Tongues are aflutter over how search darling Google might conduct its
expected initial public offering — while Google has kept a safe distance from the chatter.
The IPO, which could raise as much as $4 billion, has provided months of panting speculation about its date, terms, size and method. Will Google use the so-called book-building method, in which lead investment banks set the price behind closed doors, then use shares to reward favorite clients? Or will it embrace the spirit of the Internet and open its arms — and its IPO shares — to all the little guys that love it?
The questions are just the latest parlor game in the Internet industry over how the company would float the shares. Will Google use the Dutch auction method championed in the U.S. by investment firm WR Hambrecht & Co.? By auctioning all or part of the opening shares, at least individual investors would get a shot at a piece of the offering.
The idea of a share auction is particularly apropos for Google, which has done well by doing well for a host of small advertisers. Its AdSense program, for example, provides a level playing field for advertisers with its open bidding system on placement costs. As a result, Martha Stewart and your mom have an equal shot at getting the top rank in paid listings returned in a search for “apple pie.”
In the book-building process used in a traditional public stock offering, however, bankers have been known to set the initial price below what they think the stock could fetch, so that those who buy IPO shares are sure to profit.
“They leave a lot of money on the table, which ingratiates the
underwriters with their buy side clients but doesn’t give the issuing firm as much proceeds,” said Jay Ritter, Cordell Professor of Finance at the University of Florida. Nor are the bargain-priced shares available to all of the underwriters’ clients. “If Martha Stewart wants the shares, she’ll get
them,” he said. But an investment bank’s rank-and-file customers won’t be in the queue.
In Hambrecht’s version of the auction model, called OpenIPO, would-be investors bid for a set number of shares at a set price. When the bidding is over, OpenIPO finds the “clearing price,” the highest price at which it can sell all the shares. That becomes the maximum offering price for all shares.
All bids are treated equally; the share allocation is based on price and has nothing to do with the size of the order or the bidder’s relationship with the underwriter. Those who bid at or above the final price will get shares; anyone who bids below the final offering price is out of luck.
The Dutch auction has two advantages over book-building, proponents say. More money goes into the coffers of the issuing company, and it’s more fair, because small investors have an equal chance at getting IPO shares.
“The traditional book-building system of going out and polling out a few investors does a terrible job of pricing,” said Clay Corbus, Hambrecht’s senior managing director, who said some hot offerings shoot up as much as 40 percent the first day. “Nor does it give you the correct allocations.”
Genitope , a Redwood City startup working on
personalized cancer vaccines, is the most recent company to go public using OpenIPO. Its October 2003 offering raised some $33.3 million. Former CFO
Fred Kurland (who resigned to join another company on January 29), told
internetnews.com the process worked to the company’s benefit.
“The very nature of OpenIPO almost forces a company to see what the
market is telling it about its value,” Kurland said. His company also liked that the process was literally an open book. “Unlike in the more classical process, Genitope was able to see ‘the book,’ the list of all the institutions and individuals who had made bids on the stock. In the classic
situation, the lead investment bank…would not reveal to anyone who was on the book.” Another major factor in the choice, he said, was that the company and the bank work in partnership to decide the price and to whom the stock would be sold. “We found that very compelling.”
Some people — Kurland and Corbus included — say that Dutch auctions lead to a more stable stock price, because investors aren’t getting the stock at an artificially low price and flipping it, or selling it as soon as the price jumps during initial trading. However, the performance of the nine stocks OpenIPO has taken public since 1999 is all over the map, with its best-performing offering, Ravenswood up 180.7 percent when it was acquired last year, and its worst-performing, Salon Media Group
Dutch auction proponents say that the process is inherently fair, because bids are open to anyone who has an account at the brokerage. “If you open an account with Hambrecht, you’re on an equal footing with their best customers,” said Ritter. “If you’re willing to pay a dollar more than Martha Stewart [for an IPO share], you’re more important.”
In most cases, said Ritter, “The issuing company is probably going to be able to raise at least as much money with an auction as with book-building. If it’s a hot deal, an auction will raise more.”
While auction IPOs haven’t gotten traction in the U.S., they’re fairly common in other countries, according to Ann Sherman, a finance professor at the University of Notre Dame in Indiana. In 1993, she said, Singapore Telecom raised about US$2.7 billion in a hybrid Dutch auction/book-building IPO, while an Argentina Telecom auction raised about US$1 billion in 1992.
“If Google were to do a Hambrecht auction, it would have a good chance of setting off a bidding frenzy that would bring the company much more than it would raise otherwise,” Sherman said. “But, as with so many Dutch IPO auctions in the past, the price would probably be unsustainable and would fall later, leading to big losses for the many small investors that might be buying stock for the first time because of their enthusiasm for Google.”
Sherman thinks Google should pursue a hybrid model, selling some shares at auction and the rest through book building. She said the hybrid model minimizes the problem of what she calls “free riders.” Because all bidders pay the final clearing price, not the price they actually bid, free riders drastically overbid for shares in an auction to ensure they get some. They can really mess up the price, she said.
In the case of OpenIPO, though, brokers review each bid and, if one seems out of whack, they will contact the bidder. They may even reject a bid if they feel the bidder hasn’t performed due diligence. “It’s not like eBay,” Corbus said.
There are other reasons why even an innovator like Google might find it wise to go the old-school route. The prevailing wisdom, according to Corbus, is that an IPO is like healthcare for the kids: Not a good time for experimentation.
“People want to make sure it’s a system that works and delivers what it’s supposed to,” he said. At the same time, he pointed out, while business has been transformed by technology, the rules of investment banking haven’t changed since they were set back in 1933. “There are entrenched pattern of behavior, and it takes a while to break out of those.”
And Wall Street has its ways. Tom Taulli, a USC professor and author of “Investing in IPOs,” said the analyst coverage and research reports issued by big banks can be incredibly influential. While Hambrecht is a full-service brokerage with it own analysts, “a bank like Goldman Sachs has
an incredible network of investors, and they probably see every deal you
could see.” On the other hand, “Banks that feel dissed may tell their
brokers not to push the stock quite as much or tell their analysts not to be
as positive.” Despite SEC reforms, Taulli said, “That’s the way Wall Street
operates.”
In coming to a decision, Google’s top management may be having a bit of a
culture clash, according to Taulli. Google’s founders, Sergey Brin and Larry
Page, may want to set their IPO loose on the Internet, while CEO Eric
Schmidt is pushing the big bank route.
“He’s a very seasoned CEO who really knows Wall Street — and knows it’s
a symbiotic relationship,” Taulli said.
The Internet has its own symbiosis — and its own system of rewards. If
Google does decide to auction all or part of its IPO shares, it will be
engaging in a high-stakes test of the Internet model. Better yet, it will be
putting its money where its mouth is.
