Abstract
This paper presents and empirically tests a theoretical framework to explain auction house behavior in setting pre-sale price estimates for sellers of assets. In strategic interaction between two auction houses and sellers that are either informed or uninformed about their asset values, we demonstrate that equilibrium pre-sale price estimates of auction houses will be guaranteed to show convergence on a single price only if auction houses are informed about the beliefs of the seller regarding the value of their asset. Further, we show that the extent of price divergence will be determined by whether the seller is informed or uninformed about their asset value, and in the case of uninformed sellers by the responsiveness of seller to pre-sale price estimates of auction houses. In a novel data set of 44,912 fine art auction lots over the 2009-2021 period, we find evidence of both pre-sale price estimate convergence and divergence. However, divergence dominates convergence. Relative to the market leader auction house defined in terms of largest market share, competitor auction houses appear to predominantly provide higher pre-sale valuations of auction lots than the market leader for the same auction lot. Results are robust to the sold/unsold status of auction lots, and the degree of market concentration.

This publication has 55 references indexed in Scilit: