Real Exchange Rates
Parity Check

An important summary measure of the competitiveness of an open economy is the "real exchange rate", which measures the price level of domestic or "non-traded" goods relative to the prices of traded goods. A rise in this index, often called a "real appreciation", may be identified with a loss in international competitiveness, since encourages a substitution of resources away from the traded goods towards the non-traded goods sector, while consumers are encouraged to substitute in the opposite direction. For some value of the real exchange rate these forces would exactly offset each other, so that balance-of-payments equilibrium would obtain. This hypothetical value is often referred to as the "equilibrium real exchange rate".

The importance of these considerations is well known and the determinants of the equilibrium real exchange rate have been examined in a variety of special models. However, no compact derivation of these determinants in a general model which allows for many non-traded goods is currently available. The purpose of this paper is to present such a derivation and to illustrate its relevance to a number of policy issues.

The principal result of the paper is a formula which shows that any disturbance is more likely to lead to a rise in the real exchange rate (a real appreciation) the greater its effect on the demand for and the smaller its effect on the supply of non-traded relative to traded goods. More specifically, the proportional change in the real exchange rate equals a weighted sum of the differences between the marginal propensities to consume and to produce each non-traded good, where the weights reflect the relative importance of each non-traded good in consumption and the difficulty of substituting it for other goods.

This result can be used to analyse the effects of a transfer of income from foreign to home residents. By definition, such a transfer is effected in terms of traded goods, so it has no impact effect on the supply of non-traded goods. However, by raising income, it directly affects the demand for non-traded goods and so tends to induce a real appreciation.

A similar outcome follows from a boom in a traded goods sector which is an "enclave", i.e. a sector which has no production links with the rest of the economy. Natural resource sectors are an obvious example of such enclaves. The theory predicts that booms in such sectors are likely to lead to a real appreciation and a loss of competitiveness by (non-booming) traded goods sectors. Although the boom itself means that real national income must rise, its side-effects may pose problems of adjustment, a phenomenon which has come to be known as the "Dutch Disease". The general formula also shows that the effect of a boom on the real exchange rate is less clear-cut if production in the booming sector is integrated with the rest of the economy, since the boom now has an additional direct effect on the supply of non-traded goods.

The general result can also be applied to international comparisons of "purchasing power parity". This doctrine predicts that price levels should be equal across countries when compared using equilibrium real exchange rates. However, there is ample evidence which casts doubt on this hypothesis, suggesting that non-traded goods are relatively cheaper in low-income countries. The formula presented in this paper suggests why this may occur. The change in the real exchange rate given by the formula can be reinterpreted as the difference between the real exchange rate of a high- and a low-income country. The marginal propensities to consume and produce can also be reinterpreted as the differences between the two countries' average propensities to consume and produce non-traded goods. If cross-country productivity differences are smaller in the production of non-traded goods (such as services) than in the production of traded goods, then the production terms in the formula are likely to be small or even negative and so the higher-income country will have a higher real exchange rate (a higher relative price of services). This effect will be reinforced if non-traded goods are superior in demand so that the consumption terms in the formula are positive and relatively large. This interpretation is consistent with a considerable body of theoretical and empirical work which has attempted to explain the systematic pattern of deviations from purchasing power parity.