Competition and market structure in the South African banking industry

• Main Author: Tsegay, Yared Teka
• Other Authors: Black, P. A.
• Language: en
• Published: University of the Western Cape 2021
• Subjects: South African; Panzar-Rosse; Monopolistic competition; Banking sector; Mmethodology
• Online Access: http://hdl.handle.net/11394/7884

Magister Commercii - MCom === The South African banking industry is relatively sound and adequately capitalized high-tech service industry. It provides services to 51% of the economically active population. In the late 2001, 2002 and,2003, a number of small banks failed·due to liquidity crises, which subsequently spurred the momentum of consolidation in the industry. During this period, a great deal of mergers and acquisitions has taken place in the industry. The regulatory environment is in line with the Bank of International Settlements (BIS) guidelines. The regulations regarding foreign banks are evidently a barrier to foreign bank entry, which could spur the level of competition in the industry In this thesis the level of market concentration is calculated using concentration ratios, CR. and CRs, for four product markets. The concentration ratio R. ranges from 77% for investment products, 79% for deposit market, 89% for installment sales market to 95% for credit card markets. The Herfindahl-Hirschman Index (RH!) as calculated by reserve bank is 1750 by the end of year 2002, compared with the thresholds set by American Department of Justice; this falls short by only 50 from the range of the highly concentrated markets, which is above 1800 points. The methodology used to test competition, one of the new empirical industrial organization models, is the Panzar-Rosse reduced form revenue function. The P-R reduced form revenue function estimates the competitive conduct by the extent to which changes in factor price is reflected in revenue. The reduced form revenue function is estimated based on unbalanced panel of 15 banks for the period of 1993 to 2002 using fixed effects panel data econometrics. The estimated H-statistic is 0.516,