آیا مقامات بانک مرکزی باید به صحبتهای بازار ارز خارجی گوش دهند؟

In this paper we study the role of official statements and speeches given by central bank authorities in charge of foreign exchange policy. We investigate the impact of statements that comment on and confirm such interventions on the day of the intervention on the two major foreign exchange markets over an extended period (1989–2003). We also study the impact in terms of policy effectiveness of central bank communication before the actual operations. Our results suggest that appropriate speeches clarifying the current intervention policy can have marginally virtuous effects, both in terms of exchange rate level and exchange rate volatility. This leads us to conclude that, in general, actual interventions on the market should still be employed, provided that official statements are used to clarify the nature and purpose of these interventions.

The exchange rate policies conducted by major countries have changed radically since the mid-1990s. Most central banks, such as the Fed and the ECB, have become increasingly reluctant to rely on actual interventions and have shifted towards the use of communication policy to influence exchange rates. In recent years, only Japan has continued to intervene actively and unilaterally, both by buying currencies and by using words. As a matter of fact, the way central banks communicate with the market has also changed radically. Before, central banks had a tendency to favour secrecy by not clarifying their objectives. Along with a trend towards greater independence, and under pressure from governments, there has been a significant move towards more communication and transparency in both monetary and exchange rate policy. One underlying factor behind this trend is undoubtedly the link between communication and policy effectiveness. But through what channels does communication influence exchange rates? The signalling channel remains one of the most important theories of how the market is influenced: actual interventions, in the form of purchases and sales of currencies in FX markets, may have an impact on exchange rates by conveying a signal about the fundamentals or future monetary and exchange rate policies (Mussa, 1981, Kaminsky and Lewis, 1996 and Vitale, 2003). In the same way, communication may influence exchange rates by signalling future monetary policies or interventions. A recent alternative suggestion, related to the emerging literature on microstructure exchange rate models, is the co-ordination channel (Sarno and Taylor, 2001 and Taylor, 2004). According to this new theory, official statements may be considered as a co-ordination tool allowing market participants to align their views on a new equilibrium. While a growing literature has been devoted to the issue of the effectiveness of communication and transparency in monetary policy (see, inter alia, Geraats, 2002, Ehrmann and Fratzscher, 2007 and Sager and Taylor, 2004), a few papers have also investigated the desirability of transparency for exchange rates (e.g. Bhattacharya and Weller, 1997, Enoch, 1997, Vitale, 1999 and Popper and Montgomery, 2001). Some recent papers ( Fratzscher, 2004, Fratzscher, 2006, Fratzscher, 2008, Jansen and de Haan, 2005 and Jansen and de Haan, 2007) have dealt with this issue from an empirical perspective. In this paper, we present an empirical assessment of the economic desirability of communication in foreign exchange central bank interventions (CBIs). Specifically, we investigate the role of central bank communication as a complementary instrument to actual interventions. To this end, we focus on a key element of communication: statements by monetary authorities made after (but on the same day as) central bank interventions, which are aimed at either confirming the intervention, providing additional comments, or just declining to comment on the operation. Our work is related to other empirical studies (Fatum and Hutchison, 2002, Fratzscher, 2004, Fratzscher, 2006, Fratzscher, 2008, Jansen and de Haan, 2005 and Jansen and de Haan, 2007) that focus on the effect of official statements on exchange rates.1 However, it differs in several respects from this emerging literature. The most significant original contribution of this paper is its investigation of central bank statements as tools that complement, rather than replace, actual interventions on the foreign exchange market. While Fratzscher (2004), for instance, concludes in favour of a good communication policy as a substitute for actual interventions, we find that communication can also be beneficial once the intervention has been carried out. Our approach implies that we consider one particular type of communication, namely ex post communication, as representative of all the official statements perceived by market participants that are issued after direct interventions. We also investigate the impact of ex ante communication, i.e. statements made in the context of G7 meetings or threats of future intervention issued directly by the central banks. The nature and type of official statements studied in this paper are thus different from those analysed elsewhere. We ask the following question: to what extent do official statements make the impact of direct central bank interventions (CBIs) more effective? In other words, we examine whether an official statement made after a FX intervention is an effective policy tool, one that has the intended impact on the exchange rate. To this end, we used a new data set, based on newswire service releases, to collect and classify the daily statements made by officials of the Bundesbank (after 1999, the European Central Bank (ECB)), the Federal Reserve (Fed), and the Bank of Japan (BoJ) over the period 1989–2003 for the EUR/USD and 1991–2003 for the YEN/USD exchange rates.2 Using a standard event-studies regression approach (Dominguez, 1998), we test whether the impact of reported central bank interventions that are followed by a statement differs from the impact of those with no subsequent statement. We test the impact on both exchange rate levels and exchange rate volatility. From a policy perspective, our results can shed some light on whether and how the monetary authorities should talk to the market when they consider an FX intervention. As a robustness analysis, the econometric model has been extended in order to account for different factors likely to change our results such as the intervention size, the distinction between announced and unannounced interventions, the change in intervention regime, the size of the exchange rate misalignment, and at last the ex ante communication captured by G7 meetings and official statements indicating clearly the possibility of a future intervention. We find that, on the whole, interventions accompanied by statements had more impact on the exchange rate level. Issuing commentary statements tends to reduce the traditional increase in exchange rate volatility, which accompanies central bank interventions. Interestingly, our results also hold when we consider the intervention size of the intervention in interaction with the official statements. The analysis also suggests that ex ante statements can have virtuous effects, for instance in terms of FX volatility. This leads us to conclude that, in general, actual interventions on the market should still be considered, provided that official statements are used to clarify the nature and purpose of these interventions. The paper is organised as follows. Section 2 reviews existing theories on the impact of interventions. It also briefly describes the existing literature on the effect of communication on exchange rates. Section 3 provides information on the data for both central bank interventions and speeches. Section 4 details our econometric approach, and presents and comments on the empirical results. It also includes some robustness analysis and discusses further issues. In Section 5 we present our conclusions.

In this paper we assessed empirically how communication by the three major central banks about their exchange rate policies has affected exchange rate levels and exchange rate volatility. Communication is captured here by official statements by central bank authorities that aim at either confirming or commenting on FX operations. We find that, consistent with the signalling theory of central bank interventions in the FX markets, statements allow central bankers to improve the information conveyed by direct sales or purchases of foreign currency. In particular, we find that commenting or confirming speeches result in central bank interventions having a more virtuous impact on exchange rate levels. This result is important, as targeting exchange rate levels and reversing undesirable trends has obviously been a major aim of such operations.41 Interestingly, this result holds for the two major exchange rate markets, the EUR and the YEN against the USD. On the whole, our results indicate that communication can enhance the effects of interventions on the exchange rate dynamics. Another important result concerns the smoothing impact that commenting speeches have on exchange rate volatility. Consistent with the signalling theory of CBIs, this result suggests that talking to the markets in an appropriate way can remove at least some of the ambiguity that is associated with FX operations. The same is true when communication is provided before the operation, for instance during G7 or G8 meetings or directly by the monetary authorities themselves. To the extent that removing excess exchange rate volatility is another important goal for central bankers, the result also suggests that central bank statements can be productive. Our results turn out to be robust to a set of factors that have received recent attention in the literature. These are the variation over time of the Japanese exchange rate policy (Ito, 2003), the case for a co-ordination channel regarding the role of statements (Sarno and Taylor, 2001), the inclusion of statements as separate policy instruments (Fratzscher, 2006) and the fact that intervention size varies significantly over time. In general, our analysis – consistent with that emerging from other authors, such as Fratzscher (2008) – suggests that appropriate statements by monetary authorities on their exchange rate policy can be a valuable complementary tool to actual exchange rate operations.