scorecardresearch Skip to main content
opinion | Jonathan Schlefer

Economists’ long-held beliefs make income inequality worse

Jacob stead for the boston globe

Though many economists today are sounding the alarm over rising income inequality, one culprit somehow has been overlooked: their own wage theory.

Wage theory — one of the sacred truths of modern economics — suggests that competitive labor markets are self-regulating. Each worker is paid his or her productive worth. Unions, minimum wages, or any other interference — all just cause unemployment. Nearly all contemporary public policy is dictated by some version of this theory, but it simply no longer holds up.

Adam Smith, often called the father of classical economics, told a very different story. Smith believed that each society sets a living wage to cover “whatever the custom of the country renders it indecent for creditable people, even of the lowest order, to be without.” His successor David Ricardo similarly saw the “habits and customs of the people” as determining how to divide income between profits and wages. Marx’s class struggle was just a more confrontational version of the idea.

Around the turn of the 20th century, economists grew dissatisfied with this squishy sociologist’s answer, and some found it morally problematic. “The indictment that hangs over society is that of ‘exploiting labor,’” conceded John Bates Clark, a founder of the American Economic Association. He set out to disprove it.


Clark and other colleagues posited that firms shop for the best deal among “factors of production” — labor and capital — just as smart consumers shop for the best deal at the supermarket. Automakers, for example, could build cars by employing more workers and less machinery, or vice versa. By seeking the least expensive combination, the firms will pay only wages equal to a worker’s “marginal productivity” — the gain in output added when he or she was hired.

RELATED | Editorial series: Service Not Included

In this best of all possible worlds, output is maximized, and no willing worker is unemployed. How? Suppose workers want a job. If they offer to work for a bit less than the going wage — and if no unions or minimum wage law stop them from doing so — firms find it cost-effective to hire them.


The Great Depression did not look like the best of all possible worlds to the British economist John Maynard Keynes. He developed a second, rather nuanced theory that said capitalism only works well if entrepreneurs’ “animal spirits” for investing (that is, spending on production) are sustained alongside workers’ earnings and demand for goods.

After World War II, the American economy was managed according to Keynes’s ideas. General Motors and the United Automobile Workers would strike a bargain to raise wages in line with productivity gains. Other unionized firms would follow suit and raise their own wages — and so would non-unionized firms such as IBM in order to fend off organizing. Meanwhile, labor lobbied Congress for comparable minimum wage increases. The whole wage structure rose, sustaining consumer demand and assuring firms that if they invested in workers, they could sell their products.

There remained a little academic problem. The influence of both Clark’s and Keynes’s theories persisted, but they had nothing to do with each other. Then, in the 1970s, Robert Lucas of the University of Chicago brilliantly tore into at least American academia’s interpretation of Keynes and, with Clark as its basis, invented a whole new theory of booms and busts.


The specter of stagflation in the 1970s helped Lucas’s attack on Keynes. Inflation and unemployment rose, profits sank. Certain firms simply broke the law to stop unionization. After 18 labor proceedings against the Southern textile manufacturer J.P. Stevens, a US Court of Appeals ruled against the employer, blasting its violations as “flagrantly contemptuous.”

Unions sought legislation to make existing labor law more costly to violate, but even Democratic President Jimmy Carter gave the effort only lukewarm support, letting the bill languish in the Senate in 1978 until a filibuster killed it.

Amid this environment, policy makers looked back to the old J.B. Clark story. Carter launched deregulation, appointing the economist Alfred E. Kahn as his czar to run it. Kahn targeted airlines and said, in explicit reference to the Clark parable, “I really don’t know one plane from the other. To me, they’re all marginal costs with wings.”

The Reagan Revolution, further weakening unions and driving down minimum wages, brought surging income inequality.

Democratic economists — such as Paul Samuelson, also of MIT — gradually also turned against unions. In the 1976 edition of “Economics,” his seminal text, Samuelson concludes his discussion of them with a quote on their beneficial effects in establishing a “more orderly and defensible” wage structure. By the 1985 edition, he insists that if unions raise money wages, “the main impact is to begin an inflationary wage-price spiral,” and raising the real wage just “freezes workers into unemployment.” With enemies like this, how could Reagan fail?


Economists by the 1990s had discovered income inequality but, the Journal of Economic Perspectives noted, reached “virtually unanimous agreement” that technology was to blame. Advanced technology raised the marginal productivity of more skilled workers, so they earned more, and lowered the productivity of less skilled workers, so they earned less. The solution was more education, but since even a college education was doing little good, it was a throwaway. The women’s movement was making a difference, but it was about equity, not marginal productivity.

Clark’s theory is so well drummed into economists that they seem not to notice a fundamental problem. Firms do not have a significant choice among factors of production the way shoppers have a choice among foods at the grocery store. For example, in 2006, when Ford opened an auto plant in Chongqing, China, a spokesman said it was “practically identical to one of its most advanced factories” in Germany. Since Chinese wages were a faction of German wages, why not use more labor and less automation? Obviously, Ford had no idea how to. What would a crowd of extra workers actually do?

An old union joke puts the same point the other way around: “What’s the marginal productivity of an auto worker?” Answer: “It’s the steering wheel.” A firm can no more subtract a significant number of workers than it can add them. Economists who study the matter, such as Paul David of Stanford, conclude that production methods are essentially fixed. Innovation might find alternative methods. Then again, it might not. Either way, it is a profoundly uncertain exploration — not a market choice like smart shopping.


With an entire organization cooperating to produce goods or services, and no individual contributing any ascertainable productivity, we are back to Smith, Ricardo, and Marx. The habits and customs of the people or class struggle, call it what you will, determine the wage structure. Of course, there are limits. The sum of the slices of the pie — the profits and wages paid to different workers — cannot be bigger than the pie. But how to slice the pie is a fundamentally social decision.

In the 1970s, unions obtained real raises (despite inflation) that undermined profits. Such a situation hobbles capitalism. Business struck back, knocking most people’s wages down. Now pay has fallen too long and too far. The resulting chasm today equally hobbles capitalism. We as a society must solve the matter because markets will not.


Farah Stockman: Why income inequality threatens world order

Tom Keane: The real reason for income inequality

Editorial: Walking the walk on income inequality

Jonathan Schlefer, a researcher at the Harvard Business School, is the author of “The Assumptions Economists Make.”