Do underwriters compete in IPO pricing?

We propose and implement a direct test of the hypothesis of oligopolistic competition in the U.S. underwriting market against the alternative of implicit collusion among underwriters. We construct a simple model of interaction between heterogeneous underwriters and heterogeneous firms and solve it under two alternative assumptions: oligopolistic competition among underwriters and implicit collusion among them. The two solutions lead to different equilibrium relations between the compensation of underwriters of different quality on one hand and the time-varying demand for public incorporation on the other hand. Our empirical results, obtained using 39 years of initial public offering (IPO) data, are generally consistent with the implicit collusion hypothesis that banks, especially larger ones, seem to internalize the effects of their underwriting fees and IPO pricing on their rivals.