Why Do IPO Auctions Fail?
Social Science Research Network
We document a somewhat surprising regularity: of the many countries that have used IPO auctions, virtually all have abandoned them. The common explanations given for the lack of popularity of the auction method in the US, viz., issuer reluctance to try a new experimental method, and underwriter pressure towards methods that lead to higher fees, do not fit the evidence. We examine why auctions have failed and verify, to the extent possible, that they are consistent with what academic theory
... cademic theory predicts. Both uniform price and discriminatory auctions are plagued by unexpectedly large fluctuations in the number of participants. The free rider problem and the winner's curse hamper price discovery and discourage investors from participating in auctions. Calculating the optimal bids in large multi-unit common value auctions with endogenous entry imposes a huge computational burden. With IPOs taking place sporadically, and each firm being different, auctions are likely to end up being unstable. JEL Categories: G24, G28, G32 "Improbable as it is, all other explanations are more improbable still." Sherlock Holmes in "Sliver Blaze," 1892, by Sir Arthur Conan Doyle Book building is the primary method through which initial public offerings (IPOs) are brought to the market in the United States (US). An ongoing debate in the academic literature explores the advantages and disadvantages of the book building method, relative to sealed bid auctions. On the one hand, the greater control and flexibility of book building provides substantial benefits to issuers. 1 On the other hand, the book building procedure necessarily gives the underwriter substantial discretion over allocations. When agents are given discretion, there is always the potential for abuse, and the scandals following the internet bubble suggest that at least some abuses have occurred in practice 2 . Moreover, there is a general agency problem between underwriters and issuers that has not yet been fully explored for IPOs. 3 Thus there are both advantages and disadvantages to the flexibility offered by book building. In the search for an alternative, much of the focus has been on auctions, which have been extremely successful in a wide range of alternative settings. With sealed bid auctions, theory also offers trade-offs -auction theory predicts that sealed bid auctions will lead to very accurate pricing under some circumstances but to substantial problems under others. In this case, the theoretical differences are in the underlying assumptions regarding information structures and the determinants of entry. If information is endowed (i.e. costless) and bidder entry is predictable, auctions should be relatively efficient. But if accurate estimates of IPO share values are difficult to produce and entry is uncoordinated, theory predicts that auction outcomes may be far less desirable (see Sherman, 2005) . Because theory predicts varying outcomes for both auction and book building IPOs, it is worth examining the available evidence regarding the track records of each method. In this paper we offer evidence on overall usage patterns for many countries -the 'market test' -and then examine IPO auction outcomes in more detail. We find that, when standard auctions have 1 As first shown by Benveniste and Spindt (1989) and Benveniste and Wilhelm (1990). Ritter and Welch (2002) , Ljungqvist (2004) and Wilhelm (2005) offer reviews of the academic IPO literature. 2 See Loughran and Ritter (2004) for discussion of the scandals and overall trends in IPO underwriting. 3 with the notable except of Biais, Bossaerts and Rochet (2002), for the French regulatory regime. 2 had to compete with another method -either with fixed price public offers 4 or with book building -auctions have been driven out. The lack of popularity of auctions cannot be explained by either lack of familiarity or by differences in underwriting fees. The fees for fixed price public offers in most countries have been the same as those for auctions, leaving investment banks with no incentive to favor one method over the other based on fees. In spite of that, when issuers have been allowed to choose between fixed price public offers and auctions, the former method has prevailed and auctions have lost out 5 . And when fixed price public offers later were faced with competition from book building, the fixed price public offer method has generally lost out, although not as completely as the auction method. The observation that auctions have consistently lost out to other methods is an important piece of evidence but is not, by itself, sufficient to conclude that the predictions of auction theory are correct. We therefore examine the reasons why auctions have failed and verify, to the extent possible, that they are consistent with auction theory in an IPO setting. The auction method is old and well established, and has been particularly successful for the largest security issue markets -those for government debt, particularly US Treasury securities; and auctions have been frequently used for new preferred stock issues in the United Kingdom (UK), particularly for government-owned utilities 6 . Treasury auctions are held frequently at regular time intervals, with a core of regular participants. Further, close substitutes to the securities being issued are already trading actively in the market, making valuation relatively easy and precise 7 . Preferred stocks of regulated utilities are relatively easy to value since they resemble high quality bonds. In contrast, IPOs occur less frequently, at sporadic intervals, and their value is difficult to determine. Each issue is different and may 4 With fixed price public offers, the price is set before any information on demand is received, as shown by Loughran, Ritter and Rydqvist (1994, Table 2 ). With book building (a term coined in the 1990s), the underwriter arranges for investors to attend a road show and then collects indications of interest, which are used to fill (build) the order book. The offering price is set only after the order book is full, giving the underwriter some idea of demand. With standard auctions, pricing and allocation are based on bids, using pre-established rules. Sherman (2005) argues that the main difference between the methods, from a regulatory standpoint, is the underwriter's discretion over allocations with book building. With either fixed price public offers or sealed bid auctions, underwriters may, and sometimes do, hold road shows before the offer price is set. They are allowed to ask for feedback but, without control over allocations, they cannot give investors an incentive to offer reliable feedback. 5 The only exception that we know of is France, which used a unique auction method that discouraged free riders. 6 In the six month period from Oct.