Real exchange rates have shown very large swings that have persisted for years at a time, only to be subsequently reversed. The real value of the dollar, for example, rose 72 percent against the yen from late 1978 to early 1985 and then declined by 46 percent in a space of just three years, from early 1985 to early 1988. Exchange-rate swings of this magnitude can play havoc with investment plans and production decisions. A plan, for example, to build an automobile plant in the United States might make good business sense at today's exchange rate. But five years from now, when the plant is ready to begin production, a very different constellation of exchange rates might leave this plant without a hope of competing successfully in international markets. Faced with such uncertainty, it would not be surprising if investors were reluctant to commit to large investment projects in industries where a changing exchange rate might expose them to ruinous foreign competition.
Where investiments have been made, large exchange-rate movements can wipe out the viability of entire industries. With one set of exchange rates, production in the United States of autos, steel, or computers might be quite viable. With another constellation of exchange rates, it might be impossible. When exchange rates move, potentially productive plants may be idled. Workers will have to be retrained and will perhaps have to move to new cities in order to find jobs in new industries. Communities favored by a swing in exchange rates will have to build new schools, roads, and sewers to accommodate a growing population, while still-functional schools, roads, and sewers may be abandoned in areas hard hit by exchange-rate changes. lf a movement in exchange rates reflects a permanent change in the productivity of workers or industries in one nation relative to those of some other nation, then such adjustment costs are unavoidable. The best policy in these circumstances is usually just to get on with the necessary adjustments.
This will be painful enough. But what if the exchange-rate change is reversed in a few years? The adjustment costs already borne may be effectively lost, as we rebuild or reopen the factories and towns we just closed. Alternatively, we might simply do without the factories and jobs already lost but which might have been perfectly viable had there been no exchange-rate swing in the first place. A final diabolical element is added by the fact that it can be impossible to know whether an exchange-rate movement is permanent or likely to be reversed in a couple of years. Thus, it is hard to know whether we should undertake painful but necessary adjustments or just hold until exchange rates move back again.