Estimating the value and interest rate risk of demand deposits in concentrated markets

Abstract

PURPOSE OF THE STUDY

The purpose of this study is to determine the value and interest rate risk of funds deposited in demand deposit accounts under imperfect competition among banks. The value of a demand deposit is divided into two components, which are rent and liability. The former is defined as the profit bank receives from accepting demand deposits (by paying rates below the short-term market interest rate) and the latter as the nominal value of deposits minus the rent. The interest rate risk of demand deposits is measured by their sensitivity to shocks in the short-term market interest rate. The analysis in this thesis is carried out from the viewpoint of a case bank, which is a Finnish commercial bank, and the Finnish banking sector as a whole.

DATA AND METHODOLOGY

Historical data is needed in this thesis in order to estimate the demand functions for deposits and the processes of the variables. Most of the data series span from January 2006 to December 2010, totaling 60 monthly observations. The data was obtained from three sources: the case bank’s databases, Bank of Finland, and Statistics Finland. Monte Carlo simulation is used in generating the value and interest rate risk estimates. A majority of the variables are modeled as AR(2)-processes, whereas the short-term market interest rate is modeled using a one-factor stochastic Cox-Ingersoll-Ross model. Moreover, various assumptions concerning deposit balance dynamics are taken into account in the analysis of case bank, whereas the analysis of the whole banking sector is carried out only under AR(2) forecasted balances.

RESULTS

The results indicate that several variables measuring macroeconomic environment and market concentration play an important role in determining the demand function for demand deposits. Also, it is found that both the case bank and the Finnish banking sector as a whole exercise market power, as both of them are able to generate significant positive rents from accepting demand deposits. However, the magnitude of these rents varies a lot depending on the assumed deposit balance dynamics. The largest rent estimates are obtained assuming that future deposit balances evolve according to AR(2) forecasts, whereas under constant and decaying balances the rents are substantially lower. Finally, I find the interest rate risk of demand deposits to be significant, as their valuations are sensitive to short-term market rate shocks under all deposit balance dynamics covered.