The Panzar-Rosse Revenue Test and Market Power in Banking: An Empirical Illustration

Although the theoretical literature has shown that the Panzar–Rosse H-statistic fails as a measure of market power, it is still a widely used statistic in empirical banking studies. Such studies still rely on the erroneous belief that H > 0 is inconsistent with significant market power. This chapter provides empirical evidence against the latter belief by analyzing a US banking duopoly. We find “competitive” estimates of H but, consistent with a priori expectations, non-competitive outcomes according to an alternate measure of competition, the Lerner index. Moreover, our bank-specific estimates of H are mutually inconsistent, suggesting additional problems with the Panzar–Rosse test.