Academic journal article Journal of Money, Credit & Banking

A New Analysis of the Determinants of the Real Dollar-Sterling Exchange Rate: 1871-1994

Academic journal article Journal of Money, Credit & Banking

A New Analysis of the Determinants of the Real Dollar-Sterling Exchange Rate: 1871-1994

Article excerpt

Recent empirical work has reported evidence that purchasing power parity (PPP) deviations can be parsimoniously modeled as univariate nonlinear time series processes (see e.g., Obstfeld and Taylor, 1997, Michael, Nobay, and Peel, 1997, Taylor, Peel, and Sarno, 2001, Kilian and Taylor, 2003). (1) The Exponential Smooth Autoregressive (ESTAR) model of Ozaki (1985) captures the adjustment mechanism implied or derived in the theoretical analyses of PPP by a number of authors (see e.g., Dumas, 1992, Sercu, Uppal, and Van Hull, 1995, O'Connell, 1998, Berka, 2002). (2) In these analyses the authors demonstrate how transactions costs, transport costs, or the sunk costs of international arbitrage induce nonlinear adjustment of the real exchange rate. These nonlinear processes are globally mean reverting and also have the property of exhibiting near unit root behavior for small deviations from PPP. The reason for this feature is that small deviations from PPP are left uncorrected if they are not large enough to cover the costs of international arbitrage. Additionally, nonlinear impulse response functions derived from ESTAR models show that while the speed of adjustment for small shocks around equilibrium will be highly persistent, larger shocks mean revert much faster. This property of the ESTAR models provides some solution to the PPP puzzle outlined in Rogoff (1996)--namely, how to reconcile the vast short run volatility of real exchange rates with the glacially slow rate of 3-5 years at which shocks appear to damp out in linear models.

In the literature cited above the equilibrium real exchange rate was modeled as a constant. However, even in relatively short spans of data, real effects on the equilibrium real exchange rate may be important and therefore play a role in explaining the Rogoff puzzle. (3) A variety of theoretical models, such as those of Balassa (1964), Samuelson (1964), Lucas (1982), Stein et al. (1995), and Stein (1999) imply a non-constant equilibrium real exchange rate.

In the model first postulated by Harrod (1933), and further stressed by Balassa (1964) and Samuelson (1964) (HBS hereafter), productivity differentials between two countries can explain deviations from PPP, especially under cpi -based real exchange rates. If productivity is higher in the tradable-good sector than in the nontradable sector, but wages tend to equality within different sectors of the economy, nontradables (mainly services) will be relatively more expensive in countries with higher productivity levels. (4) Under these assumptions (the HBS effect), PPP need not hold in the short run and it could exhibit "trend-like" behavior defined by the productivity differential between the two countries. Relative real income per capita differentials also drive the real exchange rate equilibrium in the two-country intertemporal model of Lucas (1982).

Other theoretical models of the long run determinants of real exchange rates predict that real appreciations should be associated with the accumulation of net foreign assets. This relationship can be derived from a simple Keynesian model where future trade surpluses via real appreciation would be needed to offset the foreign asset accumulation (see Mussa, 1984, and Broner, Loayza, and Lopez, 1997, for reduced-form models).

A transfer from the home to the foreign country will also have an impact in an intertemporal optimizing framework through two different channels. First, if there is a home bias for domestic tradables, the transfer would reduce global demand for home goods, depreciating the home currency (see Buiter 1989). (5) Second, the effect of a wealth change on labor supply. If domestic residents receive a net transfer from abroad, their labor supply will decrease, reducing domestic output of tradables and hence its relative price must rise (see Obstfeld and Rogoff 1995).

Some attempts have been recently made to incorporate the determinants of the equilibrium real exchange rate in linear models of the real exchange rates. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.