We study a procurement auction recently analysed by Gal-Or et al. (2007). In this auction game the buyer ranks different bids on the basis of both the prices submitted and the quality of each bidder that is her private information. We emphasise the similarity between this model and existing models of competition in horizontally differentiated markets. Finally we illustrate conditions for the existence and the stability of such equilibrium. To this end we extend the model to a dynamic setting in which a sequence of independent auctions takes place. We assume bidders have bounded rationality in a twofold sense. On one hand, they use an underparametrized model of their competitors’ behaviour, best responding to expectations on average bids rather than keeping track of the entire vector of competitors’ bids. On the other they update expectations adaptively. In a general framework with more than two bidders the system may fail to converge to the steady state, i.e. to the symmetric Nash equilibrium of the original game.
Dynamics in non-binding procurement auctions with boundedly rational bidders / D. Colucci; N. Doni; V. Valori. - STAMPA. - (2010), pp. 181-193. [10.1007/978-3-642-04023-8]
Dynamics in non-binding procurement auctions with boundedly rational bidders
COLUCCI, DOMENICO;DONI, NICOLA;VALORI, VINCENZO
2010
Abstract
We study a procurement auction recently analysed by Gal-Or et al. (2007). In this auction game the buyer ranks different bids on the basis of both the prices submitted and the quality of each bidder that is her private information. We emphasise the similarity between this model and existing models of competition in horizontally differentiated markets. Finally we illustrate conditions for the existence and the stability of such equilibrium. To this end we extend the model to a dynamic setting in which a sequence of independent auctions takes place. We assume bidders have bounded rationality in a twofold sense. On one hand, they use an underparametrized model of their competitors’ behaviour, best responding to expectations on average bids rather than keeping track of the entire vector of competitors’ bids. On the other they update expectations adaptively. In a general framework with more than two bidders the system may fail to converge to the steady state, i.e. to the symmetric Nash equilibrium of the original game.I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.