Non-Interest Income and Bank Efficiency in Ghana: A Two-Stage DEA Bootstrapping Approach

Abstract

This paper examines the effect of the inclusion of non-interest income on efficiency and economies of scale of Ghanaian banks in a two-stage analysis. The data envelopment analysis technique is employed to estimate efficiency scores with and without non-interest income of 26 Ghanaian banks from 2003 to 2011 in the first-stage analysis. In the second stage, a truncated bootstrapped regression is estimated to examine the effect of contextual variables on bootstrapped efficiency scores. The findings indicate that the exclusion of non-interest income as output variable leads to the under-estimation of efficiency scores. From the second-stage regression analysis, we find a curve-linear relationship between bank size and efficiency to suggest that bank efficiency increases as size also increase due to economies of scale but only up to an optimal point after which inefficiency sets. Market concentration, leverage, and loan loss provisions are also identified as the other significant determinants of efficiency. Policy implications for improving bank efficiency are discussed.