For instructors
Teaching notes
Governments sell spectrum and mining leases by auction, firms win contracts through sealed tenders, and advertising space is sold in millions of small auctions every hour. The rules differ from case to case, and a bidder's best strategy depends on the rules.
What the game shows
The game tests how people bid under different rules and whether formats that the theory treats as equivalent produce the same prices. With private values it shows bid shading, the dominant strategy of the second-price auction and the revenue equivalence of the standard formats. With a common value it shows the winner's curse. The all-pay and dollar auctions show escalation, in which bidders together pay more than the item is worth.
The theory
Vickrey (1961) showed that in a second-price sealed-bid auction it is a dominant strategy to bid one's value, and that the English auction leads to the same outcome, since a bidder stays in until the price reaches that value. In a first-price auction bidders shade their bids below their values, and with values drawn uniformly a bidder who cares about expected earnings bids the lowest possible value plus the fraction of the distance to the own value given by the number of bidders minus one over the number of bidders. The Dutch auction presents the same decision as the first-price auction.
The revenue equivalence theorem, due to Myerson (1981) and to Riley and Samuelson (1981), states that all these formats, and the all-pay auction too, yield the same expected revenue when values are private and independent and bidders are neutral toward risk. When the value is common and each bidder holds a noisy signal, the highest signal tends to win, so a bidder must reduce the bid to allow for what winning implies. Milgrom and Weber (1982) give the general theory, and Capen, Clapp and Campbell (1971) described the problem from experience of bidding for oil leases. The dollar auction of Shubik (1971), in which the runner-up also pays, turns each further bid into a comparison with a loss already incurred.
What to expect
Bids in first-price auctions are generally above the benchmark for bidders who are neutral toward risk, a pattern that Cox, Smith and Walker (1988) relate to risk aversion. The equivalences between formats fail in a consistent direction, beginning with the finding of Coppinger, Smith and Titus (1980) that prices in Dutch auctions were about 5 percent below those in first-price auctions. In second-price auctions, Kagel and Levin (1993) found 62 percent of bids above the bidder's value and 30 percent at it, whereas Kagel, Harstad and Levin (1987) observed that prices in English auctions moved quickly to the level that the dominant strategy implies.
The winner's curse appears reliably when the value is common, and Bazerman and Samuelson (1983) document it with items worth 8 dollars sold in 48 classroom auctions. The average estimate of the value was $5.13 and the average winning bid $10.01, so that the average winner lost about two dollars. Kagel and Levin (1986) found that experienced bidders earned profits in groups of three or four and made losses in groups of six or seven. In the all-pay auctions of Gneezy and Smorodinsky (2006), the seller's revenue was commonly between two and three times the value of the prize.
Tips and things to look out for
- Revenue equivalence can be tested within one session by choosing several formats, for instance the first-price and the second-price auction. The auctions of the class then follow the two formats in rotation, and the host page reports price, revenue and efficiency for each format and for each auction, with bids plotted against the benchmark of each format. Running the formats one after the other with the whole class gives every student experience of both, although the comparison then confounds the format with experience.
- In the second-price auction, look for bids above value. Ask a student who bid above value to describe a case in which the higher bid changes the outcome, which is always a case in which the bidder loses money.
- For the winner's curse, use a common value with six or more bidders in each auction. The share of winners who lost money appears on the host page, and the comparison of signals with bids shows who failed to adjust.
- The dollar auction works with a value known to all and the whole class in one auction. Two bidders are usually left to escalate, and since points can become negative it is worth agreeing beforehand that the session is an exercise.
- In the Dutch auction the price is set by the server's clock at the moment the request arrives. Steps of three seconds or more keep the effect of a slow connection small.
- The setting for what bidders see after a round affects learning. Showing all bids lets losers see how far they were from winning, which can raise bids in later rounds of a first-price auction.
- Collect a prediction of the winning bid before each format is played. The gap between predictions for the first-price and the second-price auction shows whether the class anticipates shading.
References
- Bazerman, M. H., & Samuelson, W. F. (1983). I won the auction but don't want the prize. Journal of Conflict Resolution, 27(4), 618–634.
- Capen, E. C., Clapp, R. V., & Campbell, W. M. (1971). Competitive bidding in high-risk situations. Journal of Petroleum Technology, 23(6), 641–653.
- Coppinger, V. M., Smith, V. L., & Titus, J. A. (1980). Incentives and behavior in English, Dutch and sealed-bid auctions. Economic Inquiry, 18(1), 1–22.
- Cox, J. C., Smith, V. L., & Walker, J. M. (1988). Theory and individual behavior of first-price auctions. Journal of Risk and Uncertainty, 1(1), 61–99.
- Gneezy, U., & Smorodinsky, R. (2006). All-pay auctions: An experimental study. Journal of Economic Behavior & Organization, 61(2), 255–275.
- Kagel, J. H., & Levin, D. (1986). The winner's curse and public information in common value auctions. American Economic Review, 76(5), 894–920.
- Kagel, J. H., & Levin, D. (1993). Independent private value auctions: Bidder behaviour in first-, second- and third-price auctions with varying numbers of bidders. Economic Journal, 103(419), 868–879.
- Kagel, J. H., Harstad, R. M., & Levin, D. (1987). Information impact and allocation rules in auctions with affiliated private values: A laboratory study. Econometrica, 55(6), 1275–1304.
- Milgrom, P. R., & Weber, R. J. (1982). A theory of auctions and competitive bidding. Econometrica, 50(5), 1089–1122.
- Myerson, R. B. (1981). Optimal auction design. Mathematics of Operations Research, 6(1), 58–73.
- Riley, J. G., & Samuelson, W. F. (1981). Optimal auctions. American Economic Review, 71(3), 381–392.
- Shubik, M. (1971). The dollar auction game: A paradox in noncooperative behavior and escalation. Journal of Conflict Resolution, 15(1), 109–111.
- Vickrey, W. (1961). Counterspeculation, auctions, and competitive sealed tenders. Journal of Finance, 16(1), 8–37.