WP 2004/8 Horváth Csilla - Krekó Judit - Naszódi Anna: Interest rate pass-through in HungaryPrint
This paper studies the pass-through from the short-term money market rate to various forint denominated bank rates in Hungary between 1997-2004. The analysis is based on linear and non-linear error correction models (ECMs), using both aggregated and bank level data. According to the linear ECM results, corporate loan rates adjust to the market rate completely and reasonably fast, whereas the adjustment of the deposit rates and household loan rates is characterised by incompleteness and/or sluggishness.
We analysed the potential non-linearities of banks’ pricing by TAR (threshold autoregressive) models. The results suggest that the speed of adjustment of bank rates depends on the size of the changes in the money market rate and the distance of the bank rates from their long-term equilibrium. We found the adjustment to be significantly faster for changes above a threshold level than for smaller ones. This phenomenon can be explained by the presence of menu-costs. The sign of yield shocks also turned out to be influential to the speed of adjustment. In line with international experience, we found that corporate loan rates are characterised by downward rigidity, in accordance with the profit maximisation behaviour of banks. Surprisingly, the sharp competition in the corporate loan segment could not fully counterbalance the downward rigidity. We also found that household deposit rates adjust more rapidly to upward than to downward shifts of the market rate. This seemingly counterintuitive finding can be explained by the fact that the average size of positive shocks exceeded the average size of negative ones in the sample period. We also analysed how the volatility of money market rate affects the pass-through. At least one of the parameters determining the speed of adjustment changed towards faster adjustment when the volatility of the market rate exceeded a certain level. Intuitively, higher volatility should be accompanied by higher uncertainty and hence more sluggish adjustment; however, high volatility can be the consequence of huge rate shocks, which are not negligible by the banks. The size-effect and the effect of uncertainty are hardly separable. We think that in the volatile periods, especially in 2003, the effect of uncertainty was dominated by the size-effect. However, the faster pass-through could not offset the effect of higher money market rate shocks completely, which resulted in higher volatility of the spread between bank rates and money market rate.