The paper discusses the issue of estimating short-and long-run exchange rate pass-through to import prices in euro area countries and reviews some problems with the measures recently proposed in the literature. Theoretical considerations suggest a cointegrating relationship (between import unit values, the exchange rate and foreign prices), which is typically ignored in existing empirical studies. We use time series and up-to-date panel data techniques to test for cointegration with the

more »

... ity of structural breaks and show how the long run may be restored in the estimation. The main finding is that allowing for possible breaks around the formation of EMU and the appreciation of the euro starting in 2001 helps restore a long run cointegration relationship, where over the sample period the fixed component of the pass-through decreased while the variable component tended to increase. JEL: F14, F31, F36, F42, C23 A large number of recent papers (see for example Campa and González Mínguez, 2006; Campa, Goldberg and González Mínguez, 2005; Frankel, Parsley and Wei, 2005; Marazzi et al., 2005) have investigated the issue of exchange rate pass-through (ERPT) to domestic prices. Studies of ERPT have been conducted both for the United States and for countries of the euro area, with a particular focus on its evolution over the past two decades, in response to changes in institutional arrangements (such as the inauguration of the euro area) and to monetary and financial shocks. Several economic policy issues hang upon the determination of the rate of pass-through from exchange rates to prices, and its evolution, both in various time horizons as well as in different sectors. These include issues relating to pricing strategies of foreign exporting firms, the persistence of inflation, the accuracy of inflation forecasts, the impact of entering into a monetary union or structural reforms across the European Union. For the countries belonging to the euro area, the issues listed above are particularly relevant. A notable lacuna in the literature, we argue, is a clear disjunction between the wellworked-out theoretical arguments surrounding the key determinants of pass-through, and the techniques used to estimate import or export exchange rate pass-through equations. Thus, while almost all the theories contain a long-run or steady-state relationship in the levels of a measure of import unit values (in domestic currency), the exchange rate (relating the domestic to the numeraire currency) and a measure of foreign prices (unit values in the numeraire currency, typically US dollars), this long run is routinely disregarded in most of the empirical implementations. This may seem surprising for at least two reasons. First, proper determination of the short-run ERPT relies on appropriate assumptions about the long run. Second, as monetary policy tends to be medium-term oriented, policy actions should in principle look beyond short-term inflation developments for a better understanding of the underlying forces. Since it is commonly agreed that the time series considered are integrated, one way of defining the long run is in the sense of Engle and Granger (1987) , henceforth EG, where the long run is given by the so-called cointegrating relationship. The reason for ignoring this long run, and substituting it by an ad hoc measure, is the failure to find evidence in the data for cointegration. The difficulty inherent in such a re-definition of the long run is two-fold, first the contradiction between a theoretical prediction of a steady state that cannot be found in the data, and, second, the ad hoc measure proposed being no more than an extended version of the estimate of the short-run (and, as we shall see below, strongly dominated by the estimated short-run). We therefore look for the long-run relationship using methods which allow for changes in the long run or use more powerful panel data methods. Focusing on a specification of ERPT normally used in the literature, we argue in particular that: (a) the long run, in the sense of Engle and Granger (1987) , is restorable once appropriate testing strategies (including lag length selection) are adopted and proper account is taken of the possibility of breaks in the long-run relationship; (b) the estimate of the 'long run' used in the empirical literature is sensitive to a number of misspecification issues; (c) once the distinction is established between the long run (with a break) in the sense of Engle and Granger (1987) and the definition used in the ERPT literature, it becomes important to investigate the relative magnitudes of these alternative measures and to interpret them; and (d) it is important to allow for breaks in the long-run theoretical relationship to take due account of pass-through rates in response to changes in financial