An empirical study of liquidity risk embedded in banks' asset liability mismatches

Abstract

The correct measure and definition of liquidity in finance literature remains an unresolved empirical issue. The main objective of the present study was to develop, validate and test the liquidity mismatch index (LMI) developed by Brunnermeier, Krishnamurthy and Gorton (2012) empirically. Building on the work of these prior studies, the study undertook to develop a measure of liquidity that integrates both market liquidity and funding liquidity within a context of asset liability management. Liquidity mismatch indices were developed and then tested empirically to validate them by regressing them against the known determinants of liquidity. Furthermore, the study investigated the nexus between liquidity and profitability. The unit of analysis was a panel of 12 South African banks over the period 2005–2015. The study developed two liquidity measures – the bank liquidity mismatch index (BLMI) and the aggregate liquidity mismatch index (ALMI) – whose performances were compared to and contrasted with the Basel III liquidity measures and traditional liquidity measures using a generalised method of moments (GMM) model. Overall, the two constructed liquidity indices performed better than other liquidity measures. Significantly, the ALMI provided a better macro-prudential liquidity measure that can be utilised in dynamic stochastic general equilibrium (DSGE) models, thus presenting a major contribution to the body of knowledge. Unlike the LMI, the BLMI and ALMI can be used to evaluate the liquidity of a given bank under liquidity stress events, which are scaled by theoretically motivated and empirically supported liquidity weights. The constructed BLMI contains information regarding the liquidity risk within the context of asset liability mismatches, and the measure used comprehensive data from bank balance sheets and from financial market measures. The newly developed liquidity measures are based on portfolio management theory as they account for the significance of liquidity spirals. Empirical results show that banks increase their liquidity buffers during times of turmoil as both BLMI and ALMI improved during the period 2007–2009. Subsequently, the improvement in economic performance resulted in a rise in ALMI but a decrease in BLMI. We found no evidence to support the theory that banks, which heavily depend on external funding, end up in serious liquidity problems. The findings imply that any policy implemented with the intention of increasing bank capital is good for bank liquidity since the financial fragility–crowding-out hypothesis is outweighed by the risk absorption hypothesis because the relationship between capital and bank liquidity is positive.