Research

When exporters price their goods in a foreign currency, they are exposed to exchange-rate risk. However, they can hedge this risk by underwriting a foreign exchange (FX) forward contract, which means selling forward the currency in which they price their goods. In this paper, we study how the cost of FX hedging influences the currency choice of French exporters. Our identification strategy exploits an exogenous increase in the trading costs of FX forward contracts, that was triggered by a spike in the Greek default risk. First, we find that higher FX trading costs lower the probability of pricing in dollars and in local (i.e., buyer’s) currency for hedging firms. Second, we show that hedging firms price more their goods in dollars than in local currency. Third, we document that FX hedging affects the transmission of exchange-rate shocks to prices and find that FX hedging is associated with lower levels of exchange-rate pass-through. We conclude that FX hedging contributes to dollar dominance and to the exchange-rate disconnect puzzle.

Whether Large Language Models (LLMs) will result in a marginal productivity increase or a technological revolution in decision making largely depends on their ability to reason. LLMs with reasoning capabilities outperform vanilla ones on math and coding. However, it remains unclear whether such emergent abilities translate into improved economic insights. We evaluate state-of-the-art reasoning-enhanced LLMs by OpenAI and DeepSeek on standard financial tasks: news sentiment and earnings direction prediction. Reasoning-enhanced models fail to demonstrate a consistent advantage, while model size does. This finding indicate that improved reasoning does not necessarily translate into enhanced economic understanding, questioning the cost-effectiveness and practical utility for managers.

Using a banking model, we show that the effects of introducing a Central Bank Digital Currency (CBDC) depend on the ongoing monetary policy and the amount of excess reserves. We derive the conditions for a neutral introduction without central bank pass-through funding and find that they do not always hold with quantitative easing, as bank lending shrinks if demand for CBDC is large enough. Moreover, commercial banks optimally liquidate their excess reserves to accommodate households’ demand for CBDC. Consequently, households replace banks on the liability side of the central bank balance sheet, making quantitative tightening difficult to implement.

* This paper previousy circulated as "Central Bank Digital Currency and Quantitative Easing".

A Central Bank Digital Currency (CBDC) would reduce commercial bank deposits and provide households with a new payment technology. We develop a structural model of the banking sector, calibrate it, and introduce a CBDC to run counterfactual analyses. We find that, if the central bank compensates the commercial banks for the loss in deposits, then banks optimally push households towards the CBDC. This allows banks to capture the consumer surplus stemming from the new technology and increase their profit margin. The design of the compensation mechanism can mitigate this effect.

Cryptocurrencies and equities have exhibited a high and positive correlation since March 2020, making cryptocurrencies a poor diversification tool. We show theoretically that trading flows by retail investors can drive this correlation, even without fundamental drivers. Using a unique dataset of investor-level holdings from a bank offering trading accounts and cryptocurrency wallets, we show that retail investors tend to trade equities and cryptocurrencies in the same direction simultaneously. This behavior became prominent in March 2020. We provide evidence showing that stocks preferred by crypto-investors exhibit a stronger correlation with cryptocurrencies, especially when the cross-asset retail volume is high.

This paper examines the rationale behind the sharp increase in direct Government investments in Venture Capital (GVCs) over the past two decades. We find that GVCs are not pure profit maximizers but rather seek to capture positive externalities. Using EU data from 2002 to 2020, we find that GVCs have lower performance but that they target more innovative and geographically dispersed firms. Furthermore, we derive the conditions under which GVCs can crowd in private investments and show consistent empirical evidence. These findings suggest that governments have a role to play in promoting innovation and entrepreneurship.

Does the network of venture capital firms (VC) affect their risk-taking? We use data from Crunchbase to build a network of VCs based on syndications. Consistent with the literature, better-connected VCs have a higher probability of success. Additionally, we find that these funds take deals that are riskier ex-ante. We build a theoretical model of the VC investment process to rationalize these findings. The representative fund decides whether to invest in a startup and whether to syndicate. The results suggest that if the VC is able to internalize the value-added effect of its network, it selects riskier deals.

We find that hedge fund managers from elite universities exhibit unusually high return correlations. A network-based analysis shows that this pattern is especially strong among alumni of Columbia, Harvard, the University of Pennsylvania, Stanford, and NYU. We attribute the effect to private information sharing, supported by a quasi-natural experiment: the 2009 Galleon Capital insider trading scandal. A difference-in-differences analysis reveals a significant drop in returns for elite managers after the scandal. Finally, we show that investors value such access, as managers from elite schools raise 55% more capital when launching their first fund.