Competition in the Turkish banking industry

Abstract

This study employs the Panzar-Rosse H-statistic to assess the competitive environment of the Turkish banking industry over the period 1990 to 2000. The results indicate that for the period under consideration bank revenues behaved as if they were earned under conditions of monopolistic competition. Therefore, the observed high profitability of the Turkish banking sector does not seem to be an indication of an increase in monopoly power. The liberalization process and deregulation measures appear to have beneficial effects on competition.

Notes

¹ The high number of banks was primarily a result of entry due to excess profits and low efficiency of incumbent banks. Beginning from the second half of 1990s, the exits from the sector have also slowed down, contributing positively to the increased number of banks. Inefficient firms managed to survive owing to the fact that a heavy government subsidy in the form of a 100% deposit insurance was established during the economic crisis of 1994 (Akçay et al., 2001).

² A comprehensive survey of econometric methodologies allowing the identification of the degree of competition is given by Bresnahan (1989).

³ For details of the formal derivation of the H-statistic, see Panzar and Rosse (1987) and Vesala (1995).

⁴ Shaffer (1982) shows that the H-statistic is also equal to one for a natural monopoly operating in a perfectly contestable market or for a firm that maximizes sales subject to a breakeven constraint.

⁵ A limitation of the P-R approach is that the increasing relationship between H and competition may not hold in certain oligopoly equilibriums. See Panzar and Rosse (1987) and Vesala (1995) for more on this issue.

⁶ Inputs are determined according to the so-called intermediation approach to banking where banks are viewed mainly as financial intermediaries. The extension of the P-R methodology to banking requires one to assume that banks are single product firms which is consistent with the intermediation approach (De Bandt and Davis, 2000).

⁷ At this point, it should be recognized that the typical fixed asset investments for a banking firm take up a very small portion of its total asset portfolio. In addition, the major ‘resource’ for a bank is the funds that it collects in the form of deposits, borrowings and equity. Therefore, the PK ratio may not contribute a lot to the explanatory power of the model when the estimation is carried out for a banking firm rather than a typical manufacturing firm. Nevertheless, the PK variable is kept in the model to provide comparability with the previous studies in the literature.

⁸ Serial correlation appears to be a problem in estimations. In order to correct for this problem, the error term was modelled as an AR(1) process.

⁹ The results of the F-tests for fixed effects and time dummies are available from the authors upon request.