Auctioning Football Seats

A couple of PMBA groups submitted this article:

Jets to Auction Seats on eBay

This article is a good answer to a question I often get: "How can I price optimally if I don't know what the demand curve is?"

The Jets get to sell these seat licenses once and only once, so it's not possible for them to do pricing experiments to determine the elasticity of demand. So an auction is a good alternative.

This is pretty similar to what Google did with their IPO a few years back. As with the selling of seat licenses, Google gets to sell these shares once and only once. In a typical IPO, a firm will set a price. If the set price is lower than the market clearing price, then the IPO is "oversubscribed." In this case, the underwriter (usually a big investment bank) will determine which of the potential buyers are allowed to buy.

Oversubscribed IPOs have two big problems. First, the firm is selling its shares for less than the market thinks the shares are worth. This means the firm is leaving money on the table. Second, oversubscribed IPOs have the potential for abuse, as investment banks would often allocate the underpriced shares to favored clients.

If the set price is too high, then the issue is undersubscribed, and all heck breaks loose.

Google circumvented all this using an auction. Google asked investors to submit something like a personal demand curve. Investors gave a list of how many shares they were willing to buy at each price.

Example: I'm willing to buy 100 shares if the price is $50 each. If the price is $40 each, then I'll buy 120. If the price is $30, I'll buy 140.

Then Google took the demand curves and said "We have X shares to sell... What's the price at which quantity demanded (given these demand curves) is equal to quantity supplied?" That was then the price that each buyer paid.