Economics

• The Uncertainty Premium of Climate Tipping Points (2023 - ongoing) - R&R @ JEEM
  This paper studies how deep uncertainty about climate tipping points affects optimal climate policy and the social cost of carbon. Using an integrated assessment model with temperature feedbacks, it shows that tipping point uncertainty raises the social cost of carbon by 12–50%, adding roughly $20–40 per tonne when tipping points occur below 2.5°C of warming. While earlier detection of tipping points reduces this uncertainty premium by about 9%, the results highlight that near-term emissions reductions remain the most effective way to limit tipping-point risk.

• Endogenous Fragility of Supply Chains and Correlated Disruption Risk (2022 - 2024) - Under Review
  Models the endogenous formation of supply chains under correlated disruptions, showing that greater correlation among upstream producers reduces downstream diversification incentives and can trigger widespread under-diversification and large welfare losses; improved information about supply-chain risk can further exacerbate these losses.

Work in progress

• Reputation of Climate Policy with Inès Mourelon (2026-ongoing)
  We develop a continuous-time reputation game in which firms choose abatement based on the climate policy they expect to face, while a non-committed government re-optimizes policy sequentially. Firms do not observe the government’s type directly and update their beliefs from noisy public signals. Reputation therefore becomes an endogenous state variable that affects both firms’ abatement decisions and the government’s incentive to deviate from the committed policy path. We characterize the Markov sequential equilibrium as a function of cumulative emisions and policy reputation. A government with a stronger reputation can set a lower current climate policy while keeping firms’ expected carbon price and abatement relatively high. This policy flexibility is constrained by learning, as deviations from the committed path reveal information about the government’s type and erode its reputation over time. Noisier policy signals weaken this constraint and allow the government to deviate further. The value of reputation is greatest when reputation is already high and cumulative emissions remain relatively low.

• The Role of Firm Coordination in Delaying the Green-Transition: An Experimental Approach with Andrea Marietta Leina and Emanuele Campiglio (2026-ongoing)
  This paper studies whether firms can strategically delay decarbonisation in order to induce the postponement of a regulatory deadline. Firms are endowed with carbon-intensive capital and choose dynamic retirement paths subject to convex adjustment costs. The regulator sets an initial compliance date but postpones it if aggregate remaining carbon-intensive capital exceeds a threshold. This creates a dynamic coordination problem: firms may either invest early to meet the original deadline or coordinate on slow retirement to make postponement optimal. We test these mechanisms in a laboratory experiment in which participants represent firms in six-firm economies. The design first compares environments without communication to environments with non-binding messages about intended future capital holdings, capturing the role of net-zero-style signals in coordinating expectations. It then introduces three policy treatments aimed at disrupting coordination on delay: ambiguity about the regulator’s threshold, risk with a known threshold distribution, and redistribution from high- to low-carbon-capital firms. The main outcomes are deadline postponement, compliance with the aggregate target, individual retirement paths, within-economy dispersion, and the relationship between announced and realised choices.

• Options can stabilise markets with Donald Hagesteijn and Cars Hommes (2024 - ongoing)
  We show that trading binary at-the-money put option can stabilise markets and mitigate bubble formation, in asset pricing models with trend-following agents

Neuroscience

• Bayesian Model of Motion Extrapolation (with Philippa Johnson) - 2022
  We model the brain compensation of neural delays in motion perception.