The world economy is entering a new phase, in which achieving global cooperation will become increasingly difficult.
The US and the EU, now burdened by high debt and low growth - and therefore preoccupied with domestic concerns - are no longer able to set global rules and expect others to fall into line.
Compounding this trend, rising powers such as China and India place great value on national sovereignty and resisting interference in others' domestic affairs. This makes them unwilling to submit to international rules - and thus unlikely to invest in multilateral institutions as the US did in the aftermath of the Second World War.
As a result, global leadership and cooperation will remain in limited supply, requiring a carefully calibrated response in the world economy's governance - specifically, slimmer rules that recognise the diversity of national circumstances and demands for policy autonomy.
But discussions in the Group of 20 leading and emerging economies, the World Trade Organization and other multilateral forums proceed as if the correct remedy were more of the same - more rules, more harmonisation and more discipline on national policies.
Going back to basics, the principle of "subsidiarity" provides the right way to think about global governance issues. It tells us which kinds of policies should be coordinated or harmonised globally, and which should be left largely to domestic decision-making processes. The principle demarcates areas where we need extensive global governance from those where only a thin layer of global rules suffices.
Economic policies come in about four variants. At one extreme are domestic policies that create no (or very few) spillovers across national borders. Education policies, for example, can be safely left to domestic policymakers.
At the other extreme are policies that implicate the "global commons": the outcome for each country is determined not by domestic policies but by (the sum total of) other countries' policies.
Greenhouse-gas emissions are the archetypal case. In such policy domains, there is a strong case for establishing binding global rules, since each country, left to its own devices, has an interest in neglecting its share of the upkeep of the global commons. Failure to reach agreement would condemn all to collective disaster.
Between the extremes are two other types of policies that create spillovers but need to be treated differently. First, there are "beggar-thy-neighbour" policies, whereby a country derives an economic benefit at the expense of other countries. For example, its leaders restrict the supply of a natural resource to drive up its price on world markets, or pursue mercantilist policies in the form of large trade surpluses, especially in the presence of unemployment and excess capacity.
Because such policies create benefits by imposing costs on others, they also need to be regulated at the international level. This is the strongest argument for subjecting China's currency policies or large macroeconomic imbalances such as Germany's trade surplus to greater global discipline than currently exists.
Beggar-thy-neighbour policies must be distinguished from what could be called "beggar-thyself" policies, whose economic costs are borne primarily at home.
Consider agricultural subsidies, bans on genetically modified organisms, or lax financial regulation. While these policies might impose costs on other countries, they are deployed not to extract advantages from them, but because other domestic-policy motives - such as distributional, administrative or public-health concerns - prevail over the objective of economic efficiency.
Over-ambitious and misdirected efforts at global governance will not serve us well at a time when the supply of global leadership and cooperation is bound to remain limited.
* Dani Rodrik, a professor of international political economy at Harvard University, is the author of The Globalization Paradox: Democracy and the Future of the World Economy // Project Syndicate