Search

It has been suggested that monopolistically structured banking markets will perform competitively because banks have so many potential competitors and these markets are contestable. This hypothesis is tested on 2,861 banks in 1,286 markets. The tests compare rates of return of banks in markets with one, two, three and four banks with similar sized banks in metropolitan markets. The findings clearly reveal that rates of return in markets with few banks are significantly higher than those in markets with many competitors. These results question the validity of the theory of contestable markets in banking.

Contestability in local banking markets is tested using a structure-performance (S-P) model. OLS and TSLS tests using a 38 state sample of 5,396 banks in 1,474 markets reveal a positive concentration-profitability relationship. This implies that local banking markets are not contestable. However, tests for each state separately show that the S-P relationship exists only in approximately one-fifth of states. F-tests indicate that using a national pooled sample is inappropriate. The results suggest that market structure, including entry conditions, affect performance, consequently the theory of potential competition is more relevant than contestability theory. However, generalizations may be inaccurate.