We analyze the impact of a requirement similar to the Basel III Liquidity Coverage Ratio on the bank intermediation applying Regression Discontinuity Designs. Using a unique dataset on Dutch banks, we show that a liquidity requirement causes long-term borrowing and lending rates as well as demand for long-term interbank loans to increase. Lower levels of aggregate liquidity increase the estimated effects. Short-term borrowing and lending rates only rise during periods of lower market-wide liquidity. Further, banks do not seem able to pass on the increased funding costs in the interbank market to their private sector clients. Rather, a liquidity requirement seems to decrease banks’ interest margins.

The purpose of this study is to provide empirical evidence on the links between currency and banking crises. Panel data probit and bivariate probit models are estimated to a sample of 21 developed and developing countries having monthly observations between the years 1985 and 2010. The findings indicate that banking crises precede currency crises, and vice versa. Currency crises also indirectly influence future banking crises probability through external shocks, liberalized financial markets, or highly-leveraged banking sectors. The study also finds evidence of contemporaneous correlation between the two crises. The results not only confirm the theoretical links between banking and currency crises, but also underline the importance of higher frequency data in analyzing the relationship between various financial crises.

This paper takes a novel approach to derive a central bank intervention reaction function. A GARCH model for exchange rates is amended to allow interventions to have an effect on both the mean and the variance of exchange rate returns. An intervention reaction function is obtained by combining the model with a loss function for the central bank. Estimation results for the implied friction model reproduce the familiar ‘leaning against the wind’ policy by the Bundesbank and the Federal Reserve. Furthermore, the central banks appear to have reacted to increases in the conditional variance of daily DM/$-returns.