http://www.beyonddiscovery.org/content/view.article.asp?a=3681
The Federal Communications Commission (FCC) auctioned off a slice of electromagnetic spectrum for the use of wireless communication devices in July of 1994. What was estimated to be worth about $10 billion, turned out to garner $42 billion by early 2001 and there are still more licenses to be sold. What was successful about this auction was in its design and in the rules of the auctions. The FCC split the spectrum into thousands of licenses to be auctioned off piece by piece. There were many complexities, such as choosing the type of auction: open or sealed bid, the planners were careful to also avoid loopholes that the firms might exploit, such as conspiring to keep the prices low. To solve these issues, the FCC hired mathematical experts of game theory to figure out the best strategies in this competitive arrangement, predict how bidders would behave in different types of auctions.

Since October of 1993, the FCC has invited telecommunication firms to submit proposals on the structure of the auction, gathering preliminary reports and important papers of auction theory. Since there were over 2500 licenses to disperse, and many companies needed to buy more than one license at a time, the FCC decided to auction off the licenses altogether instead of one at a time, to prevent bidders from fearing the winner’s curse of paying too much; this decision made running the auction more complicated than selling each spectrum license one at a time. The FCC also decided to use an English auction, since it would raise the most revenue, but with a twist: the bids in each round of the auction would be placed secretly in enclosed booths, and the FCC would then announce the new high price without saying the bidder’s name. Concealing the bidder’s identity this way lessened the possibility of retaliatory bidding or conspiracies to keep prices down.

The outcome of this careful planning resulted in a great success that yielded in not only more money than anticipated, but it also sold licenses to the companies that would use the spectrum efficiently: within two years of the first auctions, wireless phones were seen in the market.

Careful planning and design based on game theory are often needed for high-price auctions such as this. Many auctions have tried to follow the FCC for similar success. While many were successful and yielded high prices for licenses, some have been exploited by bidders who’ve found loopholes in designs, such as the telecom licenses Turkey auctioned off in 2000 that resulted in a company monopoly, as well as the 10 blocks of spectrum Germany tried to sell in 1999, resulting in a collusion between two powerhouse bidders, Mannesman and T-Mobil. Auction theorists work hard to prevent such abuses from happening, mostly just through the simple idea of putting themselves into the minds of their opponents, the bidders.

This article discusses the designs of various telecommunication spectrum license auctions, in particular those by the FCC. It also talks about the basic idea and general history behind game theory, discussing many things we have learned in class, such as Nash equilibrium, zero-sum games, types of auctions, and the Winner’s Curse of paying too much. It’s a great article to see the application of game theory in real-life: not only applied to yield a specific result, but one that involves such large sums of money. Also, it emphasizes the basic concepts that we’ve learned in class, a great review for those that would like one.