What is a Dutch Auction?
A Dutch auction is a price discovery process in which the auctioneer starts with the highest asking price and lowers it until it reaches a price level where the bids received will cover the entire offer quantity. Alternatively, a Dutch auction is known as a descending price auction or a uniform price auction. Dutch auctions are appropriate for instances where a large quantity of an item is being offered for sale, as opposed to just a single item.
Dutch Auction Process
In the Dutch auction process for an IPO, the underwriter does not set a fixed price for the shares to be sold. The company decides on the number of shares they would like to sell and the price is determined by the bidders. Buyers submit a bid with the number of shares they would like to purchase at a specified bid price. A list is created, with the highest bid at the top. The company works down the list of bidders until the total desired number of shares is sold.
The price of the offering is determined from the last price covering the full offer quantity. All bidders pay the same price per share. A Dutch auction encourages aggressive bidding because the nature of the auction process means the bidder is protected from bidding a price that is too high.
Example of a Dutch Auction
Assume that the company, Compu Inc., is using a Dutch auction to price its shares for an IPO. The company is looking to sell a total of 400 shares in its IPO.
- Investor A places a bid for 200 shares at $300
- Investor B places a bid for 25 shares at $450
- Investor C places a bid for 500 shares at $100
- Investor D places a bid for 60 shares at $200
- Investor E places a bid for 100 shares at $150
- Investor F places a bid for 15 shares at $120
- Investor B: 25 shares at $450 (400 – 25 = 375 shares remaining)
- Investor A: 200 shares at $300 (375 – 200 = 175 shares remaining)
- Investor D: 60 shares at $200 (175 – 60 = 115 shares remaining)
- Investor E: 100 shares at $150 (115 – 100 = 15 shares remaining)
- Investor F: 15 shares at $120 (15 – 15 = 0 shares remaining)
- Investor C: 500 shares at $100
In this example, the IPO would be priced at $120 per share because the last bid of 15 shares at $120 filled out the total number of shares that Compu Inc. is looking to offer. Investors B, A, D, and E would be able to purchase shares for $120 instead of their initial bids of $450, $300, $200, and $150, respectively. Investor C would be out because the number of shares is already filled.
Consider the same example but with different bid prices:
- Investor A places a bid for 100 shares at $300
- Investor B places a bid for 200 shares at $200
- Investor C places a bid for 200 shares at $200
- Investor A: 100 shares at $300 (400 – 100 = 300 shares remaining)
- Investor B and C: 200 shares at $200 (300 – 400 shares total = -100 shares remaining)
Note that Investors B and C both placed a bid for 200 shares at $200 (400 shares in total). Therefore, at a price of $200, there is a demand for 500 shares (100+200+200). However, the company only wants to sell 400 shares. In this case, the company must figure out a way to allocate these shares. One way to resolve this problem is to take the percentage:
400 shares available / 500 shares demanded = 80%. Investors A, B and C would receive 80% of their requested shares:
- Investor A receives 80 shares at $200
- Investor B receives 160 shares at $200
- Investor C receives 160 shares at $200
Total shares issued: 400
Traditional IPOs and Underpricing
Setting a price for an IPO can be difficult. Traditionally, investment bankers (underwriters) would take the top management of the company on “roadshows” to meet with institutional investors and assess their interest in the IPO. These roadshows offered the underwriter an opportunity to market the stock in advance, thereby hopefully increasing demand for it, and also the opportunity to learn how much per share large, institutional investors were initially willing to pay for the stock.
However, setting a price for an IPO through roadshows is sometimes unreliable. For example, Twitter was priced at $26 for its IPO, based on gauging public interest at roadshows, but traded as high as $45 on the first day of trading. This is called mispricing, where the IPO is priced too low. The feedback the underwriter received from the roadshows was obviously misleading in terms of public interest in acquiring the stock.
A Dutch auction is used to minimize the increase between the offer price and the opening price of the offering (therefore minimizing underpricing). While it usually results in some bidders paying less for the stock than they were willing to, it at least protects the underwriter and the company from having to sell hundreds, or perhaps thousands, of shares at a ridiculously low price.
Google’s IPO: A Dutch Auction
In 2004, Google (NASDAQ: GOOG), now Alphabet Inc., decided to go with a Dutch auction IPO process. In its regulatory filings, Google’s documents stated: “Many companies going public have suffered from unreasonable speculation, small initial share float, and stock price volatility that hurt them and their investors in the long run,” and “we believe that our auction-based IPO will minimize these problems, though there is no guarantee that it will.”
Google (Alphabet Inc.) relied on a Dutch auction to minimize underpricing and to earn a fair price on its IPO. Although Google went public at $85 a share and climbed nearly 30% in two days to close at $108 a share, the IPO was considered a success due to the initial uncertainty in the effectiveness of a Dutch auction. At that time, many market analysts criticized the Dutch auction process. Many of them were afraid that investors would collectively submit a low bid and cause Google to open at an unfavorably low price. In hindsight, there is speculation on whether Google would have been able to set a higher opening price if it had gone the more traditional IPO route.
Read more from Market Watch about the reaction to Google’s dutch auction IPO.