We use standard results from agency theory to assess the ex ante and ex post efficiency of the fixed price risk transfer contracts used to procure public infrastructure under the Private Finance Initiative (PFI) in the United Kingdom. We argue that these contracts act as a revelation mechanism designed to improve ex ante contracting efficiency at the cost of ex post inefficiency: PFI contract are a case of solving the moral hazard problem (creating maximum incentives for cost reduction) at the expense of increasing the adverse selection problem and therefore the selected firm’s information rent. Hence, Risk transfer leads to ex post inefficient outcomes that are directly related to the distribution of firm type: the more this distribution is skewed in favour of a few efficient firms, the less competition there is for the risk transfer contract, the larger the rent of the efficient firm should be. Thus, the ex post inefficiency of PFI contracts may be heavily country dependent. Using a detailed database of individual construction contracts for standard and PFI school projects in the UK, we find that full risk transfer does indeed lead to self-selection by the efficient firm, and that the distribution of firm types is indeed skewed in this market. We also find evidence of lower unit construction costs for the firm under PFI contracts compared to traditional procurement, which supports the hypothesis of the self-selection of the efficient firm when the public sector creates a menu of contracts (PFI vs. TP).