Helene Mass

Publications

Worst-case  equilibria in first-price auctions, Theoretical Economics, 2023 (with Vitali Gretschko)

The usual analysis of bidding in first-price auctions assumes that bidders know the distribution of valuations. We analyze first-price auctions in which bidders do not know the precise distribution of their competitors’ valuations, but only the mean of the distribution. We propose a novel equilibrium solution concept based on worst-case reasoning. We find an essentially unique and efficient worst-case equilibrium of the first-price auction, which has appealing properties from both the bidders’ and the seller’s point of view.

Two-dimensional information acquisition  in social learning, Journal of Economic Theory, 2022 (with Nina Bobkova)

We analyze a social learning model where the agents' utility depends not only on an unknown common component but also on an unknown idiosyncratic component. Each agent splits a learning budget between the two components. We characterize the optimal learning decisions and find necessary and sufficient conditions for complete learning.  As agents learn from the decisions of previous agents, information about the common component is never fully aggregated unless agents do not have to give up any information about the idiosyncratic component to learn marginally more about the common component. Allowing agents to communicate their signals accelerates learning but does not change the learning outcome asymptotically.

Imitation perfection - a simple rule to prevent discrimination in procurement,  American Economic Journal: Microeconomics, 2020  (first author, with Nicolas Fugger, Vitali Gretschko, and Achim Wambach)                                           

Procurement regulation aimed at curbing discrimination requires equal treatment of sellers. However, Deb and Pai (2017) show that such regulation imposes virtually no restrictions on the ability to discriminate. We propose a simple rule – imitation perfection – that restricts discrimination significantly. It ensures that in every equilibrium bidders with the same value distribution and the same valuation earn the same expected surplus. If all bidders are homogeneous, revenue and social surplus optimal auctions which are consistent with imitation perfection exist. For heterogeneous bidders however, it is incompatible with revenue and social surplus optimization. Thus, a trade-off between non-discrimination and optimality exists.    

Working Papers

The role of uncertain evaluation standards in voluntary disclosure  (with Avi Lichtig)

We study a voluntary disclosure model in which the receiver can gather partial information independently, by means of a test – a signal with at most k realizations. We characterize the optimal test that balances two objectives: maximizing informativeness when the sender withholds information, and incentivizing disclosure when possible. The optimal test involves randomized evaluation standards, sacrificing precision to improve disclosure incentives. Our results highlight how coarse information structures, common in regulatory practice, can be strategically designed to enhance disclosure in environments with conflicting interests.

First-price auctions under uncertainty - maximin selection from rationalizable strategies 

I investigate the strategic problem of a player in a first-price auction who faces uncertainty about the other bidders' strategies and valuation distributions. I propose a new method to derive strategies in such a setting -- the rational maximin criterion --  which works in two steps. First, the player assumes rationality and common belief in rationality and eliminates all strategies that are not rationalizable. Second, the player applies the maximin criterion. Under a common prior and strategic uncertainty, the criterion induces pure strategies and efficient outcomes. If in addition, distributional uncertainty is present, bidders adopt mixed strategies causing inefficient outcomes.

Work in Progress

Optimal transparency in task design  (with Claudia Herresthal)

A principal has to select an agent among a group of agents. The agents exert costly effort to obtain knowledge on some subject and the principal's objective is to maximize how much the selected agent has learnt. The principal can observe how much agents have learnt but once she observed it, she cannot commit to not to choose the agent who has the learnt the most. However, the principal can let the agents work on a task and commit to observe only the results of the task. The principal can determine the transparency of the task, i.e., how informative the results are about the agents' knowledge. The principal can also determine the difficulty of the task, i.e.,  how the results on the task depend on how much agents have learnt. We characterize the optimal choice of transparency and difficulty given that agents get a utility from being selected and a utility from good results on the task. 

Gaming the test  - when should test thresholds be unpredictable?  (with Ludmila Matyskova)

We study the design of thresholds in pass/fail tests. The principal aims for the agent to pass when the agent’s natural type (e.g. ability) is sufficiently high. However, the agent can manipulate the perceived natural type at a cost. Randomizing the passing threshold becomes optimal when the principal faces significant uncertainty about the agent’sgaming costs. In particular, any positive probability that the agent cannot game (i.e., has infinite gaming costs) makes randomization optimal. Randomization allows high natural types with high gaming costs to pass, at least with some probability, without incentivizing low ability types to game. Moreover, it incentivizes low natural types to limit their gaming, thereby separating them from high natural types. We identify randomization strategies that allow the principal to achieve better outcomes than a deterministic threshold, without requiring precise knowledge of the distribution of the agent’s natural type and gaming ability. We thereby provide robust rules for when unpredictable testthresholds are justified.