Abstract
The literature examining the role of macroeconomic variables in explaining the significant amount of heterogeneity in the pass through from market rate to retail rates among countries has done so only using cross-section data and has found four major explanatory variables: inflation, market rate volatility, credit risk, and GDP. The major drawback with these studies is that they fail to account for existence of country and time-fixed effects. As a result, the estimated coefficients on the relationship between macroeconomic variables and interest rate pass through may be biased. Such bias in these coefficients will lead to under/over reaction of the central bank in adjusting its feedback parameters. In this paper we employ a fixed effects panel data model to estimate an unbiased relationship between policy rate and retail rates. In particular, we use data from 42 countries covering the period 2003-2012 and construct a panel dividing the data into two periods (period 1: 2003-2007 and period 2: 2008-2012). We summarize our findings in this paper as follows. First, through we find that higher inflation and lower policy rate volatiliy is associated with higher lending pass through in both cross-section and fixed effects estimation, the coefficients on these variables in the fixed effects regression are bigger in absolute value than those of the cross-section regression. We therefore conclude that there are country-fixed effects that lead to biased estimated coefficients in the cross section regression. Second, the effect of fixed effects is stronger in the relationship between macroeconomic variables and deposit rate pass through. For instance, both high GDP and high inflation are associated with higher deposit rate pass through in our cross-section regression. But both variables are statistically insignificant in the fixed effects regression. In fact, the only variable with explanatory power for deposit rate pass through in the fixed effects model is policy rate volatility. Specifically, higher variation in the policy rate leads to lower pass through from the policy rate to the deposit rate. Third, we find that credit market risk and time-fixed effects do not explain the pass through from policy rate in either of the estimation methods.