A Replication of “Are Competitive Banking Systems More Stable?” (Journal of Money, Credit, and Banking, 2009)

This study replicates Schaeck, Čihák, and Wolfe (2009), henceforth SCW, and performs a variety of robustness checks. Using a cross-country, time series sample of 45 countries from 1980-2005, SCW investigate the relationship between competition and concentration in the banking system, and the occurrence of country-level systemic crises. Their primary measure of competition in the banking industry is Panzar and Rosse’s H-statistic. Concentration is measured using a concentration ratio of the three largest banks. They conclude that (i) competition and concentration measure two separate dimensions of the banking sector, and (ii) greater competition is associated with fewer systemic crises. Using data and code provided by the authors, we are able to exactly reproduce the original results of SCW. However, we find that their results are not generally robust. While we confirm their results on concentration, when we extend the data to the current period and use updated variable values, we find that competition, as measured by the H-statistic, is consistently insignificant across both duration and logit models.