Job Market Paper

Black Box Credit Scoring and Data Sharing [Paper]
Abstract: Should credit scoring algorithms be transparent or opaque? I study this question in a model where a lender uses data shared by borrowers for allocating and pricing credit, and is privately informed about how the algorithm maps data to allocations. I show that revealing the algorithm's parameters exposes it to gaming in the form of strategic withholding of unfavorable information. Under opacity, data withholding emerges as a prudent strategy against the unpredictability of the black box and the risk of credit rationing. The lender's optimal transparency regime maximizes data collection and is socially inefficient as it results in excessive credit rationing.  Algorithmic opacity can be welfare-improving as it reduces the stigma around data withholding, thereby expanding credit access for privacy-concerned borrowers. I analyze the distributional impacts of recent algorithmic transparency regulations and offer policy recommendations.
Presentations: EFA Doctoral Tutorial (Bratislava), EUROFIDAI-ESSEC Paris December Finance Meeting (Paris), EEA-ESEM (Rotterdam), HEC Paris Finance PhD Workshop (Paris), ENTER Summer Jamboree (Bruxelles).
Awards: Best PhD Paper Award (EUROFIDAI-ESSEC Paris December Finance Meeting 2024).

Working Papers

Market Information in Banking Supervision: the Role of Stress Test Design [Paper] (R&R at The Review of Financial Studies)
(with Haina Ding and Alexander Guembel)
Abstract: The Basel committee views market discipline as complementing banking supervision. This paper studies how supervisors should design stress tests when markets discipline banks via price signals their traded securities provide to bank creditors. We show that the optimal stress test is coarse and lenient. Speculators have incentives to identify bad banks that erroneously passed the test, which makes markets useful at reducing the type-2, but not the type-1, error of a stress test. Our results hold even when the supervisor can intervene directly based on private information. In the limit of costless supervisory interventions, the optimal stress test is uninformative.
Presentations (including by coauthors): OxFIT Conference (Oxford), FTG Summer Meeting (Stockholm), European Winter Meetings of the Econometric Society (Manchester), FTG Summer School (Seattle), London Business School (London), Universidad Carlos III (Madrid), Toulouse Business School (Toulouse).

Agents under Pressure: Risk Governance in a Rat Race  
(with Matthieu Bouvard and Samuel Lee)
Abstract:  We study agency problems in risk management for a market with preemptive competition. Competition weakens risk governance, favoring contracts that do not incentivize compliance. Externalities that induce constrained inefficiency operate partly through the agents' contracts. Replacing human agents with (faster) machines has ambiguous effects on governance quality. There is scope for compensation regulation to correct both contractual and direct externalities but levying ex-post penalties on firms and agents may be a more practicable alternative.