In 1919 John Maynard Keynes, like many Europeans, looked back at the prewar era of free trade as a kind of golden age. “The inhabitant of London,” he recalled,

could order by telephone, sipping his morning tea in bed, the various products of the whole earth, in such quantity as he might see fit, and reasonably expect their early delivery upon his doorstep; he could at the same moment and by the same means adventure his wealth in the natural resources and new enterprises of any quarter of the world and share, without exertion or even trouble, in their prospective fruits and advantages.

In the wreckage of World War I, it was hard to imagine a return to this borderless “economic Eldorado.” But today, it’s the relatively self-contained national economies of the mid-twentieth century that may seem like a lost world. To access the products of the whole earth, you don’t even have to pick up the phone; you can just log onto Amazon.

Neoliberalism has never been totally hegemonic in trade, and its star has dimmed considerably since the 1990s.

This return to—and surpassing of—prewar levels of economic integration has been paralleled by a revival of pre-Keynesian ideas about the international economy. The vast expansion of international trade over the past forty years is often presented as the result of simply removing artificial constraints—that is, as a victory of “free trade” over “protectionism,” a realization of the cosmopolitan and liberal ideals of the nineteenth century after the aberrant nationalism and state direction of the economy of the twentieth. This victory is often claimed as one of the great successes of the neoliberal era, one whose benefits are so obvious as to hardly need stating. One of us recently attended a panel on trade at a meeting of the American Economic Association, where the chair opened the discussion by saying, “Obviously, if you are in this room then you are for free trade, as much as we can get.” No one in the room seemed to disagree.

The growth of trade and the deepening of financial links over the past several decades has gone hand in hand with the re-entrenchment of a rigid ideological vision of how the international economy functions, and of the external policies that national governments should be encouraged, or permitted, to follow—the so-called Washington Consensus. This consensus has never been absolutely hegemonic, however, and its star has dimmed considerably since its heyday in the 1990s.

While the neoliberal vision of trade has never lacked for critics, there is not one coherent opposing school. In thinking about alternatives to the neoliberal model, it’s helpful to distinguish two questions, each of which offers a different ground for opposition to the neoliberal vision. These correspond to two bodies of heterodox thought on trade which, while mutually consistent, are distinct arguments generally made by different people and on different grounds.

First, what kind of long-run pattern or structure do we imagine the world economy in terms of? That is, where do trade and financial linkages eventually lead? Second, how does trade adjustment happen in the short to medium run? That is, when the current pattern of trade departs from the long-run trend—or when countries’ international positions are inconsistent with each other or with domestic developments—how are they brought back into line?

On both these questions, the neoliberal consensus bypasses mid-twentieth century ideas about trade for an older orthodoxy—stretching back to the work of Adam Smith and David Ricardo—in which the goal of trade is specialization and market prices are a sufficient guide for cross-border exchange. This vision filtered through the free trade supporters of Victorian England and descended into twentieth-century economics through figures such as Eli Heckscher, Bertil Ohlin, Wolfgang Stolper, and Paul Samuelson. It had its most formidable proponents within political economy in the Mont Pelerin Society, especially Friedrich Hayek and Wilhelm Röpke. It was enforced politically by the World Bank and International Monetary Fund (IMF) of the 1980s and 1990s and derived its authority from the Reaganite and Thatcherite economists who held sway in the core economies.

Neoliberal theory fails as both a description of how international trade works and a prescription for how it ought to.

A recent Financial Times op-ed by former IMF chief economist Maurice Obstfeld offers a nice distillation of the orthodox position on both dimensions. It begins by dismissing the “an age-old fallacy” that “countries lose from trade unless their total exports exceed their imports.” On the contrary, he says, the trade balance has no effect on a country’s output or employment; rather, it passively “mirrors whether it spends less than its income or more.” Far from being a problem, trade deficits “can help countries finance productive long-term investments that ultimately raise national income and wealth.” Meanwhile, the specific pattern of trade (which countries are exporting what, and where) “reflects the international division of labour—based on each country’s comparative advantage.” Efforts to shift trade flows “can badly distort the international division of labour—thereby reducing the benefits all countries derive from trade.”

This all sounds very reasonable, especially if you’ve taken an economics class or two. (Obstfeld, as it happens, is also the author of a widely used textbook on international finance and trade.) But a little critical thought should raise some serious doubts about this story, as a description both of how international trade works and how it ought to.

In the rest of this piece, we will lay out what we see as the two main positions from which the neoliberal orthodoxy can be attacked. First, there is the standpoint of development, which argues that the test of the international order is not whether it allows countries to specialize according to their existing productive capacities but whether it creates space for the transformation of those capacities. Second, there is the standpoint of Keynesian macroeconomics, which rejects the distinction between international trade and domestic objectives such as full employment and questions the capacity of exchange rates and market mechanisms to balance payments between countries. After laying out these two arguments, we will briefly return at the end of the piece to prospects for the neoliberal order going forward.

Neoliberalism versus Development

The developmentalist critique challenges the idea that the existing patterns of trade and its benefits can be understood in terms of specialization according to comparative advantage. In a comparative advantage world, countries are endowed somehow or other with different productive capacities. Trade will mostly be between countries with very different endowments, since that’s where the gains from specialization are greatest.

Two hundred years ago, when Ricardo wrote down the first model of comparative advantage in terms of wool and wine, this vision had at least a superficial plausibility. Much trade then did involve primary products where differences in climate were decisive. But today most trade is in manufactured goods. In recent years, for example, the majority of Mexico’s exports to the United States have been vehicles, machinery and equipment, computers and auto parts; only about a tenth consisted of agricultural goods and oil, which depend on natural endowments. The theory of comparative advantage is harder to apply to manufactured goods, since their production does not depend on specific geographic factors in the way that agriculture and other primary products do.

On the developmental vision, the economic world is composed not of abstract individuals but of social organisms that require suitable conditions to grow.

The problem gets worse when we observe that—despite some spectacular exceptions—the world is still divided between a wealthy, industrialized North and poorer, less industrialized South. It’s not clear how specialization according to comparative advantage is supposed to explain this persistent divide. How can the lasting division of the world into a minority of rich industrial countries and majority of less developed ones reflect a rational division of labor, unless there is some deep and unchanging difference between the human beings who make them up? At one point, the suggestion was that poorer countries simply suffered from a lack of capital; free financial flows, it was promised, would bring the distribution of labor and capital into balance and close the gap between rich and poor. Few of neoliberalism’s promises have failed more clearly than this one, as financial liberalization has generally seen capital flow “uphill,” from rich to poor.

An earlier generation of neoliberals were more frank about where the international division of labor ultimately came from. As historian Quinn Slobodian’s notes in his magnificent book Globalists: The End of Empire and the Birth of Neoliberalism (2018), Röpke put it this way: “The rich countries of today are rich because, along with the necessary prerequisites of modern technology, they have a particular form of economic organization that responds to their spirit.” Industrialization in the South was impossible due to a “lack of punctuality, reliability, inclination to save and create.” Few contemporary advocates of free trade would justify the international division of labor in such naked biological or civilizational terms. (Nor did Ricardo.) But it’s hard to see how else to reconcile a comparative advantage story of trade with the persistent division of the world into rich and poor.

More sophisticated mainstream accounts of international trade recognize that little of it looks like Ricardo’s story of Britain exporting cloth and Portugal exporting wine. According to the New Trade Theory, trade mainly results from a combination of increasing returns and imperfect competition. Production of goods is concentrated in certain countries not because they are they necessarily have any relative advantage in producing them, but simply because concentrating production at one site leads to lower costs than smaller scale production in many places. This body of theory has some affinities with radical theories of underdevelopment; in one early paper Paul Krugman, who won his Nobel for work in this area, developed a model he described as “reminiscent of the Hobson-Lenin theory of imperialism.” But the shifting theoretical foundation of trade orthodoxy has had surprisingly little effect on its public pronouncements.

This is where the first of our two alternatives to neoliberalism enters. This vision makes national development, rather than the international division of labor, the ultimate goal of trade. This tradition draws on the historical record of catch-up industrialization in the United States, Germany, and East Asia. Its theoretical roots lie in the pragmatic and nationalist approach of Alexander Hamilton in the United States, Friedrich List in Germany, and the policymakers of the Meiji revolution in Japan. It extends through twentieth-century anti-colonial economists such as Raul Prebisch in the South and development economists such as Albert O. Hirschman in the North, to more contemporary heterodox thinkers like Alice Amsden, Ha-Joon Chang, and Anthony Thirlwall. It also draws on more radical Marx-influenced ideas about underdevelopment and imperialism.

In this vision, the economic world is composed not of abstract individuals but of social organisms—firms, industries, states—that require suitable conditions to maintain themselves and have the capacity to develop over time. The concept of development—borrowed originally from biology—was the master metaphor for much mid-century thinking on the international economy. For development theorists like Walt Rostow, economies were like tadpoles growing in a pond; one might metamorphose earlier than another, but given the right conditions all would end up as similar frogs.

Actually existing global capitalism, from the developmental view, is very far from the efficient, self-equilibrating system of neoliberal fantasy.

On this account, the central question about the international system is not whether it allows for optimal allocation of resources across borders but whether it provides a suitable environment for the survival and growth of these social organisms. Does it promote or hold back the productive capacities of nations? Actually existing global capitalism, from this view, is very far from the efficient, self-equilibrating system of neoliberal fantasy. International competition is red in tooth and claw; unmanaged, it is more likely to disrupt the development process of the weaker participants than to deliver mutual benefits.

If comparative advantage is a bad guide to how the greater part of trade is actually organized, it is an even worse one to how it should be. Think again about Obstfeld’s warning against distorting the international division of labor. (“Distort” is one of the many keywords that allows a narrow theoretical point to masquerade as a fact about the real world, by conflating a term’s technical meaning with its everyday one.) Intentionally or otherwise, this language suggests that it’s the proper role of some countries to make cars and computers, and the role of others to make clothes and coffee beans, and nothing should be done to change this. The countries that specialize in higher education and software and pharmaceuticals should retain their monopolies, while the countries that specialize in plantation agriculture and sweatshop clothing should keep on doing that. Everybody should stay in their lane.

The anti-development program of neoliberal trade is not just a matter of excluding the infant industry” protections that todays developed countries relied on when they were industrializing. The regime enforces the international division of labor in ways that go well beyond what we normally think of as trade. The outstanding example is intellectual property (IP). The increasing weight of IP provisions in todaytrade” agreements means they are no longer simply about setting rules for exchanges between countries. While the ideal of free trade closes off the possibility of transforming productive capabilities, IP rules prevent them from using even the capabilities they already have. They are also a reminder that the market power that the New Trade Theory takes as its starting point isnt just a fact about the world but something that has to be actively created and maintained. More broadly, IP protections, like the neoliberal trade regime they are part of, dont just reflect but actively enforce a global division of labor.

The experience of India shows how consequential these agreements can be. In 1972 the nation banned product patents in pharmaceuticals. At the time, medicine prices in the country were among the highest in the world, but critics of the ban warned that the country would lose access to imported medicines. In the decades that followed, however, India established a vast indigenous generics manufacturing industry and reverse engineered most state-of-the-art medicines developed elsewhere. Prices in the country dropped to among the lowest in the world, and by the turn of the century, Indian generic companies had become the largest supplier of affordable essential medicines outside the western world and the largest global supplier of generic medicines. Doctors Without Borders dubbed the country the pharmacy of the developing world.”

Intellectual property protections—like the neoliberal trade regime they are part of—dont just reflect but actively enforce a global division of labor.

The success of this industry was not predictable from standard narratives of export-oriented growth. This was not a case of low wage led industrialization; India did not have a comparative advantage in the labor, knowhow or raw material required for drug production. Instead, a combination of industrial policy, including early public investment, learning by doing as Indian pharmaceutical companies gained technical and technological expertise, a fortuitously large pool of scientists, and critically, no IP restrictions on the adoption of foreign technology combined to allow the country to become a low-cost producer. In theory, countries specialize in the things they are best at making. In reality, what countries are good at depends on what they make—or are allowed to.

India’s experience is far from unique. Ha-Joon Chang notes that historically, European and U.S. patent laws “accorded only very inadequate protection to the [intellectual property rights] of foreign entities. In most countries, including Britain (before the 1852 reform), the Netherlands, Austria and France, the patenting of imported inventions by their nationals was often explicitly allowed.” More recent industrializers followed the same playbook—Japan and Korea made extensive use of “creative imitation” to promote a whole range of industries and to generate enormous opportunities for indigenous companies to develop technological capabilities.

To be clear, these projects were not alternatives to international trade; on the contrary, trade was an integral part of them. Journalist Joe Studwell makes the plausible case that the only meaningful oversight the Korean state could achieve over firms like Hyundai or Posco was by testing them in the global market; and in the absence of that, there was no way to consistently pressure them to devote their surpluses to productivity-boosting investment.

This is a central irony of the neoliberal trade regime. On the one hand, the increasing importance of knowledge and organization relative to natural endowments has made it more realistic to imagine a system of trade yielding a global convergence to the material living standards and productivity enjoyed by today’s rich countries. But at the same time that this possibility has developed at the level of production, the commitment to treat knowledge as private property—and to the global division of labor as the touchstone of trade policy—has closed it off at the level of politics.

Neoliberalism versus Macroeconomic Management

If the exclusion of development is one central concern for critics of the neoliberal model of trade, the other is its claim that freely floating exchange rates are enough to balance payments between countries, without limiting their ability to pursue full employment or other domestic goals. On this front, the challenge for the neoliberal vision is not the division of the world into poor and rich countries, but rather the recurring currency crises of the neoliberal era—and the ways that domestic policies have been reoriented in response. While the development frame offers an alternative long-term vision of the international economy, this second line of criticism, drawing inspiration primarily from Keynes, focuses on the problems of the shorter term.

In the orthodox framework, there is no reason for trade to pose macroeconomic problems. Whatever labor and other resources are withdrawn from areas where a country does not have comparative advantage will be fully employed in the areas where it does. In modern terms, this means—as Obstfeld says—that a country’s GDP, employment, savings, and investment rates are determined solely by domestic factors; trade has no effect. As Paul Krugman put it in his 1993 essay “What Do Undergrads Need to Know about Trade?,” “The essential things to teach students are still the insights of Hume and Ricardo. . . . We need to teach them that trade deficits are self-correcting.” In particular, “employment is a macroeconomic issue, depending in the short run on aggregate demand and depending in the long run on the natural rate of unemployment, with . . . policies like tariffs having little net effect.” This essay remains unsurpassed as a statement of the neoliberal credo on trade. Krugman’s own ideas about trade have evolved since then, but the profession’s have not.

A second critique of neoliberal trade takes aim at the way it restricts national autonomy in the pursuit of domestic goals.

According to orthodox trade theory, the mechanism that makes deficits self-correcting, insulating growth and employment from trade, is a flexible exchange rate between currencies. When a country’s trade balance moves toward deficit, or becomes uncompetitive in certain industries, its currency will depreciate, making other industries more competitive, thus absorbing the displaced labor and bringing the country back toward balanced trade. As countries gradually abandoned the postwar system of fixed exchange rates after the breakdown of the Bretton Woods system in the 1970s, this theory has come to bear more weight.

At the start of the floating-rate era, the hope was that smoothly adjusting exchange rates would free countries to pursue full employment without being constrained by the balance of trade. Perhaps in a world where trade was the main economic link between countries, exchange rates would respond reliably to the trade balance. But at the same time fixed exchange rates were abandoned, so were controls on international financial transactions, and the volume of capital flows between nations soon came to dwarf trade flows.

In the neoliberal vision, this was not supposed to be a problem. In particular, free capital flows and floating exchange rates were supposed to work together to allow countries to combine deep international integration and pursuit of domestic economic goals. According to impossible trinity” or trilemma” models introduced by Robert Mundell and Marcus Fleming, countries could have any two of floating exchange rates, capital controls, and control over their domestic interest rate—but not all three. In the post–Bretton Woods era, they would give up capital controls, allowing financial transactions to take place freely across their borders. But by allowing their currencies to float, it was claimed, they could maintain control over their interest rates—and domestic economic policy more broadly—and thus maintain full employment and steady growth.

The theoretical models that this assertion was based on were always unrealistic, resting on strong claims about the perfect adjustment of exchange rates to their equilibrium values. But at the start of the floating-rate era, it was hoped that they would work as a reasonable first approximation. In this case, smoothly adjusting exchange rates would free countries to pursue full employment without being constrained by the balance of trade. These hopes have not been borne out. To be sure, countries with rising productivity and persistent surpluses, like Japan, have seen their currencies appreciate over time, while countries in the opposite position, like the UK, have seen theirs get weaker. But these shifts are far too slow and inconsistent to prevent major trade flows from spilling back into the domestic economy.

The result—most dramatic in poor countries, but evident in rich ones as well—is that exchange rates are dominated by financial markets. And in contrast to the smooth functions in the models, real cross-border financial flows are dominated by fickle market sentiment, which changes in ways that have nothing to do with the supposed fundamentals. When a currency is favored by the markets, it can remain unreasonably strong for long periods; alternatively, a sudden change of sentiment can cause its value to collapse overnight.

Neoliberals mistakenly thought that free capital flows and floating rates would work together to enable full employment.

Neither do trade flows respond to changes in exchange rate in the way that theory predicts. A staple of discussions of international trade is the “Marshall-Lerner condition,” which says that when a country’s currency devalues, or gets weaker, its trade balance should improve. But it is surprisingly difficult to find evidence that this is consistently true. As one survey of the empirical literature perplexedly acknowledges, “A typical finding in the empirical literature is that . . . that the Marshall-Lerner (ML) condition does not hold. However, despite the evidence against the ML condition, the consensus is that real devaluations do improve the balance of trade.”

Unless a country’s currency reliably weakens when it runs a trade deficit, and its trade balance quickly improves in turn, there is no automatic mechanism to ensure that employment in a sector lost to trade will be made up by employment somewhere else. Without reliable exchange rate adjustment, the logic that says trade must always leave a country better off as a whole no longer applies—and this is exactly what we observe. Exchange rates may have the right relationship to trade in the very long term, but countries facing balance-of-payments crises cannot wait that long. If direct restrictions on trade are ruled out, a country that has to close a trade deficit quickly—because foreign investors are no longer willing to finance it, say—has only one option: reducing domestic demand. Unlike exchange rates, domestic income (as measured by GDP) does have a reliable relationship with trade flows. Countries run trade surpluses in recessions and deficits in booms. So if you need to close a trade deficit quickly, a deep recession will get the job done.

This tool has been employed repeatedly over the past forty years in balance-of-payments crises all over the world, from Latin America in the 1980s to the Asian crisis of the 1990s and southern Europe in the past decade. Greece, for example, moved from a trade deficit of 12 percent of GDP in 2008 to essentially balanced trade five years later. Did its competitiveness improve? Not at all; Greek exports actually fell over this period. The trade deficit closed only because Greek imports fell by far more—almost half—thanks to a catastrophic depression.

In the absence of a market mechanism to balance payments between countries, it might seem that we are left with only two alternatives: delinking from the global economy entirely, or accepting that core domestic objectives like growth and employment will be hostage to the unpredictable whims of international financial markets. There is a third possibility, though: a more or less deliberate “surplus recycling mechanism” that generates payment flows back from surplus countries to deficit ones.

This term is associated with heterodox macroeconomist (and former Greek finance minister) Yanis Varoufakis, who argues persuasively that historically, periods of stable growth in the global economy have always depended on the existence of some such mechanism. A broader group of economists in the Keynesian tradition—Lance Taylor or Jane D’Arista, for example—have argued for the need for some kind of deliberate structure to ensure payments balance between countries, one that does not put the whole burden of adjustment on the countries facing deficits. An earlier generation of reform proposals focused on the idea of throwing “sand in the wheels” of cross-border financial flows with a tax on currency transactions, an idea first proposed by James Tobin in the 1970s. In the wake of the global financial crisis of a decade ago, there was a revival of interest in a more comprehensive system to manage to cross-border payments.

Exchange rates may have the right relationship to trade in the very long term, but countries facing balance-of-payments crises cannot wait that long.

Many of these proposals look back to the Bretton Woods arrangements (or to Keynes’s more ambitious proposals), but history offers a number of other examples. In the gold standard era, cooperation between central banks maintained payments balance without the need for catastrophic adjustments (in Western Europe, at least; in Latin America and elsewhere the experience was very different.) In the period after World War II, Marshall aid as well as the Bretton Woods institutions were meant to do this job; in practice, U.S. military aid also played a big part. In the European Union, the Common Agricultural Policy at one time served as a surplus recycling mechanism, sending substantial payments to the less developed Western European countries. So did the free movement of labor within Europe, which generated large flows of remittances from North to South. Critics of the euro who focus on the rigidity of the single currency as the root of Europe’s recent problems tend to ignore these mechanisms, exaggerating the role of flexible exchange rates in the earlier period of growth and convergence.

Today, there is no formal surplus recycling mechanism. But the U.S. trade deficit has performed a similar function, allowing countries elsewhere to accumulate foreign exchange reserves to protect themselves from turbulence in international capital markets. As Joerg Bibow has argued in several insightful articles, this “self-insurance” by middle-income countries in Asia and elsewhere can be seen as a kind of new Bretton Woods, allowing for controlled integration into the global economy that is consistent with domestic objectives for growth and employment. The success of this system of self-insurance was visible in the global financial crisis of 2007–2009: countries like Korea and Thailand experienced financial outflows considerably larger than in the “Asian crisis” of 1997, but unlike the grinding depressions that followed that earlier crisis, their massive foreign exchange holdings allowed them to escape this one mostly unscathed.

Reserve accumulation, and the trade surpluses it requires, are sometimes seen by U.S. critics as a violation of market norms, a form of currency manipulation or mercantilism. But they are better seen as a defensive response to the absence of any global management of international payments. There would be less need to run surpluses to accumulate foreign exchange reserves if countries took the more direct route of regulating financial flows across their borders with capital controls—but that is something that the neoliberal consensus has strenuously ruled out. Under the circumstances, persistent U.S. deficits play an essential role in maintaining a reasonably stable international trading order.

In both intent and effect, neoliberalism has closed off options for economic policy at the national level.

The incompatibility of the neoliberal trade model with stable growth and employment is most visible in crises, but it operates in normal times too. If faster growth leads to trade deficits, and if most countries cannot safely sustain large deficits, and if flexible exchange rates don’t solve the problem—well then, there is no choice but to keep growth slow enough that trade remains roughly balanced. This idea was formalized by the great post Keynesian economist A. P. Thirlwall as “balance of payments constrained growth,” and over the past forty years, he and his followers have made a strong case that it sets the speed limit on growth in much of the world. For the United States, as issuer of the world’s reserve currency, trade is not a constraint on growth. But for much of the world, surrendering the right to manage trade flows may prevent them from achieving economic growth rates they would otherwise be capable of.

Unlike the development critique, the Keynesian challenge to the neoliberal trade model is not concerned with which countries produce what, and the division between surplus and deficit countries does not correspond to the divide between North and South. But an important common thread between these two critiques of neoliberalism is their focus on the way that international integration limits political choices at the national level. Before World War I, adherence to the gold standard served as a “seal of approval” for countries aspiring to join the club of civilized nations. More recently, the euro has been explicitly defended as an effort to impose uniform, liberal policies across Europe. In general, it is striking how much classic defenses of free trade, from John Stuart Mill to Milton Friedman, focus not on economic benefits per se but on the ways free trade is supposed to promote good government.

The idea that deeper trade and financial links were desirable in part precisely because they reduced the autonomy of national governments saw a revival in the 1980s. World Bank chief economist Anne Krueger played a critical role in discrediting the idea of state-led development, on the grounds that any movement away from free trade inevitably leads to poor governance. It was Krueger who popularized the idea of rent-seeking as the critical problem for industrial policy—the notion that in practice politicians will pursue policies that don’t create wealth, but merely redistribute it to political insiders. The value of free trade, from this perspective, is not just its direct economic benefits, but as much or more the limits it imposes on predatory governments. In both intent and effect, then, neoliberalism has closed off options for economic policy at the national level—both development in the Global South and Keynesian demand management everywhere.

Neoliberalism Today

Ideologies, even dominant ones, do not always dictate developments in the real world. As in other areas, the history of international trade is subject to cross currents and eddies, with flows in one direction in one place and the other way elsewhere. It seems fair to say that through the end of the twentieth century the tide was running in the neoliberal direction. Under pressure from the IMF—but with cooperation from local elites—the developmental state has been rolled back around the world. In the decade after the 1997 crisis, for instance, Korea dismantled many of the structures that had directed credit and limited the role of foreign finance—a shift documented in an important series of articles by James Crotty and Kang-Kook Lee.

In every conflict between a national government and international markets, it’s the former that is on the side of public goods and regulation.

But if the debt crises of the 1980s and 1990s were largely resolved in favor of the neoliberal model, the picture looks more mixed today. Over the past decade, national governments have been less deferential to the Washington consensus. While someone like Obstfeld may stick to the old pieties in public, during his tenure the IMF’s research department showed a new openness to alternative approaches, particularly around capital controls. Ilene Grabel describes the current era as one of “productive incoherence”—a series of departures from the old vision that have opened space for new policies, without adding up to an alternative model. To the extent that tools like capital controls are seen as necessary but temporary distortions of an ideal liberal order, rather than natural elements of a different world economy, the ultimate goal is still to get rid of them.

The commitment to a rules-based, decentralized order has also showed signs of fracture. A prime example is the growth of preferential trade agreements. Once the floor for openness was established in 1995 by the World Trade Organization (WTO), the United States and its peers started to move away from the constraints of multilateralism and push for bilateral treaties. In an echo of his boss’s infamous “you are either with us or against us” ultimatum, Robert Zoellick, the U.S. trade representative under George W. Bush, divided the world into “can-do” and “won’t-do” countries and promised to work exclusively with the former.

While the WTO process has stalled, perhaps for good, there has been a proliferation of preferential trade agreements—over 700 according to a 2018 survey. These are bilateral and regional agreements that, in addition to offering favorable access for the signatories’ exports, typically include IP rules and other restrictions on domestic policy. These often go even further than the provisions of the WTO. For example, many allow for private arbitration so that firms can sue states for violations (whereas in the WTO, member states must bring disputes for arbitration) and cover areas well beyond those in traditional trade agreements (consumer protection laws, laws around public administration). China, of course, is the great exception, having integrated into the world market on its own terms. On this basis alone, claims of enduring neoliberal hegemony need, at least, a large asterisk.

The weakening of the neoliberal consensus does not automatically mean vindication for either of the alternative perspectives we have sketched here, of course. On the matter of surplus recycling, it is one thing to make the intellectual case for a world financial authority that would reliably channel funds from surplus to deficit countries, while preserving space for the pursuit of full employment at the national level. It’s another thing to imagine the political circumstances that could bring it into existence.

One may equally ask whether development is still viable or relevant as an alternative frame to neoliberalism. Many have doubts, on the left as well as in the mainstream. A recent article by Dissent editor Tim Barker, for example, heralds “The End of Development.” Export-led industrialization, in this view, is no longer a feasible route forward for countries of the South; services, meanwhile, offer too few opportunities for productivity-boosting investment. Other critics on the left worry that the goal of increased policy autonomy at the national level will bolster ethnic nationalisms—a view that gets some support from European politics today, where the greatest hostility to the constraints of the euro system is found on the chauvinist right.

These concerns are well taken. The past certainly does not provide a simple blueprint for the future, and economies are not really frogs in a pond. But the outlook for development does not seem as bleak to us as it does to these critics. China, again, is a glaring counterexample to claims that the project of export-led industrialization has stalled out. And other countries, like Bangladesh, continue to make good use of this strategy.

We do not see how the project of development can be abandoned.

More broadly, we do not see how the larger project of development can be abandoned. The problem of making a global economy consistent with meeting human needs is no less urgent today than in the past. National governments, for all their shortcomings, are the only place where the market is actually managed. Welfare states, credit policy, regulation, taxation and income transfers, public goods of all sorts—all are provided by national governments. We can imagine a world where supranational bodies performed these functions; fifteen years ago, the European Union might have seemed to prefigure it. But today, in every conflict between a national government and international markets or institutions, it’s the former that is on the side of public goods and regulation and the latter on the side of liberalization.

These considerations gain force when we think about the great problem of the twenty-first century: climate change. Far from making development obsolete, climate change gives it new urgency. Much like industrialization, decarbonization requires a rapid, far-reaching, coordinated reorganization of productive activity, a task that decentralized markets are particularly unsuited for. If governments in the North and South alike are going to rebuild their economies on a sustainable basis, they will need to use many of the same tools that were used to build new export sectors and shift labor and resources from agriculture to industry in the past. That means we will need an international order that leaves space for those tools to be used. Paradoxical as it might seem, it may be the global crisis of climate change that decisively shifts the center of economic authority back toward national governments.