Loading...
Please wait, while we are loading the content...
Similar Documents
Structural Estimation of First-Price Auction Models: Measuring Common Values and the Winner's Curse
| Content Provider | Semantic Scholar |
|---|---|
| Author | Shum, Matthew |
| Copyright Year | 1999 |
| Abstract | We estimate structural models of competitive bidding in symmetric low-price procurement auctions in order to address two empirical questions. First, we investigate the importance of common and private values using a general model which allows bidders' preferences to exhibit both components. Second, we see whether participants bid more aggressively (as measured by the markups of bids over expected costs) as the number of competitors in an auction increases. In common value auctions, the eeect of increased competition on equilibrium markups is ambiguous since the increased competitive pressure which tends to lower markups is counteracted by the winner's curse, which counsels greater caution and more conservative bidding. This ambiguity is not present in private value models. We nd evidence of common value components in a subset of the auctions we study. Furthermore, we estimate that bidders' markups averaged 20% in 3-bidder auctions, but rose to 50% with 5 bidders and over 70% with 10 bidders; increased competition leads to less aggressive bidding. In these auctions, then, it appears that increasing the number of competitors would not necessarily lower project costs. For other auctions in our dataset, however, this estimated markup trend is either at or even decreasing. Prepared for the 1999 Stanford Institute of Theoretical Economics: comments welcome. We thank Marty Miller at the New Jersey department of transportation for providing us with the data, and patiently answering our questions. We thank Lanier Benkard for a careful reading of an earlier draft, and Hai Che |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://www.stanford.edu/group/SITE/Shum.ps |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
