This paper examines optimally biased Tullock contests. We consider a multi-player Tullock contest in which players differ in their prize valuations. The designer is allowed to impose identity-dependent treatments - i.e., multiplicative biases - to vary their relative competitiveness. The literature has been limited, because a closed-form solution to the equilibrium is in general unavailable when the number of contestants exceeds two, which nullifies the usual implicit programming approach. We develop an algorithmic technique adapted from the general approach of Fu and Wu (2020) and obtain a closed-form solution to the optimum that addresses a broad array of design objectives. We further analyze a resource allocation problem in a research tournament and adapt Fu and Wu's (2020) approach to this noncanonical setting. Our analysis paves the way for future research in this vein. (C) 2020 Elsevier B.V. All rights reserved.
NSTL
Elsevier
