رژیم نرخ ارز و قیمت ها : موارد ایتالیا، اسپانیا و انگلستان (1998-1874)

This paper studies the relationship among Italian, Spanish and United Kingdom prices over the period 1874–1998, for most of which the currencies of these three countries maintained a floating exchange rate regime. By using cointegration techniques with broken linear trends, we find a single vector for the period 1874–1935 and two vectors and, consequently, a single common trend for the period 1940–1998. Therefore, this paper provides new evidence of no long-run monetary independence under floating regimes. Furthermore, the price differential dynamics captured by deterministic trends in the period 1940–1998, as well as agreeing with the evidence of long-run transmission of interest rates in the floating post-Bretton Woods era, fit in perfectly with the new de facto taxonomies on exchange rates

For the last decade, successive advances in the development of the European Monetary Union (EMU) have reopened the controversy about the advantages and disadvantages of floating versus fixed exchange rate regimes. Part of this controversy has been focused on the possible influence of the type of regime on inflation. Some works have empirically analyzed the link between the exchange rate regime and the persistence of inflation (Alogoskoufis and Smith, 1991, Alogoskoufis, 1992, Obstfeld, 1995, Bleany, 1999, Burdekin and Siklos, 1999 and Gadea et al., 2004). These papers have tested whether floating regimes, since they allow the accommodation of inflation differentials without losses in external competitiveness, may have influenced the wage and price setters’ expectations, encouraging wage-price spirals and leading to a higher degree of inflation persistence than under fixed regimes. A second group of works have studied the relationship between the exchange rate regime and convergence in prices or inflation rates (McDonald and Taylor, 1991, Caporale and Pittis, 1993, Crowder, 1996, Siklos and Wohar, 1997 and Jeong and Lee, 2001). In theory, fixed exchange rate regimes prevent prices across countries from evolving independently, provided trade or financial barriers do not impede goods and/or interest rate arbitrage. If, under a fixed system, a country decided to issue fiat money and provoked inflation, the rise of the domestic price level would lead to a trade deficit. By financing such a deficit, the stock of foreign assets would be reduced, offsetting the initial monetary disequilibrium and driving back the price level. This rebalancing role of the current account could be supported by the capital account (Darby et al., 1983 and Bordo and Schwartz, 1989). According to the interest rate parity condition, in open markets, when the investment risk level across countries coincides, the interest rate differentials tend to equal the expected changes in exchange rates. So, if there was a monetary disturbance such as that described above, domestic interest rates would descend and capital would flow abroad until the interest differential disappeared. Under a fixed regime, there are no expected variations in exchange rates, and, therefore, there is no room to sustain autonomous monetary policy. This is what is known as macroeconomic Trilemma (Obstfeld and Taylor, 1998), the Impossible Trinity (Frankel et al., 2001) or the Trilogy (Reinhart and Rogoff, 2004), terms all used to refer to the impossibility of simultaneously combining a fixed exchange rate, capital mobility and an activist monetary policy. Conversely, under floating regimes, prices and monetary policies across countries can evolve independently since trade and capital imbalances can be corrected by changes in nominal exchange rates. On the goods side, the depreciated exchange rate, by making imports more expensive, can rebalance the current account without the necessity of goods arbitrage leading to an internal disinflation. On the financial side, the expectations of depreciation induced by domestic inflationist policy can stop the outflow of capital before the interest rates have become equal. Thus, by allowing the implementation of autonomous monetary policies, floating rates open the possibility for domestic prices to evolve independently. However, though monetary independence is at the heart of the discussion on exchange rate regimes, studies on the issue are still scarce (Frankel et al., 2004). The possibility of implementing different monetary policies does not necessarily have to lead to long-run price isolation across countries. For this to occur, policymakers must be willing to pursue sustained autonomous policies. The present paper is concerned precisely with this question – whether, in practice, floating exchange rates have historically translated into monetary isolation. It is an issue that we study by analyzing the long-run relationship among the prices of Italy, Spain and the United Kingdom over the period 1874–19981. Our long time-span of data enlarges the usual coverage of price convergence studies, which often start with Bretton Woods, and includes the interesting experience of the gold standard (classical gold standard until the outbreak of the First World War; gold exchange standard in the post-War period). The gold convertibility of currencies that joined this standard defined, in practice, a fixed exchange rate system for the currencies involved. However, the Italian lira only belonged to it intermittently and the Spanish peseta never formally joined the system. Thus, the enlargement of the temporal coverage offers the attraction of studying a period (1874–1935) in which the Italian and Spanish authorities did not operate under an external monetary constraint and, consequently, prices could have evolved independently from British prices. Later, Spain did not enter the international financial order created at Bretton Woods until 1961, by which time Italy and the United Kingdom had already entered. Moreover, Spain did not form part of the European Monetary System (EMS) until June 1989 and the United Kingdom did not join until October 1990 and, even then, the margins for fluctuation within the bands were wide (±6% around a central rate) for both currencies. Soon afterwards, in September 1992, the pound abandoned the EMS, the same year that Italy suspended the participation of the lira in the system. Italy would not return to the EMS until 1996 and only from then on was there again an exchange commitment between the lira and the peseta. Thus, the three currencies considered evolved for long periods during the last century without any exchange rate constraint (or only a very mild constraint). As stated above, our goal is to find out, by using the available econometric tools, whether the Italian, Spanish and UK prices followed a common path between 1874 and 1998 or if, on the contrary, the long periods in which the lack of a fixed exchange regime allowed the implementation of independent domestic monetary policies was eventually reflected in long-run deviations of prices across the three countries. In Section 2 we present the data. The econometric techniques we have applied are considered in Section 3, where we refer to the multivariate cointegration approach of Johansen (1988) and the method proposed by Gonzalo and Granger (1995) to decompose each price series into its respective transitory and permanent components. Section 4 presents the empirical results, with the main finding being that, despite the predominant absence of formal exchange rate arrangements among the Italian, Spanish and UK currencies, prices moved together during the period 1874–1998. Section 5 closes the paper by relating this finding to the literature on the de facto taxonomies of exchange rate regimes.

The aim of this paper has been to analyze the long-run behavior of Italian, Spanish and UK prices between 1874 and 1998 since, for most of this period, their respective currencies have worked under floating exchange rate regimes. For the period 1874–1935, we found a single stable cointegration relationship among the three prices and, therefore, two permanent components which, in turn, differentiate the UK case from the Mediterranean experience. Furthermore, with UK prices representing the orthodoxy of monetary control in central gold standard countries, their leadership over Mediterranean prices illustrates how these peripheral countries, although formally outside the metallic standard (Italy intermittently, Spain permanently), did not remain monetarily isolated. In other words, the finding of cointegration supports the idea of Bordo and Rockoff (1996) about the de facto shadowing of the gold stability rules by the Italian and Spanish authorities. For the period 1940–1998, we found the existence of two cointegrating vectors and, as a consequence, a single common trend that was identified with the UK prices. In this respect, our results support those of others who offer evidence of no monetary isolation for the G7 countries in the second half of the last century, despite the float that began with the breakdown of Bretton Woods. Moreover, the different price dynamic that the trend coefficients estimated in this paper assign to the 1948–1974 and 1975–1998 sub-periods is in accordance with the new taxonomy of exchange rate regimes established in Reinhart and Rogoff (2004). By using the market exchange rates (black market rates included), instead of the official rates, these authors conclude that, over the second half of the last century, the nature of the classification problem has progressively passed from labeling a regime as a peg when it did not behave as such, to labeling a regime as a float when the degree of fluctuation is, in practice, very limited. This fits perfectly with the significant price differential trends that we find for the Bretton Woods era, especially pronounced for the Spanish–UK vector, which are only understandable when related to the generalized existence of black markets in Europe (with a large premium in the case of the peseta) during the 1950s. On the contrary, the irrelevance of trends in the 1975–1998 sub-period points to a stabilization of price differentials, even if the official link among the three currencies considered here was minimal. In any case, beyond differences in the degree of coordination, the equilibrium relationship found in this paper among the Italian, Spanish and UK prices during the period 1874–1898 contributes to the empirical literature on exchange rates by increasing the evidence of the non-exercise of long-run monetary independence under floating regimes.