Market power in the Swedish banking oligopoly : a game-theoretic model of competition applied to the five big Swedish banks 1989-97

Sammanfattning: This thesis aims at assessing the degree of competition among the five dominating Swedish banks during the period 1989-1997. In so doing, it also aims to develop a general oligopoly model with the specific purpose of providing a tool for an improved methodology for the evaluation of market power in oligopolistic markets, and to demonstrate the empirical application of this methodology.In the first of the three essays, An Asymmetric Oligopoly Model and a Method for Its Empirical Application, a general oligopoly model is developed that is suitable for empirical research where the data set contains firm-level data on prices, quantities and costs. This model makes it possible to gauge empirically observed prices to a range of game-theoretic equilibrium prices that are analytically deducted. It also allows for precise calculations of welfare effects from non-competitive pricing.In the second essay, Do Swedish Banks Enjoy Economies of Scale or Economies of Scope?, the extent of production economies on the cost side in the five big Swedish banks are investigated. This essay complements the empirical study of Swedish banking by, inter alia, estimating a cost function that includes not only operating costs, but also opportunity costs of equity capital. Results imply small dis-economies of scope between deposits and loans for the banks, as well as slightly negative economies of scale.In the third essay, Testing for Market Power in the Swedish Banking Oligopoly, the model in the first essay is developed into a banking context, allowing for the inherent multi-product characteristic in that industry. In the empirical part, results show that there was significant market power in both the loan market and the deposit market, although with a strongly time-varying pattern. The overall picture of conduct, is that pricing policies were less competitive in the deposit market. The economic cost level consitutes a bottom level for pricing in the loan market and attempts to establish higher spreads are not sustainable in the long run. Finally, welfare losses to the society from non-competitive pricing of loans and deposits are calculated to be ca 1.1% of GDP as a yearly average during the sampling period. The combined results of the empirical analyses in the second and third essays, indicate that there are no cost-side arguments for the many mergers and acquisitions that have taken place in the industry proper, and that small, specialised banking institutions might be competitively viable.