This paper has reviewed the elements that are normally taken into account in assessing the sustainability of industrial countries’ exchange rates. It is evident from the discussion of the preceding sections that it is a difficult matter to arrive at judgments about the exchange rate levels that are sustainable from the standpoint of the Fund’s surveillance duties.

The assessment of what constitutes a sustainable exchange rate may take a number of distinct elements into account. In this context, Section II concluded that although the analysis of disturbances in financial markets is helpful in obtaining an impression of the role of overshooting in the movements of market-determined exchange rates, it is of only limited use in indicating the level of the rate that would be consistent with a sustainable pattern for the overall payments balance in the medium term. Section III discussed assessments of the sustainable exchange rate that emphasize considerations of international competitiveness in the goods market, indicating both their usefulness and their limitations. Section IV described how a more comprehensive assessment based on underlying payments balance considerations is arrived at.

Although indicators of competitiveness and of underlying payments positions are complementary elements in the assessment of sustainable exchange rates, the latter indicator, being more comprehensive, has several advantages as a basis for such judgments. In particular, it attempts to take account of developments in international capital markets and of the effects of factors other than relative prices on both the current and capital accounts of the balance of payments. In addition, underlying payments balance considerations also take explicit account of the overall policy stance of national authorities. Thus, they help to give a measure of international consistency to assessments of external payments positions and sustainable exchange rates among member countries. In this way, they highlight the importance of the broad issues of Fund surveillance that were alluded to in the introduction.

It is also clear from the discussion of the preceding sections that there is an inherent margin of uncertainty in judgments of the sustainable exchange rate. At the technical level the main sources of uncertainty, as indicated in Sections III and IV, include (1) the margin of error in the various statistical indices of the real effective exchange rate; (2) the limited confidence that can be attached to estimates of cyclical effects and past price changes in making projections of underlying payments balances; (3) the difficulty of estimating the likely effects of an underlying change in comparative advantage (e.g., the discovery of North Sea oil) on the sustainable level of a country’s real exchange rate; (4) the sensitivity of estimates of sustainable exchange rates to assumptions concerning price elasticities in international trade; and (5) the substantial difficulties inherent in any attempt to judge the normal level of capital flows. In this context, it is again relevant to emphasize that a major uncertainty arises in connection with the very large asymmetry that is now evident in the global data on current account balances for the various groups of countries.

In addition to the uncertainties that exist at the technical level, serious issues also arise in attempting to assess the set of industrial country exchange rates that would be sustainable, in the sense that they would yield an appropriate pattern of payments balances when viewed from an international perspective. As already noted in the preceding section, there is an inherent margin of uncertainty in such assessments, owing to the fact that there is no unique pattern of current account balances that is clearly superior to all others from an international standpoint.

Given these uncertainties, it is clear that the considerations described in this paper can only be expected to yield an approximate range for the sustainable exchange rate of each member country, rather than a precise level. However, the lack of precision in such assessments should not be overemphasized. In the case of countries that peg their currencies, the need is essentially to produce a rough yardstick that will assist the authorities in choosing a reasonable exchange rate and to recognize when it is becoming over- or undervalued by a substantial margin. For countries that leave the determination of their exchange rate to the free play of market forces, there is a need for a yardstick that will signal situations where market forces have pushed the current rate widely out of line with the value that would be sustainable in the medium term.

In performing its surveillance functions, the Fund is essentially concerned with the issue of whether each member country’s exchange rate policies are appropriate from an explicitly multilateral standpoint. Thus, the Fund must reach judgments about the appropriateness of the exchange rates of all of its members in a consistent manner, while watching in particular for possible conflicts among these members. This requirement implies that the Fund must think in terms of an internationally consistent set of exchange rate ranges for member countries. In the staff’s view, the approaches to exchange rate assessment that have been outlined in this paper go a substantial way toward providing a rough yardstick for doing this for industrial countries, whatever their exchange arrangements. This does not mean that the staff’s assessment of the divergence between the actual exchange rate and the sustainable rate will be equally well defined in all surveillance cases. In some instances, the various considerations may all point toward the same conclusion, and the staff (as well as others) will therefore be able to put forward its assessment with some confidence. In other cases, the various elements in the overall assessment may conflict, and the magnitude of technical and policy uncertainties will be so large that the staff will need to be considerably more cautious in drawing a conclusion on the appropriateness of the current exchange rate. Even in the latter case, however, a discussion of why there is no consensus on the appropriateness of the current rate can be extremely useful. In any event, given the importance of the structure of exchange rates for the operation of the world economy, there is no real alternative but to try to improve existing methods so as to narrow over time the scope and size of current impediments to effective surveillance over exchange rates. Thus, the Fund staff will continue to develop and refine the methods described in this paper on the basis of research undertaken both within the Fund and elsewhere.