Qiu, Xiao Yun 

Abstract:  We study optimal allocation mechanisms that use costly signals as screening devices. A social planner seeks to minimize total signaling costs, subject to the implementation of a given monotone allocation rule. In these settings, signal recommendations based on all agents' reports (allowing information leakage) can improve coordination among losers and reduce their signaling costs. However, this comes at the expense of inducing excessive effort from potential winners. We show that when higher types have higher certainty-equivalent signals, the incentive cost of coordination outweighs its benefit. In this case, non-coordination, recommending signals without conditioning on others' reports, is optimal. Finally, we show that such non-coordination mechanisms can be implemented through coarse-ranking contests.

Abstract: We study how the design of testing institutions, encompassing both the tests themselves and the procedures used to administer them, shapes selection outcomes in environments with multiple criteria and strategic agents. We model the testing agency as either a set of independent bureaucracies (each test administered separately) or a joint bureaucracy (where test order and personalization can be coordinated). Our mechanism design analysis shows that under a joint bureaucracy, fixed-order sequential mechanisms with stringent tests are optimal for maximizing the probability mass of qualified candidates selected. Furthermore, we demonstrate that personalizing tests through upfront communication, now increasingly feasible via AI and automation, can select all qualified candidates. Finally, we compare institutional settings and quantify the value of controlling test order, showing that the benefit depends critically on the distribution of testees and the stringency of optimal tests. Our results contribute to the design of robust, efficient, and fair testing systems in both human and AI-mediated environments.

Abstract:   How much competition should a designer introduce when procuring innovation? A designer seeks solutions to an unsolved problem by funding multiple technological approaches whose feasibility is initially uncertain and learned through development. She can contract with each firm independently or organize a technology-agnostic contest. Between these two extremes, we introduce interdependent contracts, which offer approach-specific rewards while allowing the continuation of one approach to depend on outcomes realized by others. The optimal degree of competition depends critically on which technological outcomes are contractable. When both success and breakdown are contractable, interdependent contracts can extract firms' rents while allowing the designer to dynamically terminate unresolved approaches as information arrives. The mechanism-design problem then reduces to an optimal stopping problem that trades off the value of additional innovations against the cost of continued experimentation. When only success is contractable, competition becomes more valuable. For symmetric technologies, we characterize optimal interdependent contracts and show that rank-contingent contests weakly dominate them: a contest can implement the same experimentation and stopping policy while targeting rewards to realized success ranks, thereby weakly reducing expected expenditure. The advantage is strict when incentive requirements differ across attainable ranks, but vanishes as the number of firms grows. With asymmetric technologies, the ranking can reverse because interdependent contracts can tailor incentives to individual technological approaches. Thus, neither maximal nor minimal competition is universally optimal: the desirable degree of competition depends on the contractibility of technological outcomes, market size, and heterogeneity across technologies.

Abstract: Performance manipulation arises when agents exploit easily measurable, routine tasks to inflate observable outcomes without contributing genuine innovation or expert judgment. We formalize this phenomenon in a screening model in which agents allocate effort along two margins. Creative effort is non-routine cognitive labor whose return is complementary to the agent's private expertise; it is the scarce input that principals seek. Mechanistic effort is the execution of well-defined, rule-based tasks that raise performance independently of expertise, a commoditized input that AI heavily augments. We establish the existence of a symmetric, monotone pure-strategy equilibrium and show that performance-based screening remains viable so long as evaluations retain a sufficient creative component, but collapses into an uninformative pooling equilibrium once AI capability grows large enough to crowd out creative effort. Comparing contest allocations against a single-agent baseline isolates performance manipulation as the competition-induced over-investment in mechanistic effort, which we show is undertaken systematically by low-type agents but not high-type ones. We further prove that more sharply skewed reward structures mitigate this friction by eliciting greater creative effort across the participant pool. Finally, using a novel, language-model-based methodology to measure both effort types from nearly 1500 Kaggle competition scripts, we provide robust empirical support for the model's predictions.

Abstract:  We study how to allocate resources to participants who can strategically misrepresent their deservingness at a cost. A principal assigns item(s) (or money) among multiple agents on the basis of their costly signals. Each agent's signal reflects his private type, absence of misrepresentation, but can be inflated above his true type at a cost. The principal is a social planner who aims to maximize the weighted average of matching efficiency (allocating resources to the most needed) and a utilitarian objective (a measure of participants' welfare ). Strategic misrepresentation introduces novel incentive-compatibility constraints, under which we characterize the optimal mechanism. We apply our characterization to two kinds of public programs, distinguished by resource scarcity, and show that the principal strictly benefits from randomizing the allocations based on costly signals when the population of participants is large enough. Interestingly, in public programs with scarce resources, the format of the optimal mechanism converges to a winner-takes-all contest; however, there is a non-diminishing value of randomizing allocations to middle types as the population of participants grows.

Abstract: This paper studies the algorithmic complexity for computing the pure Nash Equilibrium (PNE) in Tullock Contests. The (possibly heterogeneous) elasticity parameter ri determines whether a contestant i’s cost function is convex, concave or neither. Our core finding is that the domains of ri governs the complexity for solving Tullock contents. 

• When no contestant’s ri lies between (1,2], we can design an efficient algorithm to compute the pure NE; 

• when many ri values fall within (1,2], we prove that determining NE existence can not be solved in polynomial time, assuming the Exponential Time Hypothesis (ETH); 

• when many ri values fall within (1,2], we design a Fully Polynomial-Time Approximation Scheme (FPTAS) to find an ϵ-PNE when an exact PNE exists. 

All our algorithms are efficiently implemented for solving large-scale instances, and computational experiments demonstrate their effectiveness even in complex scenarios.