A Note on Contestability in the Canadian Banking Industry

The authors examine the degree of contestability in the Canadian banking system using the H-statistic proposed by Panzar and Rosse (1987) and modified by Bikker, Spierdijk, and Finnie (2006). A modification is necessary because the standard approach of controlling for size using total assets leads to an upward bias in the H-statistic. The authors propose a variety of model specifications and test for contestability using detailed quarterly balance-sheet data from 2000 to 2006. Contrary to Bikker, Spierdijk, and Finnie (2006), the authors find that the Canadian banking sector is in equilibrium and characterized by monopolistic competition. This result is in line with earlier studies of the Canadian banking sector (Nathan and Neave 1989) as well as cross-country studies that use cruder measures of Canadian banking inputs (Claessens and Laeven 2005). As in Bikker, Spierdijk, and Finnie (2006), the authors show that projecting revenue on total assets leads to an upward bias regarding the level of competition.