우리나라 콜시장에서의 유동성 효과: 공개시장조작 자료를 이용한 분석

There seems to be a consensus among financial market participants in Korea that as the overnight call rate has been almost fixed at the target level, regardless of liquidity supply and demand conditions in the money market, its functions of market information signaling and liquidity allocation have greatly weakened. Also, restricted volatility of the call rate has led to an over-concentration of short-term funds in the overnight call market and the drying-up of term transactions in the money market. Accordingly, the transmission channel running from the call rate to short-term interest rates and long-term interest rates has not been able to function properly. Considering the drawbacks in the call rate targeting framework, the Bank of Korea (BOK) has recently reformed its monetary policy operational framework. This reform includes the change of the policy rate from the call rate to the BOK’s Base Rate applied to transactions between it and financial institutions, the regularization of open market operations and the introduction of standing facilities. In order to empirically verify this consensus by estimating the liquidity effect in the Korean call market, we use the daily aggregate level of reserves calculated by the Bank of Korea while preparing open market operations during January 2006～October 2007. We apply the EGARCH model to estimate the liquidity effect. Our key findings are as follows. First, unlike in other overnight inter-bank markets in foreign countries, we cannot find a liquidity effect on most days of the reserve maintenance period except during the one-to four-day period before settlement day (the end day of the maintenance period). Particularly, there is only a very weak liquidity effect even on settlement day - unlike the case in inter-bank markets in most foreign countries, where it is strong. This may be attributed to the BOK’s liquidity management practice of absorbing excess liquidity or supplementing for liquidity shortages at the market rate on settlement day. In addition, the higher aggregate level of reserve balances in comparison with settlement demand in the banking system is associated with the smaller liquidity effect in the Korean call market. These factors have contributed to the fixedness of the call rate. Second, we find that since the announcement of the increase in reserve requirements in November 2006, the liquidity effect has become stronger than before. This is because the BOK has changed its method of liquidity management to encourage banks to adjust their reserves in more moderate ways. The BOK has minimized liquidity adjustments on settlement day and reduced the number of market interventions during the maintenance period. This has led banks to pay more attention to changes in the aggregate level of reserves than before. In this sense, the BOK’s new framework, which has less frequent but more regular market intervention, is expected to alleviate the fixedness of the call rate. Third, what we refer to as “the banks’ call rate,” calculated based on the rates quoted by commercial banks only, responds more sensitively to a change in reserves than the overall call rate (the BOK’s operating target and policy rate), which weighs the rates quoted by all participants in the call market. This may be attributed to the fact that call loan volume and interest rate have been dominated by asset management companies, which are under no reserve requirement obligation. This means that the BOK’s operating target might not significantly represent changes in demand and supply in the reserve market. Finally, the banks’ call rate responds more sensitively to the impact of a reserve decrease than to that of a reserve increase. This asymmetric response of the banks’ call rate to a reserve change reflects the fact that most banks try to avoid the stigma attached to a bank that is short of reserves. This finding indicates that the banks may be reluctant to use the BOK’s newly introduced lending facility.