On the Effectiveness of Discrete Devaluation in Balance of Payments Adjustment
The realignment of exchange rates which have been disaligned for a long time can never be a precision job; it is usually a hit-or-miss business based on impressionistic guesses and hunches. . . . Economists who know that delays in adjustment are wasteful and that large adjustments cause unnecessary disruption have for years recommended that exchange rates be realigned without delay and by smaller changes . . . (F. Machlup 1970:69)
Almost twenty years ago, in his Horrowitz lectures, The Alignment of Foreign Exchange Rates, F. Machlup identified three important aspects of the process of exchange rate realignment that are still pertinent today: (a) that prolonged misalignment of the exchange rate is costly
in real terms, that is, in terms of output and employment in an economy; (b) that realignment calls for an adjustment whose direction is usually known, but whose magnitude is not; and finally (c) that a system of frequent and small adjustments of the exchange rate is preferable to infrequent and large changes that are usually regarded as means of last
resort. The objective of this paper is to address itself to this last point, leaving aside such important methodological issues as the measurement of exchange rate misalignment in relation to alternative definitions of the equilibrium real exchange rate.