Exchange Rate Stabilization
Off the peg

Conventional views of exchange rate management are wrong, Peter Kenen told a recent CEPR lunchtime meeting. Both the experience of the 1980s and theoretical analysis suggested that floating exchange rates do not guarantee superior economic performance, nor do they eliminate economic interdependence. The monetary system of the 1990s should be based on explicit commitments to pegged-but-adjustable exchange rates, with hard margins around the bands within which exchange rates can move and mandatory intervention to defend those margins. But the margins must be wider than those under Bretton Woods or the European Monetary System, in order to accommodate exchange rate realignments.
Peter B Kenen is Walker Professor of Economics and International Finance, and Director of the International Finance Section at Princeton University; he is currently a visiting research fellow at CEPR and at the Royal Institute of International Affairs. He based his talk on the research reported in
Discussion Paper No. 240, `Exchange Rates and Policy Coordination in an Asymmetric Model'. Financial support for the meeting was provided by the German Marshall Fund of the United States.
Kenen stressed the fundamental nature of the recent shift in economic opinion. From 1976, when the Jamaica Agreement ratified floating exchange rates, to the Plaza Agreement in 1985, four propositions commanded widespread support among academics and policy-makers: first, that floating exchange rates are the least costly way to reconcile differences in national policies; second, that governments can only achieve exchange rate stability by pursuing domestic policies, especially monetary policies, aimed at combating inflation and stabilizing expectations; third, that foreign exchange markets are fairly efficient in gathering and processing information, so that freely floating rates will promptly reflect economic funda mentals; and last, that official intervention on foreign exchange markets is thus ineffective and inappropr iate, save on rare occasions when markets become disorderly.
The appreciation of the US dollar in 1983-5 showed that foreign exchange markets are not always efficient and may behave irrationally. But there has been a more important discovery. The costs of floating exchange rates have been very high, and they have not reduced economic interdependence or the need for policy coordination. Floating exchange rates, Kenen said, cannot eliminate interdependence they can only change its form. They may indeed produce a peculiarly painful form of interdependence: because prices and wages are sticky, changes in nominal exchange rates are matched almost fully by changes in real exchange rates. Under these conditions, he said, exchange rate stability becomes an important objective in its own right, not an incidental reward for following sound policies.
Drawing on research reported in his CEPR Discussion Paper, Kenen carried this argument further. Using a two-country model, his paper examined how the choice of a floating versus a pegged exchange rate regime affected each country's ability to stabilize its output and price level without having to co ordinate its monetary policy with that of the other country. A pegged-but-adjustable exchange rate proved to be superior to a floating rate. In the face of a permanent switch in demand between the two countries' bonds, for example, a pegged rate was fully optimal. Each government could stabilize its output and price levels by open-market operations in its own bond market with no need for policy coordination. A floating rate was not fully optimal because output and price changes were inevitable under floating rates, and the governments had to coordinate their policies closely so to minimize the welfare losses stemming from those changes. The same results obtained in other cases, and Kenen found no situation in which a floating rate proved to be better than a pegged rate.

Kenen warned against placing too much emphasis on this result, which depended on the assumptions used to construct the theoretical model. Nevertheless, he said, it challenges conventional views about economic interdependence and policy coordination. As floating rates may require more, not less coordination, and as coordination is difficult in the best of circumstances, economic performance may be less satisfactory under floating than under pegged rates.
Kenen warned, however, that an agreement to stabilize exchange rates must satisfy three conditions:
First, there must be scope for small but frequent realignments, not only to offset differences between inflation rates, but also to alter real exchange rates. Governments must not back themselves into indefensible commitments because they have failed to devise suitable procedures and guidelines for making collective decisions. It would be dangerous, Kenen said, to understate this difficulty. It is in this respect that the experience of the EMS would be hardest to emulate.
Second, there must be closer collective surveillance over national policies, especially fiscal policies. Citing his own theoretical work as well as the events of the 1980s, Kenen argued that large shifts in fiscal policies are the most difficult disturbances to neutralize using monetary policies and exchange rate changes. The monitoring of fiscal policies, he said, should be the main task of multilateral surveillance among the G7 countries, and that process must be forward-looking to have any influence on policy formation.
Third, there must be major changes in the character and use of international reserves, in order to deal with two difficulties that are bound to arise when intervention is used to stabilize exchange rates. Reserve supplies must be very elastic in the short run to combat speculative pressures, but they must be limited in the long run to give each country confidence that the others are constrained. The problem can be tackled in an ad hoc manner, using bilateral swaps to impart elasticity and long-term foreign currency borrowing to pay off these swaps. Over the longer term, however, it should probably be resolved by changing the role of the IMF as a supplier of reserves and of reserve credit.
Further reform of exchange rate arrangements, Kenen concluded, should be high on the agenda of the next US administration, but it must first reduce the budget deficit, which poses a continuing threat to financial stability and short-run exchange rate stability. Without cutting the budget deficit, moreover, the new administration cannot acquire the credibility it needs to exercise leadership in international monetary matters