|
on Economic Design |
| By: | Sushil Bikhchandani; Debasis Mishra |
| Abstract: | We study the design of strategy-proof and efficient mechanisms satisfying participation constraints in the job-matching problem. Each firm can hire multiple workers and each worker can be employed at only one firm. While firm utilities over subsets of workers are common knowledge, worker disutilities for working at each firm are private information. The VCG mechanism is the unique mechanism that is strategy-proof, efficient, and individually rational for workers; however, it may not be individual rational for firms. We show that the VCG mechanism is individually rational for firms if and only if firm utilities satisfy a condition called weak substitutes. We then strengthen participation constraints of firms to {\sl strong individual rationality}, which requires that each firm has no incentive to fire some of the workers assigned to it. The VCG mechanism is strongly individual rational if and only if firm utilities satisfy submodularity. |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2605.01715 |
| By: | Alfred Galichon; Yu-Wei Hsieh; Antoine Jacquet |
| Abstract: | We propose an aggregate notion of non-transferable utility (NTU) stability for decentralized matching markets with fixed prices, where market clearing is achieved through one-sided money burning, which can be interpreted as waiting. Agents are grouped into observable types and are indifferent among individuals within type; equilibrium is defined at the type level and delivers equal indirect utility within each type. We introduce money burning into two types of NTU models: In a deterministic model, we relate our notion to classical Gale--Shapley stability and show how money burning decentralizes stable outcomes under aggregation. We then introduce separable random utility, obtaining an NTU counterpart to Choo and Siow (2006). We prove the existence and uniqueness of equilibrium and provide a stationary queueing interpretation. Finally, we develop a generalized deferred acceptance algorithm based on alternating constrained discrete-choice problems and prove its convergence to the unique equilibrium. |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2605.07528 |
| By: | Lauri Lov\'en; Sasu Tarkoma |
| Abstract: | Eliciting truthful reports from autonomous agents is a core problem in scalable AI oversight: a principal scores the agent's report using a strictly proper scoring rule, but the agent also benefits from the report through a non-accuracy channel (approval for autonomous action, allocation share, downstream control). The same structure appears in classical mechanism-design settings such as marketplace operation. Our main result is an endogeneity: the principal's optimal oversight necessarily uses a non-affine approval function to screen types, yet any non-affine approval makes truthful reporting suboptimal under the combined objective whenever deviation is undetectable. The principal cannot avoid the perturbation that undermines calibration. This impossibility holds for all strictly proper scoring rules, with a closed-form perturbation formula. A constructive escape exists: a step-function approval threshold achieves first-best screening for every strictly proper scoring rule, because the agent's binary inflate-or-not choice creates a type-space threshold regardless of the generator's curvature. Under the Brier score specifically, the type-independent inflation cost yields a welfare equivalence between second-best and first-best; we prove this equivalence is unique to Brier (the welfare gap under smooth $C^1$ oversight is bounded below by $\Omega(\text{Var}(1/G'') (\gamma/\beta)^2)$ for every non-Brier rule). Two instances develop the framework: AI agent oversight (the lead motivating setting) and marketplace operation (a parallel mechanism-design domain). The message for AI alignment is direct: smooth scoring-based oversight cannot elicit truthful reports from a strategic agent; sharp thresholds are the calibration-preserving design. |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2605.07671 |
| By: | Chiara Canta; Leonardo Madio; Andrea Mantovani; Carlo Reggiani |
| Abstract: | Online platforms connecting physicians and patients are increasingly common and often operate in heavily regulated contexts. We consider a platform that provides cost-reducing services for physicians and quality-enhancing services for patients. The platform also improves the matching between patients and physicians, thereby increasing competition among the latter. When prices are unregulated, physicians charge different prices online and offline, yet not all join the platform, which is suboptimal in terms of social welfare. The platform may also under- or over-invest in the quality level offered to patients, making their participation suboptimal as well. We then analyze price regulation. Under a single regulated price for medical visits, regardless of the booking channel, all physicians join the platform. However, the first-best allocation cannot be implemented: patient participation remains inefficiently low because patients do not internalize the platform’s cost-reducing effect. In contrast, allowing two regulated prices, one for offline visits and one for platform bookings, restores the first best. Overall, our findings suggest that an optimal pricing or reimbursement mechanism should differentiate across booking channels. |
| Keywords: | healthcare online platforms, price regulation, patient-physician matching |
| JEL: | I11 I18 L51 H75 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_12646 |
| By: | Yasunori Okumura |
| Abstract: | This note proposes a simple polynomial-time method for constructing an ex ante stable school-choice lottery satisfying equal treatment of equals. The method applies the ETE reassignment to a constrained efficient stable matching and yields a lottery that is not ordinally dominated by any other ex ante stable lottery. |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2605.06721 |
| By: | Guillermo Alonso Alvarez; Ibrahim Ekren; Liwei Huang |
| Abstract: | We study a principal-agent problem with adverse selection, where the principal does not know the agent's true cost but must design a contract to optimize a specific criterion. Unlike standard screening frameworks that allow for self-selection, we assume the principal can only offer a unique contract. We show that the agent's optimization problem can be reformulated as a stochastic target problem. After characterizing the credible domain of this target problem, we show that the principal's objective can be solved as a stochastic optimal control problem with partial information and state constraints. The description of the credible domain also allows us to obtain the value of screening contracts. |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2605.01080 |
| By: | Kentaro Kawato; Shosei Sakaguchi |
| Abstract: | This paper studies sample-size design for finite-population test-and-roll experiments, where a decision-maker first conducts an experiment on $m$ units and then assigns the remaining $N-m$ units to the treatment that performs better in the experiment. We consider welfare-aware sample-size choice, which involves an exploration-exploitation tradeoff: larger experiments improve the rollout decision but impose welfare losses on experimental units assigned to the inferior treatment. We show that the standard absolute minimax regret criterion can lead to implausibly small experiments by over-penalizing exploration in its worst-case objective. To address this limitation, we propose the Worst-case Marginal Benefit (WMB) rule, which compares the worst-case marginal benefit of adding one more matched pair to the experiment with the corresponding marginal exploration cost. We establish a simple rule-of-thirds benchmark. For Bernoulli outcomes, after excluding pathological cases, the WMB criterion yields the optimal sample size of $m \approx N/3$ through a Gaussian approximation. For Gaussian outcomes with a known common variance, the same benchmark arises exactly. These results provide a prior-free and practically implementable guide for welfare-based sample-size design. |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2605.02414 |
| By: | Zhicheng Du; Yingkai Li; Boli Xu |
| Abstract: | A principal and $n\ge 2$ agents can launch a project if the principal proposes it and at least $k$ agents accept. Their individual payoffs from the project depend on an ex ante unknown state. The principal can conduct a test to learn about the state and then communicate her findings to the agents via cheap talk. This paper focuses on comparing two communication regimes: public and private messaging. We show that public messaging is weakly dominant: any outcome implementable under private messaging can also be implemented under public messaging. Moreover, in a canonical environment with linear payoffs, we characterize the principal's optimal test in each regime and show that public messaging can be strictly dominant if and only if there exist two agents who are the principal's conflicting allies. |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2605.03621 |
| By: | Alex Chan; Ayush Gupta; Yetong Xu |
| Abstract: | Cadaveric organ shortages leave thousands without life-saving transplants each year. Countries differ in using opt-in (informed consent) or opt-out (presumed consent) systems for donor registration. Using newly assembled cross-country panel data and an event-study design, this paper provides evidence that presumed-consent laws increase organ donation only when strictly enforced and family veto power is limited; weak opt-out regimes show negligible or even negative effects. A theoretical signaling model provides a plausible mechanism when opt-in or opt-out yields more donations, emphasizing the roles of donation propensity, signaling costs, and the family’s ability to overturn defaults. A large laboratory experiment further tests these mechanisms, showing that opt-in generally produces equal or higher donation rates unless signaling is costly and family veto power is minimal. The results underscore that defaults alone rarely increase donations unless paired with strong institutional enforcement. |
| JEL: | C91 C92 D47 D64 H00 I11 I18 |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:35169 |