Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)

Legacy Department



Daniel Miller

Committee Member

Matthew Lewis

Committee Member

Tom Chungsang Lam

Committee Member

Andrew Hanssen


Highway construction projects serve as a good example where society can benefit from fast completion. A+B auctions, a type of innovative scoring auctions, address this concern by incentivizing timely completion through scores that combine price and time incentives. In the first part of the dissertation, I investigate in A+B auctions by building a theory of A+B bidding that incorporates incentives and production uncertainty, as well as structurally estimating bidding behaviors and auction performance using data from the California Department of Transportation (Caltran). I find that, in equilibrium, bidders skew the days bid below the true planned construction target days to raise the price bids. Moreover, self-selected construction time that is different from the expected social-optimal time causes ex post efficiency loss and the auction mechanism can fail at picking the socially-efficient bidders ex ante. Counterfactual analysis suggests that procuring schemes with lower incentives or even conversion back to traditional contracts are likely to yield better social outcomes. Usually, highway procurement auctions across the U.S. take the form of Unit-Price Contracts (UPCs), where the department of transportation (DOT) gives quantity estimates for different tasks of the project and the bidders attach unit-price bids on each of the tasks and form a single total bid. During the construction phase, the DOT makes progress payments, usually monthly, to the contractor for tasks completed in the previous month. Moreover, the payments are determined by the actual quantities incurred although the total bids ex ante are evaluated based on the DOT estimates. A forward-looking contractor would want to maximize the expected total present value by shifting more weight toward early tasks and tasks that the bidders believe to overrun compared to the DOT estimates. These two types of bid skewing lead to what is generally known in the industry as 'unbalanced bids', which I empirically test and quantify in the second part of the dissertation.