هماهنگی سیاست و ریسک در بازار ارز خارجی برای ارزهای اصلی اتحادیه اروپا

This study investigates cointegration, policy coordination and the risk premium in foreign exchange markets for major EU currencies since the inception of the EMU in January 1999. The results show that only the krone and the pound are cointegrated with the euro. Tests of inflation convergence and analyses of reduced-form and structural VARs indicate that the cointegration evidence reflects the relatively stronger degree of monetary policy coordination and at least the de facto fixed exchange rate regime of Denmark and the U.K. with the EMU. Additionally, cointegration of spot exchange rates can be considered one of the factors that represent the time-varying risk premium due to its explanatory power for the return to forward speculation.

The relationships among spot exchange rates have been of considerable interest to researchers, policymakers and foreign exchange market participants. The majority of prior studies (e.g., MacDonald and Taylor, 1989 and Rapp and Sharma, 1999) find no evidence of long-run equilibrium relationships, as measured by cointegrating relationships, among spot exchange rates of various currencies over the modern float. However, as policymakers in different countries closely coordinate their monetary policies to considerably limit fluctuation of the exchange rate of one currency for another, a de facto fixed regime results.1 Their spot exchange rates (expressed in terms of a numeraire currency such as the U.S. dollar) may be cointegrated because the currency prices cannot permanently diverge from each other. An example of the de facto fixed regime is a group of European Union (EU) currencies prior to the inception of the European Economic and Monetary Union (EMU) in January 1999. Haug et al. (2000) and Rangvid and Sorensen (2002) detect cointegrating relationships among spot exchange rates of several EU currencies during the pre-EMU period. This finding is consistent with the fact that the Maastricht Treaty and the Exchange Rate Mechanism (ERM), among other measures, require considerable alignment and convergence of key economic variables of EU countries including inflation rates and exchange rates before they can become EMU members and adopt the euro as their currency. To date, Denmark, the U.K. and Sweden are major EU countries which are not EMU members and hence still retain their national currencies. Denmark has been part of the ERM II since January 1999. The Danish krone is pegged to the euro at a central rate with a ±2.25% fluctuation band.2 This very narrow band is indicative of a truly fixed exchange rate regime and implies strong monetary interdependence between Denmark and the EMU. The fixed regime also suggests that the cointegrating relationship between the krone and the euro is likely present. Further, the U.K. conducted the “Five Economic Tests” to evaluate its readiness for EMU membership. The test results published in 2003 indicate that significant progress had been made towards convergence in inflation rates, long-term interest rates and government deficits and debt. However, the U.K. decided that it was not ready to join the EMU because, among other reasons, the convergence might not be sustainable. Thus, the U.K. has adopted various policies and reforms to further improve the convergence and make its economic structures increasingly compatible with the EMU.3 Because of these efforts, the de facto fixed regime between the British pound and the euro may exist and the two currencies may be cointegrated even though the pound is officially under the floating regime. In contrast, the monetary policy of Sweden is based on its own inflation target under the floating regime and there is no specific mandate that prevents the divergence of the Swedish krona and euro prices.4 Thus, it is likely that a cointegrating relationship between the two currencies is not present. Whether or not the three EU currencies are cointegrated with the euro, and more importantly, whether or not the presence/absence of long-run equilibrium relationships implied by cointegration actually reflects differences in the degree of monetary interdependence and the exchange rate regime with the EMU have not been jointly investigated by prior studies with post-EMU data. This investigation is of importance to EMU policymakers. According to the Maastricht Treaty, the monetary alignment or convergence helps ensure that price stability within the EMU area is maintained even with the inclusion of new member states. The implications of the presence/absence of the cointegrating relationship are also of importance to market participants. The absence of cointegration and its error correction representation (Engle and Granger, 1987) implies that the price of one currency is not predictable based on common long-term dynamics. This unpredictability may be consistent with weak-form market efficiency.5 Conversely, however, the presence of cointegration and the predictability as a result of an error correction representation (which requires subsequent adjustments of cointegrated currency prices to retain a cointegrating relationship once deviation from the relationship occurs) may not necessarily imply market inefficiency. Baillie and Bollerslev (1989) interpret this predictability as either a violation of weak-form efficiency or evidence of a time-varying risk premium. Because foreign exchange markets are global in scope and extremely large in scale, it is unlikely that the inefficiency exists simply as a result of policy coordination. The relevant issue is whether or not the predictability as a result of dynamic adjustments and comovements of cointegrated currency prices represents the risk premium which makes the forward exchange rate differ from the expected future spot exchange rate. Prior studies using pre-EMU data have not shown that this is the case and to our knowledge no study has examined this issue using post-EMU data. Given the niches in the literature, this study sets forth three main objectives. First, it examines whether or not the krone, pound and krona have long-run equilibrium or cointegrating relationships with the euro over the post-EMU period from January 1999 through June 2006. Second, it investigates whether or not the presence/absence of the relationships reflects the differing degree of monetary policy coordination and exchange rate regime with the EMU. The Swiss franc and Japanese yen are also included in the analysis as control variables.6 Further, convergence of inflation rates (which are directly related to or influenced by monetary policies) should also be observed if the spot exchange rate cointegration truly reflects monetary interdependence. Thus, cointegration tests of the U.K., Danish and Swedish inflation rates and the EMU inflation rate are performed and inferences about long-run inflation convergence are obtained.7 Conversely, deviation from a cointegrating relationship (i.e., cointegrating vector or CIV) of spot exchange rates can represent transitory monetary independence if it is also directly related to deviation from long-run convergence of inflation rates. Whether or not this relationship exists is examined through reduced-form and structural VARs. Third, this study examines whether or not cointegration of spot exchange rates, if transpiring, can represent a risk premium to market participants. This is done by examining the explanatory power of the CIV of spot exchange rates for the return to forward speculation which is defined in Taylor (1995) as the difference between the future spot rate and the forward rate. The remainder of this study is organized as follows. Section 2 describes the data and explains the econometric methodology used, with estimation results and related findings set forth in Section 3. Section 4 provides conclusions.

This study updates and extends prior studies related to cointegration, policy coordination and the risk premium in foreign exchange markets for the Danish krone, British pound, Swedish krona and EMU euro since the inception of the EMU in January 1999. The results suggest that only the krone and the pound exhibit cointegrating relationships with the euro. These long-run equilibrium relationships reflect the relatively stronger degree of monetary interdependence and at least the de facto exchange rate regime of Denmark and the U.K. with the EMU. The monetary interdependence implies that both Denmark and the U.K. are already informal EMU members and they possibly realize greater benefits from not seeking formal EMU membership. They can have some monetary independence at the present time (albeit limited and transitory) and can pursue totally independent monetary policies in the future (if they elect to have a floating regime to achieve domestic objectives). They can also exert some influences on EMU monetary policies. This interpretation is supported by the weak exogeneity test results that the EMU does not have the leadership role. Instead, the policy is likely influenced by interactions among monetary authorities in Denmark, the U.K. and the EMU. Additionally, the results indicate that deviation from cointegrating relationships of spot exchange rates can explain a statistically significant portion of the return to forward speculation. Because the deviation requires dynamic adjustments of relevant currencies to maintain the relationships over time, the comovements emerge and foreign exchange market participants obtain smaller benefits and thus incur greater risk from having exposure in the cointegrated currencies. Therefore, cointegration of spot exchange rates can be considered one of the factors that represent the time-varying risk premium in foreign exchange markets, particularly for the major EU currencies under investigation.