Skip to main content

Auction Theory Behind Google’s IPO in 2004

Exactly 12 years and one month ago from today, Google held its IPO (initial public offering) that attracted numerous investors from diverse financial backgrounds all over the United States. Many auction theorists, from renowned professors in Ivy League universities to young Economics-major students looking for a case study, were closely observing the event that was about to become a phenomenon in the financial market. Not only countless people stocked in to participate in IPO but also the auction process that Google chose was unique for that time: it was a Dutch auction–one of the auctions covered in class–with little modifications. First, all the investors participating in IPO fill out a form, indicating amount of stocks they wish to purchase and the price they are willing to pay for the said stock. Then the auctioneer arranges the numbers from highest to lowest bid. Beginning from the highest bid, the auctioneer looks for a market-clearing price for all the stocks out for the auction using computer software. When the market-clearing price is found, everyone who bid equal to or higher than that price will be purchasing the stocks with market-clearing price.

This odd attempt–Dutch auction was not the typical way of handling IPO for most companies at that time–merged with a fact that many individual-level, non-professional investors are also participating, aroused curiosity from many auction theorists. Many theorists interviewed by The Wall Street Journal about the issue, in fact, regarded this novel attempt by Google as “intriguing”. Some show their nervousness over the “kinks” of the auction; John Miller, an Economics professor at Carnegie Mellon University, even calls the Google IPO as a “3 billion dollar experiment”. However, despite such underlying uncertainties in the Google IPO, it’s still interesting to note that most of the auction theorists decided to stick with the classical dominant strategy of second-price bid auction: “bidding the true value”, which is the conclusion we came up in a class as a best strategy for most of the auctions.

As a class, we discussed about various kinds of auction: English auction, Dutch auction, first-price bid auction, etc. However, most of the auctions were taught in generalised, simplified versions with theoretical terms. Google IPO is, although it is more than a decade old, a good, solid example of auction theory applied to real-world Economics problem. As cited above, despite several uncertainty factors that confused the auction theorists, it is important to note that all of the scholars interviewed chose to stick with “bidding the true value” strategy–the same conclusion we reached using the theoretical auctions.

You can read the whole article from The Wall Street Journal here:



Leave a Reply

Blogging Calendar

September 2016