Near-Optimal Mechanisms for Resource Allocation Without Monetary Transfers

Aligning Incentives Without Opening the Wallet

Many organizations must repeatedly allocate scarce resources—computing capacity, scientific equipment, school seats, or food-bank supplies—when monetary payments are impractical or undesirable. Blanchard and Jaillet show how future allocations can serve as rewards and penalties, inducing strategic agents to report their values truthfully while keeping welfare close to the full-information ideal. Their mechanisms use the promised utility framework and apply to general utility distributions in both finite-horizon and discounted infinite-horizon settings. The key insight is geometric; performance depends on the shape of the attainable-utility frontier. When agents often have similar values for the resource, the planner can shift allocations at little welfare cost and align incentives more effectively. The paper establishes universal convergence guarantees, faster rates for smooth utility distributions, and even exponentially small welfare losses in some discounted settings with frequent ties. The results provide a unified, quantitative framework for deciding when nonmonetary allocation can nearly match mechanisms that use payments.