The pricing of initial public offerings: tests of adverse-selection and signaling theories

We test the empirical implications of several models of IPOunderpricing. Consistent with the winner's-curse hypothesis,we show that in markets where investors know a priori that theydo not have to compete with informed investors, IPOs are notunderpriced. We also show that IPOs underwritten by reputableinvestment banks experience significantly less underpricingand perform significantly better in the long run. We do notfind empirical support for the signaling models that try toexplain why firms underprice. In fact, we find that (1) firmsthat underprice more return to the reissue market less frequently,and for lesser amounts, than firms that underprice less, and(2) firms that underprice less experience bigger earnings andpay higher dividends, contrary to the models' predictions.