Why Have Auctions Been Losing Market Shares to Bookbuilding in IPO Markets?

This paper analyzes bidding data of IPO auctions held in Taiwan to address three questions. First, several studies on bookbuilding IPOs have shown that allocation policies favor institutional investors. Would auctions be a fairer mechanism to retail investors? Second, Sherman (2002) argues that bookbuilding has an edge over auctions in IPO pricing. What empirical evidence can be offered to suggest that auctions may not perform as well as bookbuilding? Third, Benveniste and Spindt (1989) posit that, compared to auctions, bookbuilding could increase institutional investors’ expected profits with the tradeoff of a higher allocation of shares and a lower level of underpricing. This raises the third empirical question: Do observed institutional allocations and the underpricing required by institutional investors in auctions make the tradeoff feasible? We find that retail investors win significantly higher proportions of IPO shares in auctions with negative initial returns. On the other hand, the more the shares won by institutional investors, the higher the initial returns they earn. The allocation-weighted abnormal return for retail investors is insignificantly different from zero, while the allocation-weighted abnormal return for institutions is a significant 10.5%. The results are consistent with Rock’s (1986) winner’s curse theory, and suggest that retail investors are not better off under auctions than under bookbuilding. While Ljungqvist and Wilhelm (2002) show that underpricing decreases with institutional allocations in bookbuilding IPOs, we find the opposite in auctions. The results imply that the pricing performance of auctions for IPOs that attract a great deal of institutional interest might be improved by using the bookbuilding method. Finally, we show that even though institutional investors have advantages in auctions, they could be better off under bookbuilding. The observed average institutional allocation of 19% and their required underpricing of 10.5% in auctions leave much room for the tradeoff suggested by Benveniste and Spindt (1989). In sum, our study contributes three reasons for why auctions have been losing market shares to bookbuilding in IPO markets. Why Have Auctions Been Losing Market Shares to Bookbuilding in IPO Markets?