Nicolas Eschenbaum

This paper studies reputation in the online market for illegal drugs in which no legal institutions exist to alleviate uncertainty. Trade takes place on platforms that offer rating systems for sellers, thereby providing an observable measure of reputation. The analysis exploits the fact that one of the two dominant platforms unexpectedly disappeared. Re-entering sellers reset their rating. The results show that on average prices decreased by up to 9% and that a 1% increase in rating causes a price increase of 1%. Ratings and prices recover after about three months. We calculate that identified good types earn 1,650 USD more per week.

This paper studies dynamic monopoly pricing for a broad class of settings that allows for multiple durable, multiple rental, or a mix of varieties. We show that the driving force behind pricing dynamics is the existence of trading-up opportunities. If there are no trading-up opportunities in the static monopoly outcome, then pricing dynamics do not emerge in equilibrium. With trading-up opportunities, pricing dynamics arise until these opportunities are exhausted or the game ends. We characterize the lower bound for the emerging prices and profit and study the conditions under which pricing dynamics end in finite time.

This paper develops a formal framework to assess policies of learning algorithms in economic games. We investigate whether reinforcement-learning agents with collusive pricing policies can successfully extrapolate collusive behavior from training to the market. We find that in testing environments collusion consistently breaks down. Instead, we observe static Nash play. We then show that restricting algorithms’ strategy space can make algorithmic collusion robust, because it limits overfitting to rival strategies. Our findings suggest that policy-makers should focus on firm behavior aimed at coordinating algorithm design in order to make collusive policies robust

This paper presents Multi-Agent Deep Hedging (MADH), a computational framework that extends deep learning-based financial planning and pricing techniques to markets with multiple strategic participants. Unlike traditional deep hedging, which assumes a passive market environment, MADH models the market as a multi-agent system by augmenting gradient-based optimization with a differentiable market-clearing mechanism, enabling agents to internalize their market impact and optimize policies through gradient backpropagation. In a detailed case study, we apply MADH to model prosumer behavior on electricity trading platforms. We benchmark decentralized MADH using both synthetic and real-world data against a welfare-maximizing central optimizer. Results demonstrate that decentralized agents robustly converge to high-quality policies, engaging in dynamic hedging and strategic capacity withholding while achieving near-optimal welfare. Moreover, we document that MADH performance scales well to a large setting with 32 agents.