed, promote talent from within, and cut hiring – not pay – when business conditions change (or budgets are slashed). The economics departments described by the Times are behaving normally. And Harvard is also behaving normally by not bothering Professor Katz with a salary cut. His employers may reasonably gamble that he will do better next year.
Professors Katz and Goldin are undeterred. Both were at Jackson Hole in August, where Katz presented his latest evidence on “skill biased technical change.” He was obliged to admit that the data have not been kind to this hypothesis for the past twenty-five years. Contrary to prediction, the premium separating college from high school – a key indicator in this work – has barely risen since 2000, suggesting (if the hypothesis were true) either that there has been an unseen glut of highly-skilled labor or an equally unseen decline in the rate of technological change. Neither appears supported by evidence worth speaking of, and Katz therefore describes this period as a “puzzle.”
The other possibility, amply argued outside the mainstream including by Thomas Ferguson and myself over twenty-five years ago, is that the theory wasn’t valid to begin with. Wage structures, we showed, are driven by macroeconomic and political events, which amply account for the variations across sectors and regions without reference to vague measures of “technology” and “skill.” I have further developed this argument in books and papers, including Created Unequal, published in 1998, and Inequality and Industrial Change, published in 2001. The advantage of these works over the mainstream is that they were developed from close analysis of a wide range of data, and not driven by a consuming desire to validate a preconceived model. The mainstream, alas, has the incorrigible habit of seeing only what it wants to see.
The amusing point – the moral of the story, if you like – is that the real world is finally moving against the purveyors of a fantasy theory of “labor markets,” who are wonderfully oblivious to the walls now closing in. It would be nice to believe that those making the hiring decisions are conscious that they are saving themselves unnecessary expense on miseducated people. But that would be, I fear, too much to expect.
- 1. Thomas Ferguson and James K. Galbraith, “The American Wage Structure, 1920-1947,” Research in Economic History, Vol. 19, 1999, 205-257.
- 2. James K. Galbraith and Maureen Berner, eds., Inequality and Industrial Change: A Global View, New York: Cambridge University Press, 2001. Spanish edition, DisigualdaBed y Cambio Industrial: Una Perspectiva Global, AKAL, Economia Actual, 2004. Translated by Sergio Cámara Izquierdo.
- 3. James K. Galbraith, Created Unequal: The Crisis in American Pay, New York: The Free Press, 1998. A Twentieth Century Fund Book. Paperback edition, University of Chicago Press, 2000.
- 4. James K. Galbraith, Dangerous Metaphor: The Fiction of the Labor Market, Policy Brief of the Jerome Levy Economics Institute of Bard College, 1997.
- 5. Claudia Goldin and Lawrence Katz, The Race Between Education and Technology, Belknap Press for Harvard University Press, 2008.
- 6. Lawrence Katz, Beyond the Race Between Education and Technology, Jackson Hole Economic Policy Symposium, Federal Reserve Bank of Kansas City, August 21-25, 2025.
