AFTER the second world war, the leaders of the Western world tried to build institutions to prevent the horrors of the preceding decades from recurring. They sought to foster both prosperity and interdependence, to “make war not only unthinkable but materially impossible”.

Their work has borne fruit. There has been no armed conflict in western Europe since.

Expanded global trade has raised incomes around the world. Yet, as the Brexit vote demonstrates, globalisation now seems to be receding. Most economists have been blindsided by the backlash. A few saw it coming. It is worth studying their reasoning, in order to work out whether a retrenchment is inevitable or might be avoided.

Even economists realise that free trade can be a hard sell politically. The political economy of trade is treacherous: its benefits, though substantial, are diffuse, but its costs are often concentrated, giving those affected a strong incentive to push for protectionism. Since 1776, when Adam Smith published “The Wealth of Nations”, those pressing for global openness have won more battles than they have lost. Yet opposition to globalisation seldom disappears, and often regroups. And a position once considered near-heretical, that globalisation itself seems to create forces that erode political support for integration, is gaining currency.

Dani Rodrik of Harvard University is the author of the best-known such critique. In the late 1990s he pointed out that deeper economic integration required harmonisation of laws and regulations across countries. Differences in rules on employment contracts or product-safety requirements, for instance, act as barriers to trade. Indeed, trade agreements like the Trans-Pacific Partnership focus more on “non-tariff barriers” than they do on tariff reduction. But the consequences often run counter to popular preferences: the French might find themselves barred from supporting a French-language film industry, for example.

Deeper integration, Mr Rodrik reckoned, will therefore lead either to an erosion of democracy, as national leaders disregard the will of the public, or will cause the dissolution of the nation state, as authority moves to supranational bodies elected to create harmonised rules for everyone to follow. These trade-offs create a “trilemma”, in Mr Rodrik’s view: societies cannot be globally integrated, completely sovereign and democratic—they can opt for only two of the three. In the late 1990s Mr Rodrik speculated that the sovereignty of nation states would be the item societies chose to discard. Yet it now seems that economic integration may be more vulnerable.

Alberto Alesina of Harvard University and Enrico Spolaore of Tufts University presented a different but related view of the trade-offs entailed by global economic integration in “The Size of Nations”, published in 2003. They note that there are advantages to being a large country.

Bigger countries can muster more resources for national defence, for instance. They also have large internal markets. But bigness also carries costs. The larger and more heterogeneous a country, the more difficult it is for the government to satisfy its citizens’ political preferences.

There is less variation in political views in Scotland, to take one example, than across Britain as a whole. When policy is made by the British parliament (rather than in Edinburgh, Belfast and so on) the average Briton is slightly less satisfied with the result.

Global integration, Messrs Alesina and Spolaore argue, reduces the economic cost of breaking up big countries, since the smaller entities that result will not be cut off from bigger markets.

Meanwhile the benefits of separatism, in terms of being able to cater better to the preferences of voters, are less diminished. So the global reduction in barriers to trade since the second world war, the pair contend, at least partly explains the simultaneous growth in the number of countries, even if national fractures often involve, or lead to, political instability and violence.

And then there is the question of how the benefits of globalisation are shared out. Joseph Stiglitz, a Nobel prizewinner, has warned that rent-seeking companies’ influence over trade rules harms workers and erodes support for trade liberalisation. Raghuram Rajan, the head of India’s central bank, has argued that clumsy government efforts to compensate workers hurt by globalisation contributed to the global financial crisis, by facilitating excessive household borrowing, among other things.

David Autor, David Dorn and Gordon Hanson have documented how the costs of America’s growing trade with China has fallen disproportionately on certain cities. And so on.

Open and shut
 
Branko Milanovic of the City University of New York believes such costs perpetuate a cycle of globalisation. He argues that periods of global integration and technological progress generate rising inequality, which inevitably triggers two countervailing forces, one beneficial and one harmful. On the one hand, governments tend to respond to rising inequality by increasing redistribution and investing in education; on the other, inequality leads to political upheaval and war. The first great era of globalisation, which ended in 1914, gave way to a long period of declining inequality, in which harmful countervailing forces played a bigger role than beneficial ones. History might repeat itself, he warns.

Such warnings do not amount to arguments against globalisation. As many of the economists in question are quick to note, the benefits of openness are massive. It is increasingly clear, however, that supporters of economic integration underestimated the risks both that big slices of society would feel left behind and that nationalism would continue to provide an alluring alternative. Either error alone might have undercut support for globalisation—and the six decades of relative peace and prosperity it has brought. In combination, they threaten to reverse it.