Working Paper: NBER ID: w16214
Authors: Ravi Jagannathan; Andrei Jirnyi; Ann Sherman
Abstract: In this paper we present a comprehensive comparison of IPO placement methods in over 50 countries. We find that out of the three primary methods, fixed price public offers, auctions, and book building, auctions are least popular with issuers. Since auctions allow for price discovery while avoiding the potential conflict of interest between issuer and underwriter, this is a surprising finding that is not adequately explained in the existing literature. We propose a new explanation: namely, that participating in auctions is substantially more difficult for investors compared to the other methods, and that this complexity can lead to investor behavior that is undesirable for the issuer. We suggest that this effect could be mitigated through a hybrid mechanism that resembles the one that is used in US treasury auctions.
Keywords: Initial Public Offerings; IPO Auctions; Book Building; Mechanism Design
JEL Codes: D02; D44; G12; G24; G3; H0; H10; H26; H29
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Complexity of auction participation (D44) | Undesirable investor behavior for issuers (G24) |
| Complexity of auction participation (D44) | Deterrence of issuers from using auctions (D44) |
| Preference for book building (D44) | Not solely due to underwriter market power (D49) |
| Long-term relationships with underwriters (L14) | Incentivizes truthful revelation of valuations (G19) |
| Lack of transparency in book building (G10) | Does not fully account for its dominance over auctions (D44) |