MNB WORKING PAPERS 2005/6. Exchange Rate Smoothing in Hungary PÉTER KARÁDI - PDF

MNB WORKING PAPERS 2005/6 PÉTER KARÁDI Exchange Rate Smoothing in Hungary Exchange Rate Smoothing in Hungary* Péter Karádi November, 2005 * I would like to thank the participants of the conference of

Please download to get full document.

View again

of 33
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.

News & Politics

Publish on:

Views: 13 | Pages: 33

Extension: PDF | Download: 0

MNB WORKING PAPERS 2005/6 PÉTER KARÁDI Exchange Rate Smoothing in Hungary Exchange Rate Smoothing in Hungary* Péter Karádi November, 2005 * I would like to thank the participants of the conference of the Applied Econometrics Association and the seminar at the Central Bank of Hungary, especially Gábor Kõrösi, Zsolt Darvas and Csilla Horváth for their useful comments. All remaining errors are mine. PhD candidate, New York University, 269 Mercer Street 7FL, New York, NY 10003, USA, I prepared the paper as an economist of the Central Bank of Hungary. The views expressed here are mine and do not necessarily reflect the official view of the bank. The MNB Working Paper series includes studies that are aimed to be of interest to the academic community, as well as researchers in central banks and elsewhere. Starting from 9/2005, articles undergo a refereeing process, and their publication is supervised by an editorial board. The purpose of publishing the Working Paper series is to stimulate comments and suggestions to the work prepared within the Magyar Nemzeti Bank. Citations should refer to a Magyar Nemzeti Bank Working Paper. The views expressed are those of the authors and do not necessarily reflect the official view of the Bank. Magyar Nemzeti Bank Szabadság tér 8 9, H 1850 Budapest ISSN (online) Contents 1. Introduction 5 2. Exchange rate smoothing 7 3. Estimates using announced targets Unobserved variable framework Conclusion References Appendix Unit root tests Obtaining the starting values 26 List of figures 1 The exchange rate, the announced target and interest rates 10 2 HUF/EUR exchange rate, the estimated targets and the Reuters expectations 19 3 Forecast errors (surprises) in the observed exchange rate and interest rate developments 20 4 Exchange rate target shock 21 5 An interest rate rule shock 22 6 A premium shock 22 List of tables 1 Estimates of the exchange rate equation using announced targets 12 2 Interest rate differential rules using announced targets 12 3 Unobserved variable estimation 17 4 Unit root tests 26 5 Restricted unobserved variable estimation no Restricted unobserved variable estimation no Abstract The paper proposes a structural empirical model capable of examining exchange rate smoothing in the small, open economy of Hungary. The framework assumes the existence of an unobserved and changing implicit exchange rate target. The central bank is assumed to use interest rate policy to obtain this preferred rate in the medium term, while market participants are assumed to form rational expectations about this target and influence exchange rates accordingly. The paper applies unobserved variable method Kalman filtering to estimate this implicit exchange rate target, and simultaneously estimate an interest rate rule and an exchange rate equation consistent with this target. The results provide evidence for exchange rate smoothing in Hungary by providing an estimated smooth implicit exchange rate target development and by showing significant interest rate response to the deviation of the exchange rate from this target. The method also provides estimates for the ceteris paribus exchange rate effects of expected and unexpected interest rate changes. JEL classification: E52, F31, F41 Keywords: exchange rate smoothing, interest rate rules, Kalman filter Árfolyamsimítás Magyarországon Az empirikus tanulmány strukturális modellkeretben vizsgálja az árfolyamsimítás hatásait Magyarországon az inflációs célkövetés rendszerének 2001-es bevezetése óta. A megközelítés egy idõben változó, és közvetlenül nem megfigyelhetõ árfolyamcélt feltételez. A modell alapján a központi bank kamatpolitikájával az implicit árfolyamcél középtávú fenntartására törekszik, miközben a piaci szereplõk racionális várakozásokat alakítanak ki a célról, és ennek megfelelõen befolyásolják az árfolyam-alakulást. A tanulmány a nem megfigyelhetõ változók egy módszerét az ún. Kalman-szûrõt alkalmazza az implicit árfolyamcél számszerû becslésére, miközben a céllal konzisztens szimultán becslést készít az árfolyamegyenletrõl és a központi bank által követett kamatszabályról. Az árfolyamsimítás magyarországi jelenlétét megerõsíti egyrészt a becsült árfolyamcél stabil alakulása, másrészt, hogy az árfolyam eltérése a céltól szignifikáns kamatreakciót vált ki. A megközelítés számszerû becslést ad a várt és váratlan kamatváltozások árfolyamhatására. 4 1. Introduction This paper is part of the transmission mechanism research of the Central Bank of Hungary, and its main motivation is to explore futher the interrelationship of short term interest rate and exchange rate development in Hungary using high frequency data. It argues that this relationship is strongly influenced by the systematic exchange rate smoothing behavior of the central bank, and applies a methodology which is capable of numerically assess some of its effects. In open economies, like Hungary, exchange rate can play an important role in the transmission mechanism, and its volatility can cause various, potentially substantial welfare distortions. These considerations provide incentives for the central bank even in pure inflation targeting frameworks to pay attention to the exchange rate developments (Svensson, 2000, and Taylor, 2001). The empirical exchange rate smoothing or fear of floating literature presents evidence that many open economy central banks indeed use direct and indirect methods to influence exchange rate developments in practice (Calvo- Reinhart, 2000). One strand of this empirical literature examines the exchange rate smoothing behavior of central banks by estimating open economy interest rate rules with an exchange rate term (Clarida, Galí and Gertler, 1997 and Mohanty and Klau, 2004). This term is usually proxied by the depreciation of the exchange rate theoretically motivated by some vague lean against the wind interest rate policy. This paper suggests a more structural approach which assumes that, on the one hand, the central bank has a preferred exchange rate level and uses its interest rate policy to maintain this target in the medium term, and, on the other, market participants form rational expectations about the monetary policy and influence the current exchange rate accordingly. In most cases, this preferred exchange rate level is not directly observable, because strategic considerations make the central bank silent, or sufficiently vague about it. But given it is persistent past interest rate decisions of the central bank provide informative signals about this unobserved target variable. Furthermore, assuming rational expectations and forward looking price determination in the foreign exchange market (e.g. uncovered interest parity), current exchange rate developments will reflect the market participants' best assessment of this target. This means that the econometrician can also use the behavior of the exchange rate to gain a better estimation of the unobserved target. The paper applies Kalman filtering to gain the best linear estimate of the unobserved exchange rate target variable and simultaneously estimate an interest rate rule and an exchange rate equation consistent with this target. The estimated system is able to provide interest rate and exchange rate impulse responses to known changes in the exchange rate target, and premium shocks. Furthermore, it can also provide estimates for the ceteris paribus exchange rate effects of expected and unexpected interest rate changes, where the latter takes into consideration that an unexpected interest rate move changes the market participants' beliefs about the central bank's exchange rate target. The paper uses weekly data during a fairly homogeneous period from the widening of the exchange rate band in May 2001 to the end of The frequency of the data provides a moderately large sample, but hinders the usage of more fundamental variables like inflation or output in the estimated rules. Not considering these variables in the interest rate rule can be partly justified by the communi- 5 cation of the Central Bank of Hungary (CBH) during the period. The CBH stressed its belief on the small direct influence of the interest rate on the aggregate demand, and argued that the major influence of its interest rate policy is exerted through its effect on the exchange rate. The framework does not estimate the effects of these fundamental variables on the exchange rate target either. The paper is organized as follows. Section 2 presents some theoretical results about the welfare implications of exchange rate smoothing in the recent New Open Economy literature. The section also presents some empirical results emphasizing the widespread use of exchange rate smoothing in small and open economies. Section 3 presents some evidence for exchange rate smoothing in Hungary using the explicit exchange rate target announcements of the Central Bank of Hungary available up to the end of Section 4 presents the unobserved variable framework and the estimation results. Section 5 concludes. 6 2. Exchange rate smoothing In small and open economies, exchange rate development can have a substantial influence on prices and output, which makes it a potentially important indirect channel of the transmission mechanism. It does not necessarily mean, however, that open economy central banks should also smooth the exchange rate, i.e. react directly to exchange rate developments by their interest rate policy. Along these lines have Clarida, Galí and Gertler (2001) and Galí and Monacelli (2002) questioned the seminal arguments of Ball (1997) and Svensson (2000) supporting exchange rate smoothing. Applying microfounded models capable of welfare analysis CGG and GM argued that the existence of the exchange rate channel does not theoretically imply any reason for exchange rate smoothing. In their frameworks, there are only quantitative differences between closed and open economy monetary policy problems: the exchange rate channel makes the aggregate demand more responsive to interest rate policy and it increases the importance of output gap stabilization. But as the exchange rate variability causes no distortions in these models, there are no reasons for any nominal exchange rate smoothing; or for stabilization of wider inflation aggregates than domestic inflation. Justification for exchange rate smoothing, however, reappears in the microfounded models as well, if the various potential welfare distortions caused by exchange rate volatility are also considered. Benigno and Benigno (2001), for example, argued for the positive welfare effects of exchange rate smoothing in a two country open economy model. In their model, deviations from the flexible terms of trade resulting from excess nominal exchange rate fluctuations and sticky prices cause distortions in production among the countries, leading to lower joint welfare. According to their centralized welfare criterion, rules with interest rate smoothing and explicit reactions to exchange rates similarly to the one estimated in this paper can approximate the first best optimal policies. Microbased models considering only the welfare of an individual country have also found justifications for exchange rate smoothing. One line of recent research, for example, justified it by the welfare consequences of imperfect exchange rate pass-through a general finding in empirical works (see e.g. Campa and Goldberg, 2004), which makes the producers' profits dependent on the exchange rate development. Corsetti and Pesenti (2001) found reasons for exchange rate smoothing in their two-country open-economy framework, and similar results were replicated in small open economy models as well. McCallum and Nelson (2001), for example, consider imports as inputs for production, which implies that exchange rate development influence the costs of production. Excess volatility of the exchange rate reduces welfare, because in the sticky price framework firms adjust their prices only gradually. Optimal monetary policy is found to smooth exchange rate also in the small open economy model of Monacelli (2003), which introduces imperfect pass-through to the model of Galí and Monacelli (2002) considering imports as consumption goods. The reason is that deviations from the law of one price due to imperfect pass-through cause welfare costs, which an optimizing monetary policy should counteract. Other justifications also exist in the current literature: in a novel example, Faia and Monacelli (2002) examine the role of external borrowing in a model with imperfect capital markets and find further justifications for exchange rate smoothing. 7 Magyar Nemzeti Bank If one assumes that exchange rate stability is beneficial for welfare, the quantitative importance of a rule-based direct reaction to exchange rate development might increase further, as argued by Monacelli (2003). Rule-based decision-making can have important advantages in dynamic monetary policy settings, as, thereby, the central bank can have essential effects on expectations. Clarida, Galí and Gertler (1999) argued that a rule based policy can improve the short-run inflation-output tradeoff in case of persistent supply shocks. The reason is that the stronger interest rate response to shocks influences the inflation expectations by promising tough responses to further shocks. Woodford (1999), similarly, argues that the observed interest rate smoothing of central banks can be justified by assuming that they would like to meet their inflation and output targets without causing much interest rate volatility, which they can obtain by committing to a rule which helps them influence inflation expectations. Following these lines of reasoning, commitment to exchange rate smoothing might improve the efficiency of the central bank in influencing exchange rates by guiding exchange rate expectations. As an illustration, consider a generalized UIP equation with a persistent premium shock where s t is the nominal exchange rate, i t is the interest rate differential and μ t is the premium shock. Ruling out bubbles and solving the difference equation forward we can get: showing that the expected exchange rate is influenced by an expected long term nominal exchange rate (E t s T ) and the expected interest rate determination and premium shock development up to date T. The UIP equation implies that by following an exchange rate smoothing interest rate rule, the monetary policy can influence the exchange rate (s t ) not only through the current interest rate differential (i t ), but also through the exchange rate expectations [E t (s t+1 )]. If one assumes that besides exchange rate stability, monetary policy also prefers smooth interest rate developments, it can improve the short-run trade-off between these two variables by committing to an interest rate rule which reacts to a premium shock more strongly, than its optimal discretionary counterpart. Turning to the empirical results, in a widely recognized article, Calvo and Reinhart (2000) found overwhelming evidence about exchange rate smoothing or with their words fear of floating in a wide range of open mostly, but not exclusively emerging economies. Examining volatilities and correlations of various variables, countries were shown to use direct foreign exchange market interventions and interest rate policy to avoid large exchange rate fluctuations. They present it as evidence supporting their claim that countries do not do what they actually say: despite the evidence of real life exchange rate smoothing, IMF classification shows a clear tendency towards floating exchange rate arrangements. But this evidence can be approached from a different angle: that countries tend to move towards reduced transparency about their exchange rate policy. A floating exchange rate regime, by itself, is a non-transparent arrangement, with no explicit information about any exchange rate target or accepted exchange rate volatility. This general tendency towards reduced transparency with the notable exception in Europe can be justified by the increasingly liberalized capital markets and the existence of 8 Exchange rate smoothing speculative capital, which seem to have contributed to the excess exchange rate volatility in countries with transparent arrangements (Obstfeld and Rogoff, 1995b). The reduced transparency, thereby, might be seen as a further measure to contribute to the smooth development of the exchange rate. In a seminal paper estimating monetary policy rules in the G3 (US, Japan, Germany) and E3 (UK, Italy, France) countries, Clarida, Galí and Gertler (1997) also find evidence for exchange rate smoothing, especially in the E3 countries. In their estimated interest rate rules with reactions to anticipated inflation, and output gap, they also examined the role of exchange rate or which also implies exchange rate smoothing the interest rate of a partner country. They have found some, but quantitatively not important evidence for exchange rate smoothing among G3 countries, but this effect was much stronger among the E3 countries and Germany, which eventually initiated a hard ERM policy with fixed exchange rates. In a more recent paper, Mohanty and Klau (2004) examined robust interest rate rules in 13 emerging countries countaining also an exchange rate term. The significant estimated exchange rate depreciation terms make the authors conclude that their result support the hypothesis of exchange rate smoothing or fear of floating in these countries. 9 3. Estimates using announced targets The verbal announcements of the central banks about their preferred exchange rate target, if they are specific enough, can provide a straightforward proxy for the unobserved exchange rate target variable. By announcing its exchange rate target, the central bank, on the one hand, might be able to influence expectations and thereby might gain some leverage over the exchange rate developments. On the other hand, however, announcing explicit exchange rate targets can have serious drawbacks, as they can easily be mistaken as nominal anchors potentially in conflict with other anchors (e.g. inflation target), and an explicit target as a fixed exchange rate could provide a focal point for speculation against the exchange rate. But if the central bank chooses to announce its exchange rate target, announcements, in most cases, might be considered truthful, as it can be assumed that reputational considerations prevent central banks from lying to financial markets with fast learning abilities. It might be noted, though, that even if the announcements are considered truthful, their influence on the rates and market expectations might be limited, as the ability and willingness of the central bank to influence exchange rates might be considered questionable by market participants. The Central Bank of Hungary was unusually explicit about its preferred exchange rate target even after abandoning its narrow band crawling peg exchange rate regime in May Functioning in a shadow ERM II ±15% exchange rate band and following inflation targeting, it periodically Figure 1 The exchange rate, the announced target and interest rates (June 2001 December 2003, weekly data) HUF/EUR % June July Aug Sept Oct Nov Dec Febr Mar Apr May June July Aug Oct Nov Dec Jan Febr Mar May June July Aug Sept Oct Nov HUF/EUR exchange rate (left-hand scale) Announced HUF/EUR exchange rate target (left-hand scale) Policy rate differential (right-hand scale) 3 month rate differential (right-hand scale) 10 Estimates using announced targets announced an exchange rate target it considered preferable and in line with its inflation targets. It maintained this policy of exchange rate transparency up until the end of 2003, when it fully abandoned it. Using the Statements by the Monetary Council, a
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks