Assistant Professor, Department of Economics, UCL
d.kattwinkel@ucl.ac.uk
Costless Information and Costly Verification: A Case for Transparency
with Jan Knoepfle @ Journal of Political Economy 2023
Comment on Jackson and Sonnenschein (2007) “Overcoming Incentive Constraints by Linking Decisions”
with Ian Ball and Matthew O. Jackson @ Econometrica 2023
Corrigendum to “Role of Linking Mechanisms in Multitask Agency with Hidden Information” [JET, 2010]
with Ian Ball @ Journal of Economic Theory 2023
Quota Mechanisms: Finite-Sample Optimality and Robustness (with Ian Ball)
Conditionally accepted @ Journal of Political Economy
A quota mechanism, such as a mandatory grading curve, links together multiple decisions. We analyze the performance of quota mechanisms when the number of linked decisions is finite and the designer has imperfect knowledge of the type distribution. Using a new optimal transport approach, we derive an ex-post decision error guarantee for quota mechanisms. This guarantee cannot be improved by any mechanisms without transfers. We quantify the sensitivity of quota mechanisms to errors in the designer's estimate of the type distribution. Finally, we show that quotas are robust to agents' beliefs about each other.
Probabilistic Verification in Mechanism Design (with Ian Ball), SSRN
Extended abstract accepted at EC'19, accepted @ Theoretical Economics
We introduce a model of probabilistic verification into the standard mechanism design setting. The principal verifies the agent's type using a statistical test. The result of the test is stochastic; its distribution depends on the agent's true type. The principal commits to a mechanism that assigns a test to each message and then a decision based on the test result. In our framework, the revelation principle holds. We characterize whether each type can be identified with a test. If so, the principal's problem becomes an optimization subject to incentive constraints. Under quasilinear preferences, we solve for revenue-maximizing mechanisms by introducing a new expression for the virtual value that reflects the precision of the tests.
Robust Robustness (with Ian Ball)
Reject and Resubmit @ American Economic Review
The maxmin approach to distributional robustness evaluates each mechanism according to its payoff guarantee over all priors in an ambiguity set. We propose a refinement: the guarantee must be approximately satisfied at priors near the ambiguity set (in the weak topology). We call such a guarantee robust. The payoff guarantees from some maxmin-optimal mechanisms in the literature are not robust. We show, however, that over certain standard ambiguity sets (such as continuous moment sets), every mechanism's payoff guarantee is robust. We give a behavioral characterization of our refined robustness notion by imposing a new continuity axiom on maxmin preferences.
Who and How? Adverse Selection and flexible Moral Hazard (with Henrique Castro-Pires, Jan Knoepfle) New Version
We characterize incentive compatible mechanisms in environments with hidden types and flexible hidden actions. Our approach introduces extended recommendation schedules that specify prescribed actions also off-path, after misreports. This approach yields a tractable and complete characterization of incentive compatibility, which includes a generalized integral monotonicity condition capturing the interaction between adverse selection and moral hazard. We demonstrate the usefulness of the characterization across a range of contracting problems.
A Theory of Auditor Selection ( with Peter Achim, Jan Knoepfle, Alexandra A. Scherf)
How do private interactions between auditors and clients shape auditor selection and audit quality in competitive markets? We develop an empirically grounded economic model of auditor selection in which auditors compete for audit engagements by offering fees. Auditors differ in the match-specific interpersonal chemistry with the client, a quality that influences audit process efficiency and the likelihood of certifying financial reports. We find that the structure of the selection process and the resulting audit quality depends crucially on the value of chemistry in equilibrium. If this value is low, then the audit fee becomes a dominant factor in the client's choice of auditor, leading to lower fees and higher audit quality on average in equilibrium. Conversely, if the value from chemistry is high, then the firm prioritizes good chemistry with the auditor, resulting in lower overall audit quality and higher fees.
The Division of Surplus and the Burden of Proof (with Justus Preusser) New Version
A surplus must be divided between a principal and an agent. Only the agent knows the surplus' true size and decides how much of it to reveal initially. Both parties can exert costly effort to conclusively prove the surplus' true size. The agent's liability is bounded by the revealed surplus. The principal is equipped with additional funds. The principal designs a mechanism that allocates the burden of proof and divides the surplus. In principal-optimal mechanisms, the principal's effort to acquire proof decreases in the revealed surplus. The agent's effort initially decreases, but then the sign of its slope alternates across five intervals. Applications include wealth taxation, corporate finance, and public procurements.
Optimal Decision Mechanisms for Juries: Acquitting the Guilty (with Alexander Winter)
A group of privately informed agents has to choose between two alternatives. How should a principal design the decision rule if agents are known to be biased in favor of one of the options? We address this question within the classical Condorcet jury setting. Applications include the optimal decision mechanisms for boards of directors, political committees, and trial juries.
The optimal mechanism is a non-monotonic voting mechanism. In the terminology of the court example, when jurors are more eager to convict than the lawmaker (principal), then the optimal mechanism convicts the defendant if and only if neither too many nor to few jurors vote to convict. This kind of mechanism accords with a judicial procedure from ancient Jewish law as recorded in the Talmud.
Mechanisms without transfers for fully biased agents (with Axel Niemeyer, Justus Preusser and Alexander Winter)
Extended abstract accepted at EC'22
A principal must decide between two options. Which one she prefers depends on the private information of two agents. One agent always prefers the first option; the other always prefers the second. Transfers are infeasible. One application of this setting is the efficient division of a fixed budget between two competing departments. We first characterize all implementable mechanisms under arbitrary correlation. Second, we study when there exists a mechanism that yields the principal a higher payoff than she could receive by choosing the ex-ante optimal decision without consulting the agents. In the budget example, a profitable mechanism exists if and only if the information of one department is also relevant for the expected returns of the other department.
When types are independent this result generalizes to a setting with n agents. We apply this insight to derive necessary and sufficient conditions for the existence of a profitable mechanism in the n-agent allocation problem.
Allocation with Correlated Information: Too good to be true
Extended abstract accepted at EC'20
A principal can allocate an indivisible good to an agent. The agent privately learns the value of the good while the principal privately learns the cost. Value and cost are correlated. The agent wants to have the good in any case. The principal wants to allocate whenever the value exceeds the cost. She cannot use monetary transfers to screen the agent.
I study how the principal utilizes her information in the optimal mechanism: when the correlation is negative, she bases her decision only on the costs, and when the correlation is positive, she screens the agent. To this end, she forgoes her best allocation opportunities: when the agent reports high valuations but her own costs are low. Under positive correlation, these realizations are unlikely; the principal will find them too good to be true. In contrast to standard results, this optimal mechanism may not allocate to a higher value agent with higher probability. I discuss applications to intra-firm allocations, task-delegation, and industry self-regulation.
Strategic Understatement (with Carl Heese)
We study the design of quality certificates. The certificate holder (sender) tries to convince a receiver of his quality. The receiver benefits from more precise knowledge about the sender’s quality and can costly verify the sender's quality. We show that if the certificate is too selective, in all equilibria, senders of very high quality do not present the certificate; they pool with low-quality types who do not hold a certificate. The high types forgo a signaling opportunity ("understatement"), but thereby motivate the receiver to costly verify their high quality herself. Conversely, when the certificate is not too selective, there are monotone equilibria where higher types are more likely to present their certificates.
Policy Uncertainty and Signaling in Referenda (with Carl Heese)
We study voting on a reform for which the implementation is at the discretion of a politician who is biased against the reform. Valuable information about the efficacy of the reform is dispersed among citizens, and by voting, citizens can transmit their information to the politician. When the politician has no discretion, the Condorcet Jury Theorem holds (Feddersen & Pesendorfer, 1998). When the politician has full discretion, in all equilibria, there is no information transmission at all (Battaglini, 2017). We consider a scenario of "intermediate discretion" akin to political referenda. When the majority votes against the reform, the status quo is kept. Otherwise, the politician is bound to implement the reform but can choose the degree of implementation. This creates one-sided signaling incentives for the voters. Even when the politician is only slightly biased against the referendum, there are equilibria with "oversignaling" for which the reform always passes, even in the situations where all citizens and the politician prefer it not to.
I'm co-organizing the yearly Southeast Theory Festival with Ludvig Sinander: Southeast Theory Festival (2022,2023,2024,2025,...)
I'm co-organizing UCL's Centre for Theoretical and Behavioural Economics seminar with Duarte Gonçalves
I co-organized a conference with Vasiliki Skreta, Amir Habibi, and Mia Guo Bai in remembrance of Konrad Mierendorff this June: Conference for Konrad
I'm teaching ECON0029: Economics of Information (undergrad) @ UCL
I'm teaching ECON0083: Foundations and Frontiers of Mechanism Design (PhD) @ UCL