Dynamic Auction Model with Subcontracting

We study a dynamic procurement environment with capacity constraints where contractors may re-sell part of the work in the subcontracting market. The ability to subcontract is a prevalent feature of many procurement auction markets but its analysis is absent from the literature. We find that this omission has important consequences, since the ability to subcontract has a substantial and nontrivial impact on the market. In our model projects are allocated to contractors in the primary market through first price sealed bid auctions. Contractors’ costs vary stochastically across projects and independently across contractors. They are contractors’ private information. Capacity constraints impact contractors’ costs through carried-over work commitments (backlogs). We develop a novel numerical strategy for solving this dynamic game. We use this method to characterize the equilibrium and to isolate and quantitatively assess the magnitude of various mechanisms through which the presence of subcontracting shapes this market. We show that failing to account for subcontracting in estimation results in significant bias to the estimates of the distribution of private costs and capacity constraints. We also find that outcomes of procurement policy change drastically if subcontracting becomes available and discuss an optimal procurement policy for this environment.