Dutch Auctions and Democracy: Two Things That Don’t Make Sense For IPOs

Almost all post mortem accounts of Google’s now infamous IPO almost reflexively refer to the event as Google’s “Dutch Auction” IPO. Indeed many proponents of Dutch Auctions are now holding out Google’s IPO as supposedly some kind of validating watershed that will lead to increasing use of Dutch Auctions for IPOs (just take a look at the article on WR Hambrecht today in the San Francisco Chronicle). There’s even a new start-up trying to build an entire underwriting business around conducting Dutch Auctions on the American Stock Exchange. The only problem with this hubbub is that, no matter how you look at it, Google’s IPO was clearly not a Dutch Auction. The most egregious violations of Dutch Auction etiquette perpetrated by Google include:

1. Google didn’t fill 100% of the orders. In Dutch auctions, almost all the bidders above the clearing price receive 100% of the shares they bid on. Typically, the only people that don’t receive 100% of their orders are those that “tie” for the lowest bid. It has been widely reported that investors only got about 75% of the shares they bid for, regardless of what bid they made. This is a clear violation of the principles of a Dutch Auction.

2. Google changed the number of shares offered. In what is perhaps the most egregious violation of Dutch Auction principles, Google substantially reduced the amount of shares offered after the auction had already started. Dutch Auctions are supposed to have fixed inventories. Fixing the inventory assures investors that the issuer won’t be able to manipulate the clearing price by simply reducing inventory. This is of course exactly what Google did. They started the auction, didn’t like the way the winds were blowing and promptly cut the size of the offering by 24%. This may be good deal management by the underwriters, but it’s clearly not consistent with a Dutch Auction.

Despite being faced with the obvious fiction of Google’s “Dutch Auction” IPO, many pundits and policy wonks still insist calling the IPO a “Dutch Auction” solely because it used a “democratic” means of allocating shares. From this perspective, it doesn’t matter that Google bent a few rules, it just matters that the process was more “democratic” with the supposition being that a democratic distribution of shares is much better than one in which the underwriters and issuers hand pick which investors get shares.

Tyranny of Idiocy
The problem with this argument is that Dutch Auctions are not democratic, they are more like the “Tyranny of The Biggest Idiot” in that any idiot can get shares if they are willing to pay a high enough price. While having such an allocation process will clearly maximize short term pricing, it does not guarantee a strong syndicate of stable, long-term shareholders which should be one of the primary goals, if not the primary goal, of any IPO.

The painful truth for all the academics and Wall Street critics is that democracy isn’t necessarily a good thing when it comes to IPOs. The fact is that the old-fashioned method of allocating IPO’s is a far more reasoned and reliable way of allocating shares than Dutch Auctions because it gives issuers the discretion to allocate shares to specific investors. This flexibility allows the issuer to avoid giving shares to obvious flippers and to give more shares to investors that clearly want to be long-term shareholders.

It’s true that many banks take advantage of this discretion to reward prized clients preferred access to under-priced shares but in the grand scheme of things such “payoffs” are really just a franchise tax that issuers must pay to an underwriter in return for renting their brand. For every issuer that gets screwed by having a portion of its deal allocated to settle old-debts or curry new favors, there’s another issuer that get a huge benefit when an underwriter calls in its chits and gets a few big investors to step up to the plate and help them move a soft deal.

Issuer Beware
The ultimate solution to the problem of under-priced IPOs and underwriter pay-offs is not to blindly adopt Dutch Auctions as some kind of saving grace, but to better educate Issuers about their rights and responsibilities when it comes to determining the issue price and allocating shares. Issuers need to make sure that they stay on top of their order book and that they develop an opinion on which investors they’d really like to have at their annual meetings.

One potential solution that may bridge the gap between traditional syndicate book building and Dutch Auctions may be to simply use the Dutch Auction process as a non-binding way to improve the price discovery process that syndicate desks already do today. Rather than simply taking share orders, syndicate desks should just ask investors explicitly about pricing levels and then make this data available to issuers. The reality is this is basically what Morgan Stanley ended up doing in the Google IPO.

In the past underwriters have been reluctant to do this kind of research as such “price discovery” chit-chat could potentially be construed as an attempt to “ladder the deal”, which has been a major “No! No!” since 1934. Given that the SEC seems to be a big supporter of Dutch Auctions I don’t see why they would object to providing Wall Street with a safe harbor letter that would allow them to solicit such variable pricing data prior to going effective.

However things evolve, the mindless championing of Dutch Auctions and “democratic” allocations is not doing anything to improve the IPO process. Rather it is simply lulling issuers into believing that the only thing that matters about an IPO is getting the highest price possible. Getting a high price is important, but it’s also important to get a stable core of long-term shareholders and to avoid allocating shares to flippers that might needlessly destabilize a stock’s trading early on. Issuers and underwriters have to balance both of these needs and as result retaining the flexibility to allocate shares remains incredibly important no matter what the academics say.