For all their differences, populists on left and right, both in the United States and in Europe, blame mainstream parties for creating a world of crisis, inequality, and insecurity. The story differs from country to country, but in the United States, the white working class, a traditional mainstay of the Democratic Party, propelled Donald Trump to the presidency.

The mainstream center-left might well have lost voters it would have otherwise claimed because it had grown too rich for its own good, as New York Times opinion-page contributor Thomas B. Edsall and others have argued. With widespread corporate support, endorsements from the likes of former Secretary of the Treasury Hank Paulson, and a majority among the richest voters, the argument goes, Democratic nominee Hillary Clinton paid too little attention to working-class whites or socialist-leaning youth.

But ideas about the economy also shape the directions of politicians and even how the rest of us conceive of our interests. In the late nineteenth century, British faith in laissez faire helped propel a great experiment in global markets and finance across much of the world. That experiment collapsed in World War I, struggled back in the 1920s, and failed for good amid the crises and inequality of the Great Depression. The socialist left and fascist right, not so unlike today’s populists, attacked the underlying faith in laissez faire.

In the decades after World War II, an essentially pragmatic belief in governments’ ability to create a better economy underpinned confidence in the welfare state. During his 1964 campaign, Lyndon B. Johnson promised a Great Society to end poverty—and won a larger majority of the richest five percent of voters than did Clinton in 2016. After a brutally contentious fight to pass the Civil Rights Act, which was filibustered for 60 days, he also won a majority of working-class whites.

Ideas about the economy shape the directions of politicians and even how the rest of us conceive of our interests.

Economic thinking shifted again amid the stagflation of the 1970s. British Prime Minister Margaret Thatcher and U.S. President Ronald Reagan, looking back to the nineteenth century, promised that the “magic of the market” would deliver prosperity. Their center-left successors—Bill Clinton in the United States, Tony Blair in the United Kingdom, Gerhard Schröder in Germany—signed on to neoliberalism as well: Markets, they said, with a little apology, are the best way to manage an economy. Governments might enlist some sensible regulation and limit markets’ harshest effects, but in general, they intervene at their own peril. 

Much of academia approved the market parable. Most economists who studied worsening income inequality in the 1990s shrugged it off as driven by inexorable market forces. The political scientist Adam Przeworski asked a question in one essay, “Could We Feed Everyone?” His answer: no. Socialism can fairly redistribute income but fails because it is inefficient. Capitalism is efficient but must embrace the market’s unfair income distribution.

Even as most macroeconomists and mainstream politicians adopted this market model as their baseline, some theorists were quietly tearing it apart. By 2003, MIT microeconomist Franklin Fisher and Jesus Felipe of the Asian Development Bank were warning economists that “bad habits and bad science breed bad economics and bad policy advice.”

Salt pans and dams on farmland near Perth, Australia, November 2016.
Salt pans and dams on farmland near Perth, Australia, November 2016.
David Gray / Reuters


To better understand the market model and its shortcomings, we must look back again. Economists Adam Smith and David Ricardo saw markets as playing important roles but society as determining the structure of income distribution. Each society establishes a sort of living wage, Smith argued, sufficient to purchase “whatever the custom of the country renders it indecent for creditable people, even of the lowest order, to be without.” More clearly, Ricardo saw “market” wages as fluctuating around a “natural” wage level. Varying greatly from place to place, that level “essentially depends on the habits and customs of the people.” Profits are just what is left to capitalists after they pay wages. Karl Marx turned the story about social custom into one about class struggle, but his core economic theory came straight from Ricardo.

Reacting against this “classical” view of economies—and supporting the late nineteenth century move toward laissez faire—a generation of “neoclassical” economists began trying to model how competitive markets alone might determine optimal profits, wages, and growth. William Stanley Jevons at University College London, Leon Walras at the University of Lausanne in Switzerland, and Carl Menger at the University of Vienna are particularly credited with initiating this turn. They employed Newton’s calculus and envisioned an economy like the harmonious universe of Newtonian physics. Jevons even proposed replacing “the old troublesome double-worded name of our Science”—political economy, as Smith and Ricardo had called it—with “the single convenient term Economics.” That name sounded more like the physics of the day. 

The parable evolved over the decades. In 1956, Robert Solow of MIT published the canonical macroeconomic model capturing the market parable in the Quarterly Journal of Economics. It describes how a quantity of capital, K, and labor, L, are converted into output, Y. Given the state of technology and individuals’ savings, perfectly competitive markets determine profits, wages, growth—all the big things—and do so optimally. The Solow model still underlies mainstream theories of phenomena ranging from long-run growth (the way Solow himself used it) to short-run business cycles. 

Of course, economists have always known that real markets are far from perfectly competitive. Unions or legislation push wages above their supposed equilibrium market level, and monopolistic firms jack up prices. Politicians run unsustainable deficits. Interest rates rise too high, causing recession, or fall too low, causing inflation. Mainstream macroeconomics is largely an exercise in adding market “imperfections” and other ancillary assumptions to an underlying Solow or similar model to try to approximate real economies.

Economists have always known that real markets are far from perfectly competitive.

There were dissenters. Even before Solow published the canonical version of the model, Joan Robinson of Cambridge University, who had developed principles of imperfect competition still featured in textbooks, had the temerity to ask: What is a quantity of capital, K? Of course, capital is machine tools, corn seed, software. But how do you measure it? In what unit?

A quantity of labor, L, might be roughly gauged by the number of workers. It can be argued that shoppers’ trucking and bartering, as they try to maximize their famous utility, determine the relative values of consumer goods. After markets settle down, perhaps an orange trades for ten milligrams of gold, a bottle of shampoo for twenty. In effect, consumer goods could be valued as quantities of gold—or any other commodity. 

Capital is trickier. Capitalists have no inherent preferences for machine tools versus corn seed but only for the profits they earn. To hint at Robinson’s point, you must know the profit rate to determine a value for K. But then how can you plug K into the model to determine, among other things, the profit rate? The logic seems circular. The model does not provide a coherent account of capital—the very core of capitalism.

Robinson’s critique triggered what was called the “Cambridge capital controversies.” Raging on the pages of top journals for a decade, they pitted her and her Cambridge University allies against Solow, MIT economist Paul Samuelson, and their allies in Cambridge, Mass. Cambridge University won the intellectual battle. Samuelson conceded in 1966 in the Quarterly Journal of Economics: “If all this causes headaches for those nostalgic for the old time parables of neoclassical writing, we must remind ourselves that scholars are not born to live an easy existence.”

But MIT won the war for the profession. Although Solow was content to relegate models like his own to playing a role as “merely cheap vehicles for interpreting data,” Samuelson shifted ground, claiming that Robinson’s critique was merely technical and that the neoclassical parable still works well in practice. Cambridge University theorist Frank Hahn, who did not participate in the debates, argued: “On purely theoretical grounds there is nothing to be said in its favour. The view that nonetheless it ‘may work in practice’ sounds a little bogus and in any case the onus of proof is on those who maintain this.”

Today few economists have the remotest idea what happened. Macroeconomist Paul Romer wrote a few years ago in the American Economic Review: “Economists usually stick to science. Robert Solow was engaged in science when he developed his mathematical theory of growth. But they can get drawn into academic politics. Joan Robinson was engaged in academic politics when she waged her campaign” against the Solow model.

A Ford India assembly line in Chengaipattu, March 2012.
Babu Babu / Reuters


Robinson’s was not the first critique of the utopian market model, and later critiques strengthened it. But she identified the core problem. In depicting how capital K and labor L combine to produce output Y, the Solow model reduces a whole economy to a single market writ large. This metaphor appeals to our poetic fancy, and a single market can be modeled, with a straight face, as a harmonious Newtonian world. But an economy with many actors and goods comprises the very opposite: a complex system. Nineteenth-century neoclassicals did not even have a word for it.

The substance of the Cambridge critique therefore deserves a look. Both sides accepted the famous supply-and-demand story in a market for a single good. Demand increases, and the price goes up. Supply increases, and the price goes down. The question was what happens if there are many capital goods. A simplified approach to answering that question, taken by Samuelson in his 1966 concession article, supposes that capital goods are produced with labor alone, thus ignoring for simplicity the need for existing capital goods and natural resources. If the model fails despite such a wild simplification, it fails even worse in a more realistic case. 

The value of capital goods cannot be the pure labor needed to produce them because they take time to make, and profits accumulate over time. Competitive markets are supposed to guarantee a uniform economy-wide profit rate (another gross simplification). Say the profit rate is ten percent. Thus, if a supplier used ten units of labor to build a machine the previous year, then the supplier sells it for ten units of labor plus a ten percent profit, or eleven units of (current) labor. Of course, more profits accumulate if the labor had to be applied even earlier.

Now fancifully suppose (still following Samuelson) that one winemaker buys a capital good from a supplier who produced it with seven units of labor two years earlier. A second winemaker buys one capital good from a supplier who produced it with six units of labor a year earlier and another capital good produced with two units of labor three years earlier.

An economy with many actors and goods comprises a complex system. Nineteenth-century neoclassicals did not even have a word for it.

Which winemaker’s capital equipment has a greater value? It depends. At a low profit rate, the second winemaker’s two capital goods cost more, at eight units of labor plus just a little profit. At a higher profit rate, the first winemaker’s capital good costs more—now, greater profits accumulate over two years on the seven units of labor needed to build it. But at a still higher profit rate, the second winemaker’s capital goods cost more again—now, yet greater profits accumulate over three years on the good made with two units of labor. (You can trust Samuelson’s arithmetic or check it in Excel.) The relative value of different sets of capital equipment can keep “double switching.” No fixed gauge of their value exists.

This point is fundamental. The “enormous who’s who” of capital goods, as Robinson put it, simply cannot be evaluated until the profit rate is known. Since one must know the profit rate to determine the value of an economy’s capital stock, K, one cannot plug K into the model to determine the profit rate. Trying to do so would be circular nonsense. 

Where does the profit rate come from, then? Though debated on technical grounds, the Cambridge capital controversies were ideologically charged. Robinson’s answer: not markets but social custom, as Smith and Ricardo said, or class struggle, as Marx said. Income distribution depends on social and political decisions.

Still, was Samuelson right that the Solow model works well enough in practice? Fisher drew on a so-called aggregation literature to investigate the question. This literature posits an economy with many firms of the type assumed in standard microeconomics and asks if it can be summarized in a simple Solow model or some similar model. The answer Fisher reached: only if you make ridiculous assumptions—for example, if every firm produces the same output at different scales, such as one barn and one legal brief, or several barns and legal briefs, per year. 

Fisher then tested a strategy that has been used by more macroeconomists more times than could ever possibly be counted. Assume an economy with diverse firms (of the standard microeconomic sort) is represented by a Solow model (even though the model is a distortion). Draw statistical conclusions based on the Solow model. How badly do those conclusions distort conclusions based on a true model? The answer: very.

Economist Robert Solow (second from left) with other Nobel Prize recipients, October 1994.
Economist Robert Solow (second from left) with other Nobel Prize recipients, October 1994.
Paul McErlane / Reuters


If you think about it, even at first glance, the Solow model looks doubtful. Embodying the usual supply-and-demand story, it says that as a nation accumulates a greater supply of capital per worker, the profit rate falls. But as the U.S. economy accumulated far more capital per worker over the past half a century, the profit rate rose. Lance Taylor of the New School quips: “This history is consistent with the Solow model running in reverse.”

As a last resort, defenders of the simple market parable claim it is just shorthand for a rigorous “general equilibrium” model of markets with many individuals and goods. Theorists Kenneth Arrow and Gerard Debreu published the best-known such model in 1954. Both Robinson and Fisher proved, in effect, that the simple market parable cannot be a valid shorthand: it cannot validly summarize a diverse economy. And then general-equilibrium research itself drove the final nail into the coffin of the utopian market parable.

Making a dozen assumptions about human behavior and technology, Arrow and Debreu developed a brilliant proof that their economy always has some “equilibrium” set of prices and wages. At those levels, the amount of each good freely supplied by firms and the amount of labor freely offered by each worker would just equal the amounts freely demanded. All would be for the best in this best of all possible worlds. So it would seem.

One problem is that there can be multiple equilibria. In effect, demand curves can wiggle up and down like snakes in a hurry, as Australian economist Steve Keen put it, allowing many possible prices and wages. But the fatal problem is that even those “equilibria” are unstable, as three papers published in the 1970s, one by none other than Debreu, concluded. Markets would never actually lead an economy to any such equilibrium—optimal, dismal, or otherwise. And should an economy happen to touch there, markets would yank it right away. In short, markets cannot stabilize an economy or even determine profits and wages, whether well or badly. Hahn warned Thatcher about those problems to no avail.

How then how are such matters settled? Since markets alone cannot do it, politics and society must somehow step in. Perhaps force of convention, with some help from Federal Reserve Chair Janet Yellen on a good day, keeps the economy on a fairly even keel, except when it isn’t. Markets cannot decide that some CEOs should be paid several hundred times more than average workers, as they are today, rather than a few dozen times more, as they were in the post-World War II decades. Executive compensation committees, labor law, and a host of other political and social institutions somehow determine the matter.

Even where populist protest is cultural, it is fueled by way of mainstream neoliberalism.

Of course, nothing says that society will make these decisions well. In the 1970s, as unions pushed up wages and firms jacked up prices, they drove an inflationary spiral. Perhaps worse, wage pressure on top of oil shocks pushed profit rates down from about 20 percent to ten percent across the advanced nations, according to Andrew Glyn of Oxford and other authors in an article about why economies did well earlier but badly during that decade. Capitalists do invest to make profits, so this collapse surely tended to sap growth. Without quite knowing why, Reagan and Thatcher may have exhibited some implicit sense in attacking labor.

Decades later, again without quite knowing way, populism warns that those policies are worse than exhausted. Of course, populist protest is only partly economic. But even where that protest is cultural, it is fueled by way of mainstream neoliberalism, sanctimoniously pronouncing its policies to be based on scientific economics. Yet populists’ views about the economy are, to put it kindly, often ambiguous. Tea Party activists generated rage against the Affordable Care Act as a big-government imposition on them and a benefit to undeserving others—until Trump tried to repeal it and many noticed that it provides their health insurance. Some version of national U.S. health insurance looks like it is here to stay as a pragmatic welfare-state institution (like Social Security) that is meant to benefit all hard-working taxpayers. 

Populism might be interpreted more consistently as holding that society, rather than government per se, can and must circumscribe the economy within its compass. Hence the opposition to trade treaties. This view is not obviously wrong. “Ideas, knowledge, science, hospitality, travel—these are the things which should of their nature be international,” John Maynard Keynes wrote. “But let goods be homespun whenever it is reasonably and conveniently possible, and, above all, let finance be primarily national.”

The North American Free Trade Agreement launched a far more extensive international economic regime than Keynes even imagined. It focuses less on trade per se—by the time it was negotiated the United States had already lifted nearly all obstacles to manufactured imports—than on protecting investors’ property rights and thus supporting their ability to move across borders. Of course, a race to the bottom runs into obstacles. For example, where labor is cheap, governments may be unstable, infrastructure poor, and crime endemic. Still, increasing international flows of capital, enshrined in treaties, and flows of labor, often via undocumented immigration, cannot but erode each society’s ability to determine income distribution according to “the habits and customs” of its people.

Whatever the original pros and cons of these international economic regimes, shutting them down and uprooting established trade and investment is profoundly dangerous. Beggar-thy-neighbor policies just raise employment at home at the expense of the whole world, as none other than Robinson prominently argued. (She called them “beggar-my-neighbour” policies.) In the face of opposition on many sides, after six months in office, Trump’s promised slashing of NAFTA has narrowed to some proposed whittling.

Sanders, on the other side, would raise minimum wages and better regulate Wall Street. Financial regulation is eminently consistent with general-equilibrium theory’s findings of market instability. Instability does not even have to be financial, but finance is often the worst culprit. His proposal to raise wages makes economic sense, too. If every fast-food chain pays $7.50 an hour, no single outlet can pay $15. But since the lowest-paid workers generally work in services that face no international competition—nobody travels to Mexico to buy hamburgers—and the market parable of wage determination is bunk, legislation can usefully establish a more equitable wage structure. On the one hand, the 1970s warn against too-large rapid moves. On the other hand, since it took half a century to reach our current nadir of income inequality, efforts to make distribution fairer must be persistent.

Like Plato’s republic ruled by philosophers, the market parable is dangerously alluring. It demands policies that strive for a utopian future—a future we can never approach because it lacks any reality—and meanwhile delivers disaster. Economies are so complex that we do not know much about them, but we know at least one thing. From the late-nineteenth to the late-twentieth centuries, the best economic minds tried to make the market parable work and failed by their own criteria. It does not work. Without quite knowing why, populists on left and right, sometimes reactionary, sometimes quixotic, demand that society step in to create a fairer economy. Until serious politicians accept that responsibility, we are in for a long, dark night. 

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now