A quarterly magazine of the IMF

March 2003, Volume 40, Number 1

Trying to regiment coordination of dollar, yen, and euro monetary policy isn't worth the risks and costs

AT THE END of the 2001 Academy-award winning film, A Beautiful Mind (based on Sylvia Nasar's biography of 1994 Nobel Prize winner John Nash), we are told that one of the major applications of Nash's theory of strategic interaction is to international economics. Admittedly, the film never actually depicts Nash being consulted by the U.S. Federal Reserve Chairman, in fact or in fancy. And the scene used to illustrate Nash's theory of cooperation takes place in a bar rather than a central bank boardroom. Nevertheless, that international economic policy coordination is even mentioned in a mass-marketed film shows how far the topic has penetrated the public arena.

Certainly, cooperation among the world's major central banks is something the public generally favors. It is reassuring that the U.S. Federal Reserve, the European Central Bank, the Bank of Japan, and other major central banks regularly exchange information, macroeconomic assessments, analysis, and policy ideas. But is this enough? Some think not. In our increasingly globalized world, where Group of Three (G-3) monetary policy decisions have potentially huge spillover effects around the world, don't we need a multinational mechanism to arbitrate and coordinate interest rate policies?

That isn't a new idea. Ever since the shift to floating exchange rates in the early 1970s, many leading thinkers have tried their hand at developing grandiose mechanisms and institutions for guiding global monetary policy. These range from Stanford University economics professor Ronald I. McKinnon's 1984 proposal for having the G-3 set world money targets to proposals for a global central bank. But just how big are the potential gains? And are they worth the risks and transition costs? My perhaps heretical answer is no. To my mind, the upside to grandiose G-3 monetary policy cooperation schemes is quite limited, no matter how well they are designed and no matter how seamlessly they are implemented. The upside is small, that is, compared with the overall potential benefits that would flow from having the G-3 central banks simply follow good domestic monetary policies.

Why, then, despite increasing globalization, isn't there scope for more structured and institutionalized forms of international monetary policy coordination? Over the past 20 years, researchers have studied this question, trying to quantify the benefits of idealized international cooperation in the sense of jointly setting interest rate policy. Some have combined old-style Keynesian models with Nash-style "game theory" (where the outcome for each participant depends on the actions of all, and where no participant can benefit by changing his strategy unless others cooperate). And, most recently, many have drawn on "new open economy models" that relegate older Keynesian models to the economic equivalent of Jurassic Park. From almost every angle, with a few exceptions, they have found the benefits of coordinated interest rate setting to be quite small.

How can purveyors of popular grand cooperation schemes be so far off base? Let's first correct a few popular fallacies.

• Fallacy number one is that G-3 exchange rate volatility is a disaster. Yes, it is a problem for some, but empirically, there is little evidence that exchange rate volatility has a significant impact on trade, and interest rates appear to matter a lot more than exchange rates for G-3 economic activity. Of course, this is partly because the euro area, Japan, and the United States are relatively closed economies. But it is also because the "pass-through" from exchange rates to domestic prices is stunningly small, so that consumers are effectively insulated in the short run. (Sustained misalignments of currencies are another matter, but these are generally due to real economy factors rather than to monetary policy.)
  
  
• Fallacy number two is that international monetary policy coordination always leads to less exchange rate volatility. Not necessarily. In fact, theorists have shown that effective monetary policy cooperation can easily imply greater exchange rate volatility. One of the reasons people associate cooperation with exchange rate stabilization is that the Plaza and Louvre accords of the 1980s, announced with great fanfare by the top industrial powers, were aimed at achieving sustainable parities, thereby avoiding disorderly adjustments and excessive volatility. The results were mixed. But even if significant exchange rate stabilization were achievable—and given today's large and moody foreign exchange markets, it may not be—it is far from obvious that such a policy is always desirable.
  
  
• Fallacy number three is that the stronger the international financial links, the stronger the need for institutionalized coordination. In other words, the argument goes, as Europeans boost their holdings of U.S. assets through direct foreign investment and equities and vice versa, Europe and the United States will have a greater direct interest in each other's growth. With greater potential international spillover effects, there needs to be greater coordination of monetary policy. But this isn't necessarily the case. International investment and equity linkages already enhance authorities' incentives to look at the global effects of their national monetary policies, with or without centralized cooperation.

The heart of the matter, though, is that monetary policy is most effective when it is clearly targeted, and it can't be used to fix everything. Despite all the glamour and mystery surrounding central bank decisions on interest rates, the chief role of monetary policy should be to minimize the costs to the economy—in terms of lost efficiency—of the less-than-perfect flexibility of nominal wages and prices. Indeed, if all prices and wages were perfectly flexible, they would move automatically to offset demand shocks, and monetary stabilization policy would be superfluous. Admittedly, real world economies exhibit many imperfections besides price rigidities, not least monopoly power and imperfect information. However, as modern macroeconomic theory decisively demonstrates, monetary policy is generally poor at dealing with such distortions. Thus, to the extent that uncoordinated national monetary policies already do a pretty good job of reducing the costs of price and wage rigidities globally, there isn't much need for grandiose international coordination. At least, that is what most quantitative models seem to show.

Of course, one could argue that coordination may still be worth doing, even if its benefits are slight. This might be true except that, in the real world, an attempt by major countries to institutionalize joint monetary policy decision making could easily take them far afield from idealized notions of cooperation. In the worst case scenario, a push to create a new cooperative international monetary policy institution might provoke a political debate that could cause us to forfeit some of the big gains in monetary policy design we have seen in recent years—for example, by reducing central bank independence. Or, we might end up following an ill-conceived plan. Even if the design of a new international institution were well conceived, there would likely be a transition period as markets learned to understand it. During this period, enhanced uncertainty and weaker credibility could more than offset the benefits of coordination. Certainly, the European Central Bank has experienced growing pains as it has worked to gain credibility and hone its communication strategy. So, if we are going to shift paradigms for G-3 monetary policy, we want to be sure that the long-run benefits are pretty large. Unfortunately, the weight of the recent empirical debate suggests that they won't be.

One caveat is that we don't yet know how the rest of the world (outside the euro area, the United States, and Japan) would benefit from G-3 coordination—definitely an area for further study. But keep in mind that, as we have already noted, central bank cooperation could result in more volatility, not less. Moreover, a recent study (Reinhart and Reinhart, 2002) argues that G-3 interest rate volatility hurts developing countries far more than exchange rate volatility.

For now, though, my message is this: Having three "A" quality G-3 central bank boards (from a technical perspective) that focus mainly on national welfare turns out to accomplish much of what monetary policy can do. Based on current research, it is awfully hard to make the case that moving toward a more cooperative form of G-3 interest rate setting would bring large benefits, even in an ideal world, once domestic monetary policy institutions in all three regions were properly structured. It is on improving domestic monetary institutions and on better implementing domestic monetary policies that we should be concentrating our main efforts.

References:
  Ronald I. McKinnon, 1984, An International Standard for Monetary Stabilization (Washington: Institute for International Economics).
  Maurice Obstfeld and Kenneth Rogoff, 2002, "Global Implications of Self-Oriented National Monetary Rules," Quarterly Journal of Economics, Vol. 117 (May), pp. 503–36.
  Carmen Reinhart and Vincent Reinhart, 2002, "What Hurts the Most? G-3 Exchange Rate or Interest Rate Volatility?" in Preventing Currency Crises in Emerging Markets, edited by Sebastian Edwards and Jacob Frenkel (Chicago: University of Chicago Press and the NBER). See also, "Is a G-3 Target Zone on Target for Emerging Markets?" Finance & Development, March 2002.
  Kenneth S. Rogoff, 1985, "Can International Monetary Cooperation Be Counterproductive?" Journal of International Economics, Vol. 18 (May), pp. 199–217.