Combination Bidding in Multi-Unit Auctions

This paper considers the problem of identification and estimation in the first-price multi-unit auction. It is motivated by the auctions of bus routes held in London where bidders submit bids on combinations of routes as well as on individual routes. We show that submitting a combination bid lower than the sum of the bids on the constituent routes does not require cost synergies and can instead serve as a tool to leverage market power across the different routes. As a result, the welfare consequences of allowing combination bidding in the first price auction are ambiguous, and depend on the importance of the cost synergies. We provide conditions for non-parametric identification of the multidimensional private information in the multi-unit first price auction and derive partial identification results when they are not satisfied. We propose an estimation method consisting of two stages: In the first stage, the distribution of bids is estimated parametrically. In the second stage, the (set of) costs and distribution(s) of costs consistent with the observed behavior are inferred based on the first order conditions for optimally chosen bids. We apply the estimation method to data from the London bus routes market. We quantify the magnitude of cost synergies and assess possible efficiency losses arising in this market.

[1]  William Vickrey,et al.  Counterspeculation, Auctions, And Competitive Sealed Tenders , 1961 .

[2]  Simeon M. Berman,et al.  Note on Extreme Values, Competing Risks and Semi-Markov Processes , 1963 .

[3]  E. H. Clarke Multipart pricing of public goods , 1971 .

[4]  Theodore Groves,et al.  Incentives in Teams , 1973 .

[5]  Betty Jones Whitten,et al.  Estimation in the Three-Parameter Lognormal Distribution , 1980 .

[6]  Interval Estimation for the Three‐Parameter Lognormal Distribution Via the Likelihood Function , 1980 .

[7]  Roger B. Myerson,et al.  Optimal Auction Design , 1981, Math. Oper. Res..

[8]  L. Hansen Large Sample Properties of Generalized Method of Moments Estimators , 1982 .

[9]  Thomas R. Palfrey,et al.  Bundling Decisions by a Multiproduct Monopolist with Incomplete Information , 1983 .

[10]  Richard Schmalensee,et al.  Gaussian Demand and Commodity Bundling , 1984 .

[11]  Richard L. Smith Maximum likelihood estimation in a class of nonregular cases , 1985 .

[12]  C. D. Kemp,et al.  Density Estimation for Statistics and Data Analysis , 1987 .

[13]  P. J. Green,et al.  Density Estimation for Statistics and Data Analysis , 1987 .

[14]  M. Whinston Tying, Foreclosure, and Exclusion , 1989 .

[15]  M. Whinston,et al.  Multiproduct Monopoly, Commodity Bundling, and Correlation of Values , 1989 .

[16]  D. Pollard,et al.  Simulation and the Asymptotics of Optimization Estimators , 1989 .

[17]  D. McFadden A Method of Simulated Moments for Estimation of Discrete Response Models Without Numerical Integration , 1989 .

[18]  Stephen Glaister,et al.  Bidding for tendered bus routes in London , 1991 .

[19]  Harry J. Paarsch,et al.  Piecewise Pseudo-maximum Likelihood Estimation in Empirical Models of Auctions , 1993 .

[20]  Quang Vuong,et al.  First-Price Sealed-Bid Auctions with Secret Reservation Prices , 1994 .

[21]  R. Rosenthal,et al.  Simultaneous Auctions with Synergies , 1996 .

[22]  J. Laffont,et al.  ECONOMETRICS OF FIRST-PRICE AUCTIONS , 1995 .

[23]  Jean-Jacques Laffont,et al.  Structural Analysis of Auction Data , 1996 .

[24]  Lawrence M. Ausubel,et al.  Synergies in Wireless Telephony: Evidence from the Broadband PCS Auctions , 1997 .

[25]  Jonathan D. Levin An Optimal Auction for Complements , 1997 .

[26]  Bernard Lebrun Comparative Statics in First Price Auctions , 1998 .

[27]  K. Judd Numerical methods in economics , 1998 .

[28]  Jean-Charles Rochet,et al.  Multi-dimensional screening:: A user's guide , 1999 .

[29]  C. Avery,et al.  Bundling and Optimal Auctions of Multiple Products , 2000 .

[30]  E. Maskin,et al.  Equilibrium in Sealed High Bid Auctions , 2000 .

[31]  Q. Vuong,et al.  Optimal Nonparametric Estimation of First-price Auctions , 2000 .

[32]  Susan Athey,et al.  Identification of Standard Auction Models , 2000 .

[33]  M. Armstrong Optimal Multi-Object Auctions , 2000 .

[34]  Paul R. Milgrom,et al.  Putting Auction Theory to Work: The Simultaneous Ascending Auction , 1999, Journal of Political Economy.

[35]  Philip A. Haile,et al.  Inference with an Incomplete Model of English Auctions , 2000, Journal of Political Economy.

[36]  Martin Pesendorfer,et al.  Estimation of a Dynamic Auction Game , 2001 .

[37]  F. Wolak Identification and Estimation of Cost Functions Using Observed Bid Data: An Application to Electricity Markets , 2001 .

[38]  Michael Visser,et al.  Econometrics of Share Auctions , 2002 .

[39]  C. Manski,et al.  Inference on Regressions with Interval Data on a Regressor or Outcome , 2002 .

[40]  Jeroen M. Swinkels,et al.  Communication and equilibrium in discontinuous games of incomplete information , 2002 .

[41]  Q. Vuong,et al.  Structural Estimation of the Affliated Private Value Auction Model , 2002 .

[42]  Quang Vuong,et al.  Asymmetry in first-price auctions with affiliated private values , 2003 .

[43]  Susan Athey,et al.  Comparing Open and Sealed Bid Auctions: Theory and Evidence from Timber Auctions , 2004 .

[44]  Yoav Shoham,et al.  Combinatorial Auctions , 2005, Encyclopedia of Wireless Networks.

[45]  Harry J. Paarsch,et al.  An empirical model of the multi-unit, sequential, clock auction , 2006 .