The Rise and Fall of the Dollar: Some Explanations, Consequences and Lessons

Abstract

The depreciation of the dollar by some 25 % from February I 985 throughJune i986 poses a problem for our understanding of exchange rate behaviour. This is not to say that the decline was unexpected. On the contrary, the problem is that it was so widely expected. The notion that the dollar was overvalued and headed for a correction was widespread for as much as two years prior to the onset of depreciation. A central puzzle is why forward-looking financial markets did not internalise this expectation. Why did financial markets persist in bidding up the dollar in the face of widespread recognition that this course was unsustainable that sooner or later the paper gains from speculating on the dollar's continued rise would be reversed when the increasingly inevitable correction came to pass? The easy answer to this question, for the non-economist and (increasingly) for the economist as well, is that the markets simply got it wrong; and from here it is a short step to the conclusion that exchange rates are too important a matter to be left to the vagaries of error-prone financial markets. But the economist comes to this answer if indeed he comes to it at all only with some regret. The notion that markets do not always work is one that is not easy to swallow. Some would reject it a priori, whatever the incriminating evidence, in favour of the view that the market is right, but economists can be wrong. Even those with weaker priors would want to be able to track the market failure back to its lair, with the idea that it should be possible to make the markets behave properly once this source of market failure is understood and dealt with. Finally, neither economists nor, perhaps more importantly, policy-makers would want to leap uncritically from the statement that market failure can occur to the conclusion that major policy action is needed. What are the costs of market failure? What are the alternatives to unfettered market determination of the exchange rate? Would these alternatives be better in practice than pure floating? This paper explores some of these issues from the perspective of the i980-6 experience with the dollar. Part I considers the actual experience in relation to both conventional and evolving explanations of exchange-rate determination. Part II assesses this experience in an inevitably speculative fashion in terms of what it has implied, and may imply in the future, for the ability of governments to achieve their domestic economic policy objectives. Part III