Working Paper: NBER ID: w17647
Authors: Gregory Lewis; Patrick Bajari
Abstract: Deadlines and penalties are widely used to incentivize effort. We model how these incentive contracts affect the work rate and time taken in a procurement setting, characterizing the efficient contract design. Using new micro-level data on Minnesota highway construction contracts that includes day-by-day information on work plans, hours actually worked and delays, we find evidence of moral hazard. As an application, we build an econometric model that endogenizes the work rate, and simulate how different incentive structures affect outcomes and the variance of contractor payments. Accounting for the traffic delays caused by construction, switching to a more efficient design would substantially increase welfare without substantially increasing the risk borne by contractors.
Keywords: Moral Hazard; Incentive Contracts; Procurement; Highway Construction
JEL Codes: D86; H57; L92
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Time penalties (J33) | Contractor work rates (L74) |
| Time penalties (J33) | Completion rates (I21) |
| Larger time penalties (C41) | Less likely to be late (C41) |
| Current contract design is suboptimal (D86) | Shift to linear contract structure yields welfare gains (D69) |
| Current contract design is suboptimal (D86) | Increased risk borne by contractors (G32) |
| Observed adaptation in contractor work rates (J33) | Aligns with theoretical predictions of model (C52) |
| High-powered incentives (M52) | Relatively small risk for contractors (L74) |