The More the Merrier? Number of Bidders, Information Dispersion, Renegotiation and Winner's Curse in Toll Road Concessions

Abstract

We empirically assess the winner's curse effect in auctions for toll road concessions. First, we investigate the overall winner's curse effects on bidding behaviour. Second, we account for differing levels of common-value components. Third, we investigate whether the possibility of renegotiation affects the winner's curse effect. Using a unique dataset of 49 concessions, we show that the winner's curse effect is particularly strong, i.e. bidders bid less aggressively when they expect more competition. In addition, we observe that this effect is larger for projects where the common uncertainty is greater, and is dampened in weaker institutional frameworks, in which renegotiations are easier.

Renegotiations in Public-Private Partnerships: Theory and Evidence

“…A standard interpretation is that this corresponds to cases of Winner's Curse. However, there is empirical evidence that lowballing bids is higher under weaker institutional frameworks, where renegotiations is easier, see Athias and Nuńez [2008] and Athias and Nuńez [2009]. This is consistent with our prediction of the association of opportunistic renegotiations and lowballing.…”

The Transaction Costs of Semi-Public Institutions: Railway History as a Challenge for Coase

“…20 In order to assure practical applicability, theory will be illustrated with actual jurisprudence whenever possible, instead of using simplified hypotheticals as economists frequently do. 21 Finally, economic models will be cast in terms of legal concepts as much as possible.…”

“…296 Cf. Athias & Nuñez (2008) for an interesting study of the concept in the context of tenders for toll roads.…”

Analysis of the distribution of the number of bidders in construction contract auctions

“…Similarly, with the exception of additional studies such as by Athias andNuñez (2009), Skitmore (2008) and Costantino et al (2011) there has been no further empirical work concerning the statistical nature of N and therefore previous assumptions will not be considered further here.…”