BEGIN:VCALENDAR
VERSION:2.0
X-WR-CALNAME;VALUE=TEXT:Tony Kwasnica: "Why Sellers Should Prefer Sequential Mechanisms"
PRODID:-//Harvard events data//EN
BEGIN:VEVENT
UID:event_71481_0
SUMMARY:Tony Kwasnica: "Why Sellers Should Prefer Sequential Mechanisms"
DESCRIPTION:<p><strong>PLEASE NOTE: DATE, TIME AND LOCATION</strong></p><p><strong>TUESDAY, April 24, 2012</strong></p><p><strong>CRCS Lunch Seminar</strong></p><p><strong>Date: TUESDAY, April 24, 2012<br>Time: 11:00AM – 12:30pm<br>Place: Pierce Hall 100F</strong></p><p>Speaker:  Tony Kwasnica, Penn State</p><p>Title:  Why Sellers Should Prefer Sequential Mechanisms</p><p><drupal-media data-entity-type="media" data-entity-uuid="27fddab6-3657-487d-a9b1-91deb7ad03b8"></drupal-media></p><p>Abstract:</p><p>We analyze two mechanisms commonly used for selling an asset or a contract, in a setting in which bidders must incur an entry cost in order to learn how much the asset is worth to them: an English auction, and a sequential bidding process.  A theory developed by Bulow and Klemperer (2009) predicts that sellers should prefer the auction, because it generates higher average revenues, while bidders should prefer the sequential mechanism, because it generates higher average bidder profits. We compare the two mechanisms in a controlled laboratory environment, varying the entry cost, and find that, contrary to the theoretical predictions, average seller revenues tend to be the same or higher under the sequential mechanism, while average bidder profits are approximately the same.  We identify three behavioral causes that explain our result: (1) bidders do not enter the auction 100% of the time, and (2) in the sequential mechanism, bidders do not set pre-emptive bids according to the threshold strategy, and (3) the subsequent bidders tend to over-enter in response to pre-emptive bids by the first bidders.  We develop a model of noisy bidder entry decisions, related to quantal response equilibrium, and show, using maximum-likelihood parameter estimation techniques, that our model does a good job in organizing the experimental data, both, in terms of point predictions, as well as in terms of qualitative comparisons of the two mechanisms.</p><p>Joint work with Andrew Davis (Cornell), Elena Katok (PSU).</p>
LOCATION:Pierce Hall 100F
STATUS:CONFIRMED
DTSTART:20120424T150000Z
DTEND:20120424T163000Z
END:VEVENT
END:VCALENDAR