What Lessons can be learned from Google’s Dutch Auction?

Thomas W. Hildreth
Of Counsel, Corporate Department
Published: New Hampshire Business Review
October 15, 2004

There are many ways to measure an IPO. One commonly watched metric is the first day performance of the stock price once the shares begin trading on a public exchange. All market participants – the issuer, the underwriters, the investing public – are curious to compare the first day public market closing price to the price paid to the issuer and the selling insiders in the offering. A closing price substantially lower than an offering price would suggest that the buyers from the issuer overpaid. Imbalance in the other direction – a closing price substantially higher than an offering price – would suggest that the issuer and underwriters under priced the offering, in favor of those who bought from the issuer. Synchronicity between offer and close would suggest an offering that fairly and accurately valued the issue.

In a traditional firm commitment, underwritten public offering, investment banks hired by the issuer company gauge interest in a prospective offering normally through a series of formal private “road shows” played to institutional and substantial individual investors. In the traditional IPO, the underwriters normally buy the issue from the company at the final agreed upon price and then turn around and resell it to the public, often at a substantial markup over what the underwriters paid the issuer. Many IPOs are priced so that the stock price rises appreciably in the first day of trading providing fast profits for the underwriters and their best customers.

In the 1990s a number of companies saw meteoric rises in first day stock price. Yahoo shares rose from their IPO price of $13 a share to as high as $43 during the first day of trading. Netscape shares went up 108% on the first day of trading. The Globe.com shot up 600% on their first day. In 1999, the average first day return for those who received shares at an IPO price was 69%.

Recent offerings have seen first day performance numbers more closely tied to offering prices. According to Renaissance Capital, the average first day price pop during the first quarter of 2004 was 13%.

A major difference in Google’s recent public offering was its use of a Dutch auction process to price and size the offering. A Dutch auction uses investor bids to find the lowest price at which all available shares will be sold. Also called a “descending price auction”, the Dutch auction, according to WebFinance’s Investorwords.com, was named after the famous auctions of Dutch tulip bulbs in the 17th century. It is also used for some US Treasury auctions.

In theory, Google’s Dutch auction was intended to reduce first day price spread, put more of the proceeds of the offering in the coffers of the company and more of the shares sold into the hands of small investors. As Google explained in its registration statement with the SEC,

“As part of this auction process, we are attempting to assess the market demand for our Class A common stock and to set the size and price to the public of this offering to meet that demand. As a result, buyers should not expect to be able to sell their shares for a profit shortly after our Class A common stock begins trading. We will determine the method for allocating shares to bidders who submit successful bids following the closing of the auction.”

Google employed the Dutch auction mechanism to create a pre-IPO market mechanism to determine the IPO price point based on the fundamental supply and demand economics of the auction.

A second value that Google hoped to obtain from use of the Dutch auction was to democratize the allocation of shares, to make the IPO shares directly available to small individual investors who were prepared to bid for as few as five shares.

The Google IPO was one of the most widely anticipated and closely scrutinized in recent years. Curiosity about the auction process itself was enough to guarantee a tsunami of an offering. However, a number of well publicized negative news stories may have had the effect of slowing the winds some. For example, the same day that the company finally filed an amendment to its registration statement to state the expected range for the offering price ($108 – $135), the company’s website was hit by a virus which took it offline for a period of time. Around the same time, allegations surfaced that millions of shares issued by the company to employees and consultants prior to the offering may have violated state and federal securities laws. There was also an ill-timed Playboy interview by company insiders and a negative analyst’s report which, coming during the “quiet period” of the offering, could not be answered.

These factors, and others, led a number of observers to predict failure of one degree or another. In July, Kiplinger’s and Newsweek both warned that auctioning the shares would guarantee that the offering would garner a premium price and that the stock price would then promptly fall.

Google itself, in its SEC registration statement, included nine separate risk factors under the heading of “Risks Related to the Auction Process for Our Offering,” including the first which read “Our Stock Price Could Decline Rapidly and Significantly.”

!As it turned out, however, these negative predictions did not come to pass.

After the close of the auction period, and on the day before public trading began, on Wednesday, August 18, Google set the opening price at $85.00 per share and reduced by 5 million the total number of shares to be issued. As a result, Google and its early investors and executives sold a total of 19.6 million shares at the $85.00 price, yielding $1.2 billion dollars to Google and $464 million to Google insiders.

Then, and contrary to the predictions that the price would promptly fall, the market price of Google shares immediately rose 18% on the first day of trading, closing at $100.34.

Google’s was the third richest U.S.-based IPO on the NASDAQ after Charter Communications in 1990 and Genuity in 2000. Google’s first day market capitalization of approximately $23 billion made Google one of the most valuable companies listed on a U.S. market and placed it in the ranks of General Motors and Federal Express on the Fortune 500 list. Google’s first day 18% rise suggests that it might have fallen short of its objective of using the Dutch auction as a perfect pricing mechanism and that the company left some money on the table that was picked up by the first buyers reselling shares on the same day. However, by comparison with the Yahoo experience, Google’s pricing mechanism was successful in reducing such an extreme spread.

And how many IPO shares wound up in the hands of small investors at 5 and 10 shares a piece? It is hard to say, but it is clear that many of them did not hold on to their shares long. After Google and its insiders sold the 19.6 million shares which were finally subject to the offering, 13 million shares changed hands within the first hour of trading; a full 22.4 million had traded before the first day closed.

Are there lessons here for the rest of us? Google was in a unique position to structure a unique offering. It is a widely recognized brand with a readily accessible concept. Its market clout as a private company probably gave it IPO options that would not be available to many new issuers. Rising oil prices, war in Iraq, and other sources of economic uncertainty in a presidential election year have contributed to make 2004 a relatively soft year for new securities issuances. In that context, the financial success of the Google IPO may provide a glimmer of hope that demand is building for new technology stocks in 2005.

After one full month of trading (as of September 17, 2004) Google has not traded below $95 a share and has closed as high as $115.80.

Tom Hildreth is a Director at the McLane Law Firm and Chair of its Corporate Department. McLane Associate, Colleen Karpinsky also contributed to this article.