ذخایر اضافی و اجرای سیاست های پولی بانک مرکزی اروپا

This paper explains how excess reserves are relevant for today's central banks in the implementation of monetary policy, focusing on the specific case of the policy framework of the European Central Bank (ECB). In particular, this paper studies the impact that changes to the operational framework for monetary policy implementation have on the level and volatility of excess reserves. A transaction cost model that replicates the intra-reserve maintenance period pattern of excess reserves is developed. Simulation results presented show that both the level and volatility of excess reserves may increase considerably under some changes to the operational framework. This is important as errors in forecasting excess reserves weaken the central bank's control of short-term interest rates.

Excess reserves refer to the current account holdings of banks with their central bank beyond required reserves. In the past, excess reserves were regarded as playing the key role in the transmission of monetary policy. Banks were perceived as being more inclined to provide loans to corporates or households when the volumes of excess reserves were high. Open market operations were therefore – at least in theory – conducted with the main objective of steering the level of excess reserves. This view on monetary policy implementation was referred to as ‘Reserve Position Doctrine’ (RPD).1 While it plays less of a role in current academic debate, it still populates monetary policy textbooks in the form of the money multiplier concept, which, although justifiable only in a RPD context, seems still to be considered as a useful pedagogical tool. The role of excess reserves was redefined at the beginning of the 1990s when central banks returned to more explicit interest rate targeting. Although they are no longer seen to play a role in the transmission of monetary policy, excess reserves still represent a significant challenge to the implementation of monetary policy. A central bank's objective of steering interest rates is achieved by managing the conditions that equilibrate supply and demand in the market for bank reserves. So when assessing the liquidity needs of the banking system, it is necessary to take into account the expected value of excess reserves. Even very small errors in the forecast can have an immediate impact on the overnight rate. Systematic errors could have an effect on the rest of the short-term yield curve. It is therefore a precondition of effective monetary policy implementation to understand the logic which drives the level of excess reserves. This paper therefore explains what is behind excess reserves and explains how they are relevant for monetary policy implementation in the Eurosystem. It also studies the impact that changes to the operational framework for monetary policy implementation have on the level and volatility of excess reserves. A simple ‘transaction costs’ model of excess reserves is developed to address these issues. Starting from the observation that in the euro area excess reserves can in principle always be avoided by recourse to the remunerated deposit facility, transaction costs are modelled as the costs to remain in the office until the last payments have been settled. The results show not only that excess reserves may increase considerably under some changes of the framework for monetary policy implementation, but also that such changes could lead to higher volatility and unpredictability. This would cause an increase of the volatility of the overnight rate if not addressed through further policy measures, such as an increase of the frequency of open market operations. The model developed in this paper follows the precautionary demand models of Orr and Mellon (1961) and Poole (1968) that suggested that the demand for excess reserves should be negatively correlated with the level of interest rates and positively correlated with the magnitude of payment shocks. This result was also confirmed for the US by Dow (2001). Section 2 provides a general short review of the role of excess reserves in monetary policy. Section 3 presents the relevant aspects of the liquidity policy of the Eurosystem and highlights the importance of understanding the dynamics of excess reserves in this context. Section 4 develops a simple economic model of the daily pattern of excess reserves within the maintenance period in the euro area. Section 5 reports a number of simulation experiments that illustrate how changes in the operational framework of monetary policy and changes in the level of short-term interest rates could potentially impact on excess reserves and on the conduct of monetary policy. Section 6 concludes.

Excess reserves still play a fundamental role in the implementation of monetary policy, although in a completely different way than that presumed by the exponents of Reserve Position Doctrine. A simple transaction cost model was able to identify the key factors and mechanisms that influence the accumulation of excess reserves in the euro area money market and to replicate the intra-reserve maintenance period pattern of excess reserves. The model was mainly based on the cost–benefit analysis by treasurers of using the deposit facility. This was exemplified by the choice of either bearing a daily cost of staying in the office until money markets close to fine tune the end of day position or leaving somewhat earlier and letting end of day payment shocks impact on reserve holdings. The model thus allowed to demystify further the concept of excess reserves and to provide further evidence that the role of excess reserves presumed in monetary policy between the 1920s and 1990s was based on a misunderstanding of the logic of money markets and the conduct of monetary policy operations. While the academic debate and central banks no longer devote much attention to RPD concepts, ample space is still devoted in textbooks to the money multiplier, a concept inherently linked to RPD. Our model provides a further perspective underlining the inadequacy of this concept. Simulations of the model allowed to investigate the effects of changes of policy parameters on the level and volatility of excess reserves. First, the simulations revealed that, as expected, the level of excess reserves increases when the level of interest rates decline. However, the resulting negative correlation between interest rates and excess reserves should not be misinterpreted as an excess reserves channel of monetary policy transmission because the excess reserves in the model cannot be used by banks to expand loans and hence to create additional money. Excess reserves are not a stable quantity at the level of individual banks, but just a stochastic ex-post residual from payment shocks. Therefore, it does not make any economic sense to expect individual banks to expand loans if this residual increases on average. Excess reserves should not be considered part of the transmission mechanism, but as an issue to be addressed at the very beginning of the transmission mechanism: excess reserves forecasts need to be mapped into allotments in the central bank's open market operations, such that their level and volatility has no effects on short-term interest rates. Apart from making monetary policy transmission more noisy, a higher interest rate volatility can have adverse implications for the efficiency of the money market, for example, by increasing risk premiums, driving participants out of the market, and reducing liquidity. Further policy simulations revealed not only that excess reserves may increase when parameters of monetary policy implementation are changed, but indeed also that excess reserves can become more volatile and unpredictable, with corresponding implications on interest rate volatility (if not addressed properly). By allowing predicting changes in excess reserves patterns resulting from changes in policy parameters, our model thus allows the central bank to prepare in advance adequate responses, be it in the form of a higher or lower liquidity supply through open market operations to address changed levels of excess reserves, or be it in the form of an increased frequency of operations to address their increased volatility. Finally, it should be mentioned that our modelling framework could also be used to understand better the individual recourse to standing facilities of banks, as such recourse is to a significant extent driven by the transaction cost considerations underlying the existence of excess reserves, as captured by the model. It can thereby also help central banks in their decisions on whether to establish a deposit facility and what should be the width of the corridor set by standing facilities.