Randomization, Endogeneity and Laboratory Experiments: The Role of Cash Balances in Private Value Auctions

From a theoretical perspective, cash balances are thought to play a role in common value auctions because of limited liability. However, they have also been found to be important in common value auctions where limited liability is not an issue. This paper investigates whether this effect carries over to private value auctions, since limited liability issues do not arise in private value auctions. We address the issue that cash balances are likely to be endogenous. We show that additional randomization can be used to reduce endogeneity problems and to improve the performance of our estimators. Further, we find that standard panel data econometric techniques are very useful in the analysis of data from laboratory experiments. Finally, we find that the cash balance effect does indeed carry over from common value auctions to private value auctions. * We have benefitted from discussions with Guillaume Frechette, Hidehiko Ichimura, Whitney Newey, Jean-Francois Richard, the comments of Robert LaLonde, three anonymous referees, and seminar participants at the Carnegie-Mellon/Pittsburgh applied micro workshop, the 1999 Public Choice/ESA meetings, the 2000 World Congress of the Econometric Society, and the 2001 meetings of the Economic Science Association in Barcelona, Spain. Ham would like to thank the NSF for financial support through grants SRR-951-2001 and SBR-980-9S46. Part of this paper was written while Ham was a visitor in the Economics Department at the University of Pennsylvania and he thanks that department for its hospitality and support. Kagel would like to thank the Economics Division and the DRMS Divisions of NSF for financial support. Lehrer gratefully acknowledges research support from the University of Pittsburgh. We are responsible for all errors.