انتظارات غیر منتظره در مقابل انگیزه های سیاست های پولی و نرخ بهره عبور از بازار بانکداری منطقه یورو خرده فروشی

This paper investigates the interest rate pass-through in the euro-zone’s retail banking markets by differentiating between expected and unexpected monetary policy impulses. The paper introduces interest futures as measures of expected interest rates into pass-through studies. By allowing various specifications of the pass-through process, including asymmetric adjustment, we find a faster pass-through in loan markets when interest rate changes are correctly anticipated. In contrast, deposit markets are found to be more rigid. Overall, our results suggest that a well-communicated monetary policy is important for a speedier and a more homogenous pass-through but may also be complemented by competition policies.

Investigating the pass-through of monetary policy impulses onto retail banking interest rates has become an important part of the research on the financial part of the monetary transmission process in the euro zone. The results of such analyses have direct implications for assessing the efficiency of the monetary policy, the heterogeneity (or a possible trend towards homogeneity) of the euro-zone financial system, and the competitive situation in different segments of the banking markets. While this research area is fairly well developed, the complete absence of interest rate expectations and monetary policy anticipation from the literature is striking. In the presence of forward looking financial markets, and in particular with the existence of interest rate futures, which may reflect expected future interest rates, pricing behaviour in retail banking is likely to be forward looking too. This study explores this issue by disentangling the impact of expected and unexpected monetary policy impulses. Following the pioneering pass-through study by Cottarelli and Kourelis (1994) who apply a dynamic lending-rate model in an international context, this approach had soon been adopted within the European context (Cottarelli et al., 1995, BIS, 1994 and Borio and Fritz, 1995). Additionally, since Sander and Kleimeier (2000) pass-through studies are now regularly based on an error-correction specification (e.g. Mojon, 2000, Heinemann and Schüler, 2003 and Toolsema et al., 2001). Most recently, asymmetric adjustment of retail bank interest rates to monetary impulses has also been considered. This still relatively small literature (see e.g. Sander and Kleimeier, 2000, Sander and Kleimeier, 2002, de Bondt, 2002 and de Bondt et al., 2002) builds on Tong, 1983, Scholnick, 1996, Scholnick, 1999, Balke and Fomby, 1997, Enders and Granger, 1998 and Baum and Karasulu, 1998 and Enders and Siklos (2001). Euro-zone pass-through studies commonly find the following pattern: First, bank interest rates are sticky. Thus, monetary policy rate changes typically lead to a less than one-to-one change of retail rates, i.e. the short- and medium-run multipliers are taking values often far below unity. Second, there are considerable differences in the pass-through across different bank lending and deposit rates. Third, studies that allow for asymmetric adjustment typically find ample evidence for asymmetries but without any consistent pattern. Fourth, while there is no consensus yet regarding a possible full pass-through in the long run, most authors agree that the pass-through is most complete for short-term lending to enterprises. Fifth, most studies find significant differences in the pass-through mechanism across the euro-zone countries. Finally, there is increasing though still weak evidence that the pass-through mechanism has become faster and more homogeneous in the recent years, eventually pointing to a homogenising role of the Euro and the single euro-zone monetary policy. Nevertheless, the remaining differences in existing literature are still substantial. Sander and Kleimeier (2004) have tried to unify this research by arguing as follows: First, these differences are caused predominantly by differences in (1) the choice of exogenous market interest rate, (2) the length and timing of the investigated periods, (3) the treatment of possible structural breaks, and (4) the chosen methodology for the pass-through study. Second, these differences could be reconciled by focussing on a unifying approach that uses an appropriate measure of the monetary policy stance such as short-run money market rates, incorporates endogenous determination of structural breaks, and introduces an automatic model selection out of a vast variety of possible pass-through models allowing for asymmetric and threshold adjustment. Doing so, it is shown that breakpoints are typically occurring some years before monetary union has commenced, thus pointing to the important role of forward-looking behaviour in financial markets. Comparing pre- and post-break pass-through behaviour, the results of previous studies regarding short-run stickiness of retail interest rates are confirmed, but there is evidence for an increased speed of the pass-through in the post-break period. Moreover, a large number of banking markets has a less than perfect long-run pass-through. This suggests that market imperfections such as credit rationing may have remained an important feature in European retail banking. The view that euro-zone banking markets are still fragmented is supported, with the eventual exemption of short-term lending to enterprises. Finally, analysing the determinants of the pass-through shows that competition, banking market integration, a stable monetary policy regime, or a more homogeneous growth performance are all important variables for homogenising the pass-through and thus monetary transmission in the euro zone while legal and cultural differences may, however, preclude full convergence. In this paper we extend this unifying approach to the pass-through of monetary policy impulses onto retail banking by introducing an explicit distinction between expected and unexpected monetary policy changes. In doing so, we follow Kuttner (2001) who uses federal funds future markets to obtain information on expected and unexpected monetary policy changes. We adapt this approach to the euro zone by making use of EURIBOR futures. This methodology provides an alternative to the VAR literature, which retrieves monetary policy shocks as orthogonalised innovations and can be subject to an endogeneity problem.1 The rationale for doing so is that in the presence of forward-looking financial markets, change in the policy rate may already be priced into retail banking interest rates well before the actual changes in the monetary policy rate occur. For example, in the context of the US mortgage market Sellon (2002) argues that in the recent years “both mortgage rate and market interest rate appear to anticipate monetary policy actions”, i.e. they rise somewhat before episodes of monetary tightening and fall well before decreases in federal funds target rates. Consequently, if retail interest rates anticipate changes in monetary policy rates, the measured short- and medium-term multipliers may not convey a correct picture of the pass-through process. A related issue is the impact of correct anticipation of the ECB’s policy actions on the speed of the pass-through process, thus pointing to the role of a frictionless communication between the central bank and the market. Moreover, the irregular pattern of asymmetries across countries and banking products found in previous studies may reflect a differential behaviour with respect to expected and unexpected changes. By differentiating between both we want to shed some more light on the stickiness of bank product prices. Finally, it could be argued that with a single monetary policy the anticipation of monetary policy action would also become more uniform across the euro zone and thus contribute to a more homogenous pass-through. The plan of the paper is as follows. In Section 2 we introduce our measurement of expected and unexpected monetary policy changes by making use of EURIBOR futures to measure expected interest rates which is novel in the pass-through literature. In Section 3 we describe our methodology. Section 4 reports the results of our analysis and Section 5 concludes with a view on improving the pass-through process in the euro zone

Our paper extends the traditional pass-through literature by introducing a distinction between expected and unexpected monetary policy impulses. Among the most important conclusion are: First, we can establish a significant positive impact of interest rate anticipation on the pass-through in loan markets. Thus we conclude that explicitly measuring interest rate expectation is an important addition to the pass-through literature. Second, given the faster response of retail interest rates to anticipated monetary policy impulses, our results point to the important role of a good communication policy by the ECB for monetary policy effectiveness. Third, we find severe asymmetries in retail banking price behaviour, in particular a more sluggish behaviour of deposit rates, thus highlighting the importance of price rigidity in the banking industry. Moreover, our results contest the findings of earlier studies that asymmetries are present, but show irregular pattern. In particular with respect to anticipated interest rate changes we find that the asymmetries show a quite clear pattern which points to role of imperfect competitive markets. Fourth, and consequently, we conclude that a transparent central bank policy is helpful for speeding up the pass-through, but it is no substitute for competitive markets and competition policy. Finally and under similar reservations regarding competitive banking markets, our results point therefore to the potential role of a good central bank communication policy in harmonising the euro-zone pass-through.