Università degli Studi di Salerno; CSEF c Università degli Studi di Salerno ABSTRACT
In this paper we analyse the efficiency differentials among Italian small banks located in different geographical areas and belonging totwo great institutional categories: BC's and other banks. We apply DEA throughout the 1996-2004 period, highlighting the effect of some environmental factors (asset quality, capitalisation, local GDP shocks) on banks' performance. The evidence shows that that local shocks, as proxied by value added per capita, affect the territorial differentials for technical efficiency, especially for BC's.This can be easily rationalised, as current regulations hamper BC's vis-à-vis other banks in their capability to diversify territorially. Our estimates provide us with a tentative quantitative measure of the costs of missing diversification, ranging between 2 and 5 percentage points. On the other hand our evidence suggests that there is potentially strong endogeneity in some currently available bankperformance indicators.
PROVISIONAL DRAFT, PLEASE DO NOT QUOTE WITHOUT AUTHORS' PERMISSION 1. Introduction In the literature concerned with the quantitative evaluation of bank performance the themes of regulation and proprietary forms have always played a powerful role in shaping the analysis of the determinants of bank efficiency (Berger et al., 2008, 2010). These factors have almostinvariably been taken in account without explicit allowance for changes in the socio-economic environment of banks. The latter have been often associated, however, with the treatment of risk and of risk management within the productive process of banks. The aim of this paper is to bring together the two strands of the banking literature dealing with regulation and risk, within a frontier efficiencyanalysis of Italian small banks. As a matter of fact, we focus on Italian cooperative banks, whose regulatory structure is particularly suited to the analysis of the interaction between regulation and risk. Other Italian small banks will mainly be considered for purposes of comparison. We believe that our analysis may be of relevance, not only because European cooperative banks have recently spurred aconsiderable policy interest (see, for instance, Fonteyne, 2007), but also because we produce some quantitative estimates of the impact of (missing) risk diversification upon bank efficiency. Estimates of this kind are not yet widely available, although there are some similarities between our analysis and those of Hayden et al. (2006) and of Rossi et al. (2009). It is well known in the literature(Hughes and Mester, 1993, 1994) that in order to measure
accurately bank efficiency, proper account must be taken for the role of output quality. For instance, a scant level of credit screening or monitoring may induce savings in the resource invested by banks at the cost of poor loan quality. Hence, a higher level of cost efficiency would turn out to be associated with poorer performance, andvice versa. Equally, if bank managers are not risk neutral, their degree of risk aversion is likely to affect, in a stochastic environment, their choices about the production set. Loosely speaking, a higher level of equity reduces insolvency risk, urging riskaverse managers to rely upon an equity level larger than what would be otherwise needed. Once again, were risk aversion not allowed for, theresulting production set could be described as costinefficient. Generally speaking, these considerations lead researchers interested in the correct measurement of bank efficiency to include in their analysis not only the usual array of bank inputs and outputs, but also indicators of credit quality and risk aversion (Hughes and Mester, 1994; Rossi et al., 2009; Fiordelisi et al., 2010). Similarly,...