International policy coordination is like the Loch Ness monster: much discussed but rarely seen. Going back over the decades, and even further in history to the period between the Great Wars, coordination efforts have been episodic.
Coordination seems to occur spontaneously in turbulent periods, when the world faces the prospect of some calamitous outcome and the key players are seeking to avoid cascading negative spillovers. In quieter times, coordination is rarer--though not unheard of; the Louvre and Plaza accords are examples.
Today, policy coordination has resurfaced as a hot topic: while the worst of the global financial crisis is behind us, no one would claim that a return to "Great Moderation" is in the cards, and policymakers around the globe appear worried about policy transmissions across many dimensions.
Views on the size of cross-border policy spillovers have evolved over time, but today no one doubts that we live in an interconnected world. In the 1980s, the literature often concluded that these cross-border effects were small, but more recent evidence suggests that spillovers are sizable, reflecting the increase in trade and financial integration. And spillovers are generally larger in turbulent times.
The externalities associated with cross-border spillovers also reflect a paucity of policy instruments relative to targets--which means that it is difficult, not to say impossible, for a country to inoculate itself against the cross-border policy transmission. If the number of instruments equals the number of targets, cooperative and non-cooperative outcomes will be the same, and there will be no gains from international coordination. The legacy of the global financial crisis--high public debt, near zero interest rates, and at times what looks like domestic political dysfunction--suggests that nowadays policymakers have fewer policy tools to achieve their manifold objectives. In such circumstances, gains from policy coordination across countries are likely to be larger than during the Great Moderation.
Coordination is seen, occasionally. The global fiscal stimulus enacted in the early days of the financial crisis is a case in point. The fiscal expansion, exerting a positive cross-border spillover (and thus an insufficient stimulus in the non-cooperative equilibrium), is generally seen to have raised global welfare. There may have been other factors at work at the time as well: the stigma of rising public debt may be less when everyone is engaging in expansion than when only some countries are doing so. Other examples of coordinated policies during the crisis include the pursuit of tax havens and the commitment to eschew protectionism and competitive depreciation to support aggregate demand.
But there may be cases where countries don't pursue the potential gains from coordination. A case in point is unconventional monetary policies. The major currency areas (the U.S. Fed, the European Central Bank, and the Bank of Japan) did not coordinate their monetary expansions, which reflected domestic objectives. Yet spillovers from such policies are likely to have been large, through both trade and financial flows. What might account for the international community's failure to pursue the potential gains from coordination?
In related work, Ostry and Ghosh emphasize the role of uncertainty and disagreement on the size and even the sign (that is, positive or negative) of cross-border policy effects. Take the simplest case of a monetary expansion, which raises demand for exports in the rest of the world (positive spillover) but appreciates the currencies of other countries (negative spillover). Source and recipient countries may have quite different views on the size, or even the sign, of the net spillover (output plus exchange rate effect), as the debates surrounding "currency wars" showed. These debates are nothing new, but uncertainties may be larger now: the policy instruments are unconventional and their effects less certain.
Uncertainty makes it harder to negotiate and sustain a coordinated outcome. Country A may claim its policies help the rest of the world, while country B may claim harm. As a negotiating tactic this may be sensible, but it is hardly a recipe for arriving to the cooperative outcome. The paper by Ostry and Ghosh also discusses other impediments to realizing the gains from coordination, including a lack of focus on domestic policy tradeoffs (such tradeoffs are indeed a necessary condition for successful coordination) and inherent asymmetries in the global economy.
What can we learn both from the successful coordination cases and the less successful ones? First, there have to be large and clear gaps between cooperating and not cooperating to overcome the tendency of countries to solve their problems unilaterally. And it must also be clear how each player can contribute to the global outcome. When coordination implies that all players have to do the same thing (say, expand fiscal policy, or refrain from competitive depreciations), coordination is likely to be easier to explain and to achieve. When the quid pro quo is more complex, coordination is likely to be more difficult and may just not happen--even if on paper the gains are large.
Can something better be done? There would seem to be a role for a 'neutral assessor' to clarify the nature of spillover effects and thus make it easier to establish and sustain cooperative outcomes. The IMF already plays a role as an assessor through its country surveillance and, by recognizing the increased interconnectedness across countries, it is playing a greater role in identifying spillovers. Here it can help identify policies that would be in both the national and global interest.
Is there more that could be done? There may be times when the scope for mutually advantageous policies simply is not there: for example, when a large country's policies exert adverse spillovers across a swath of smaller countries but the actions of the smaller players do not materially affect economic outcomes in the large country.
Would some guideposts or rules of the road (analogous to the rules that originally underpinned Bretton Woods--e.g., against competitive depreciation) help to limit the scope for adverse spillovers be desirable in such cases? The logic of such rules seems compelling. The guideposts could take the form of warnings, or stronger measures, when current accounts deviate too much from current account norms, or when some aspects of a country's capital account become too unbalanced, contributing to credit or asset price boom-bust cycles abroad.
Our sense is that, like the Loch Ness monster, international macro policy coordination may continue to be heard about more often than it is seen. There are important obstacles to reaping the gains from coordination. To help overcome those, the circumstances when gains are likely to be large need to be identified and a push made for specific coordinated outcomes in such cases. And when the gains from coordination are not there, this needs to be acknowledged. But even in such cases, if large players in the global economy are responsible for significant adverse spillovers across a swath of smaller countries, this needs to be acknowledged as well, and feasible remedies considered.