Essays in three design issues in experimental auctions

Abstract

The objective of this thesis is to investigate three design issues in experimental auctions: 1) the effects of allowing negative bids for a privately valued good, 2) the effects of introducing additional alternatives (substitutes) for the auctioned good in an endowment auction, and 3) respondent behavior in acquiring information. The thesis consists of three papers examining those issues. The first paper examines participants’ bidding behavior when negative bids are allowed for privately valued goods in an experimental auction. We focus on two questions: i) whether subjects with negative values tend to bid strategically – either overbidding or underbidding in an effort to enhancing earnings, and ii) the performance of random nth and 5th price auctions. We find that: a) WTP bids are demand revealing, b) subjects tend to underbid WTA values, c) controlling for risk attitude partially explains the bias in WTA bids, and d) negative values from random nth auctions tend to be below those from 5th price auctions. In the second paper we 1) investigate the effect of the availability of varying numbers of alternatives (substitutes) for a privately valued good on participants’ bidding behavior, and 2) identify whether the availability of additional alternatives: a) impacts the value of product information, and b) impacts the effect of new information on product valuations. We find that: a) allowing additional alternatives in a private value auction does not significantly decrease subjects’ bids, and b) the presence of additional alternatives in the auction decreases both the value and effect of product information. The third paper examines the effect of acquired information on auction participants’ bidding behavior. We focus on three questions: i) how subjects choose/value different types of information, ii) whether the value of acquired information about a product influences the subsequent valuation of the product itself, and iii) whether the effects of acquired information differ from those of exogenously provided information. We find that: a) subjects’ behaviors of acquiring different types of information about the product are influenced by their heterogeneous characteristics (i.e. prior beliefs, risk attitudes, prior knowledge, etc.), b) subjects place more weight on acquired information than on provided information in their decision-making process, and c) individual subjects have different values of information which caused different impacts on product valuation.