Academic journal article The Journal of Developing Areas

Does Ownership Affect the Efficiency of African Banks?*

Academic journal article The Journal of Developing Areas

Does Ownership Affect the Efficiency of African Banks?*

Article excerpt


In the last few years, there has been an extensive debate as to whether ownership matters for bank performance in less developed countries. This paper investigates whether privately-owned banks outperform state-owned banks and whether foreign ownership enhances bank performance. Based on a range of performance ratios as well as parametric and non-parametric estimations, the results show that in Africa, on average, privately-owned banks do not appear to outperform state-owned banks. However, where private ownership involves foreign ownership then this does seem to have a positive effect on bank performance. Both sets of results are affected by high variance in the data suggesting that in state-owned and privately-owned banks and in domestic and foreign-owned banks there are widely differing levels of efficiency. In addition, we also test for the effects of ownership taking into account environmental, including regulatory, variables. The study reports results for banking across Africa and in two separate regions, North Africa and sub-Saharan Africa. We also test for country-level effects.

JEL Classification: G21, G32, N26

Keywords: Africa, Banking Efficiency, Ownership, Performance

(ProQuest-CSA LLC: ... denotes formulae omitted.)


In the late 1980s and early 1990s, a number of African countries began to restructure their financial sectors in order to boost banking efficiency (Brownbridge and Harvey, 1998). Major reforms were introduced, such as the privatization of state-owned banks, the abolition of entry barriers to foreign capital, the removal of interest rate controls, and the development of new regulatory frameworks aimed at limiting opportunities to commit fraud and abuse depositors' funds. At the same time, a growing literature has developed on the impact of ownership on economic performance, including the banking sector, although mainly in developed countries (for a review, see Megginson and Netter, 2001).

This paper investigates the performance of banks in Africa and specifically to what extent the ownership structure of banks affects their performance. It also introduces variables to capture the effects of regulatory structures on bank efficiency. The analysis is cross-sectional and includes an exceptionally large data set for 2001/02. While a cross sectional analysis of banking performance has limitations, it provides an insight into the results of reforms in ownership during the 1990s. A time series analysis to chart the impact of ownership change on performance since the 1980s was ruled out by a lack of reliable historical data, especially on the ownership of banks. Nevertheless, if ownership changes introduced from the 1980s had a significant impact on bank performance, we should expect this to be reflected in the relative performance of banks under different forms of ownership in 2001/2. The study considers relative performance under different forms of ownership for banks across the African continent as a whole and for two specific regions within Africa, North Africa and sub-Saharan Africa. We also test for country-level effects. In total, 40 African countries are included in the sample.

The paper is structured as follows. The next section discusses the existing literature on bank ownership and performance and then sets out the main research hypotheses tested in the paper. This is followed by a description of the data and performance measures used in the study. The empirical analysis and findings are then presented. Three methods are used to assess performance, namely the use of bank performance ratios, a data envelopment model and a stochastic cost frontier analysis. The results from each method are compared to assess the robustness of the results. Where different measures provide the same or similar results, we should have more confidence in them (Bauer et al., 1998). In addition, a switching regression model is also applied to the performance results in order to test whether or not the performance of banks is affected (a) by ownership - private versus state and domestic versus foreign, and (b) by industry and country-specific characteristics, as well as regulatory governance variables. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.