Christian Leuz pointed out that in Germany's Neuer Markt issuers have a choice between auctioning off their shares or going through a more traditional underwriting process, but that out of the 350 initial public offerings since inception, only one chose to go with an auctioning mechanism. He also noted that since more than thirty lead banks were involved in these initial public offerings, one would assume that competition eventually would whittle down the kickbacks if any were involved. Hence perhaps the incentives of issuers and managements, issuer stupidity, or practical problems with the auction mechanisms could explain this lack of interest in auctions.
Leuz also questioned whether Ausubel's mechanism is coalition- or coercion-proof since the dominant strategy could be clinched by two or more firms acting together. Ausubel responded that if in his mechanism or in any auction procedure bidders were allowed to collude, this would change the incentive properties of the auction and could obviously degrade its performance. Hence any auction procedures he described need to be used in conjunction with good anticollusion mechanisms. A participant added that many institutional investors and individual investors are involved in buying initial public offerings, and hence the collusion of a small number of parties is not a likely scenario.
Reena Aggarwal commented that although academics treat initial public offerings as isolated events, in reality much more complex processes are involved and started before the offering takes place, and they continue with the research analysis, banking relationships for the future, margin advisers, and so on. She added that an isolated view would ignore these other important characteristics. [End Page 340]
David Crosen noted that a standard result of auction theory is that, in order to get bidders to reveal what their true values are, issuers have to leave some information rent on the table. He asked whether Ausubel had calculated how much information rent is actually being left on the table for bidders collectively and how efficient his auction might be compared to other auction mechanisms, in terms of raising money for the issuers rather than allocating it to the right people. He suggested that perhaps IPO issuers do not like auctions since they involve giving a lot of information rent to the people who are contracting for large quantities. Ausubel replied that his theorem assumes private information and that the mechanism is subject to the same incentive constraints on information, hence his results encompass information rents. He added that, in terms of seller revenues, it is not possible to raise any more revenues than is possible from an efficient allocation given that an efficient secondary market follows the auction.
In response to a question about WR Hambrecht auctions, Ausubel stated that there is nothing fundamentally defective about the procedure, although the current wisdom generally prefers ascending-bid auction formats over sealed-bid auction formats. He added, however, that there are problems, such as the fact that all bids are indicative and do not have any binding effect on bidders; he also noted that this might be due to Securities and Exchange Commission rules preventing someone from running an auction where the bidder can be directly held to the bid. He suggested that the Securities and Exchange Commission get rid of any rules or reinterpret those that might have the effect of dooming an auction from working properly.
One participant noted that if a firm with a good reputation does due diligence on the company, there is true value added, and this might explain why auctions should not take place. Ausubel responded that there is no reason why the issuing firm could not hire a reputable investment bank to do due diligence, compensate the bank for its services based on the work done, and then have an auction.
Another participant observed that in an auction large investment banks that would then market the securities, not retail investors, are the bidder. Ausubel responded that he envisions a system somewhere in between, where institutions are bidders as well as retail investors. [End Page 341]