Abstract: Investors act as a liquidity back-stop in the corporate bond market. By providing liquidity, investors help ease dealers’ balance sheet constraints, especially during market stress. During the March 2020 Dash-for-Cash, in bonds where investors stopped providing liquidity, transaction costs rose by 38%. We find the composition of types of liquidity providers – rather than just their presence – shapes trading costs. Dealers relying on flexible-mandate investors, such as hedge funds, are more resilient to liquidity shocks. Dealers offer discounts to investors for past liquidity services to maintain liquidity provider networks. These discounts represent two-thirds of relationship discounts.
Does big data devalue traditional expertise? Evidence from Active Funds Managers. With Maxime Bonelli. Latest draft: May 2026. Forthcoming in the Journal of Finance.
Abstract:We investigate how the availability of alternative data affects the performance of active mutual funds that rely on traditional expertise to produce information. To do so, we evaluate the impact of the release of stock-specific data, which provide new information but require data science expertise to leverage. We find that this release significantly reduces mutual funds’ stock-picking abilities in covered stocks, with a stronger effect for funds that rely on traditional expertise, like industry specialization, leading them to divest from covered stocks. Alternative data can therefore reshape the determinants of fund performance by reducing the value of traditional information sources.
The horizon of investors’ information and corporate investment. With Olivier Dessaint and Laurent Frésard. Latest Draft: March 2026. Accepted for publication at the Review of Financial Studies.
Abstract: We study how the quality ofinvestors’information across horizons influences investment. In our theory, managers care about how investment is impounded in current stock prices. Because prices imperfectly reflect investment’s value, they under-invest. However, they under-invest less when investors have better information about thehorizon matching that of their projects. Using a measure of projects’ horizon obtained from the text of regulatory filings, we find that improvements in investors’ long-term (short-term) information induce firms with long-term (short-term) projects to invest more, especially when managers focus on current stock prices. Therefore, the quality ofinvestors’information across horizons has real effects.
Trade-off? What trade-off: Information Production without Illiquidity. With Kostas Koufopoulos and Roman Kozhan. First Draft: November 2024. Latest draft: March 2026.
Abstract: Private information in financial markets improves the informativeness of asset prices and guides resource allocation. However, informed trading generates rents at the expense of uninformed traders, creating a trade-off between price informativeness and liquidity. In addition, private incentives to acquire information need not align with the value of that information for agents who rely on prices to make decisions. We show that this trade-off can be eliminated by a market structure that separates information production from liquidity provision. In such a structure, prices remain informative while liquidity is preserved, and incentives to acquire information align with its value for decision-makers.
The price of Exchange data. With Vincent Maurin. First draft: July2026.
Abstract: Exchanges increasingly generate revenues from selling transaction data, in addition to their traditional business of matching traders. Given exchanges’ market power over their data, this raises questions about the quality and pricing of these data, and exchanges’ contribution to market transparency. We study the optimal design of an exchange that sells both access to its trading platform and transaction data generated by traders’ activity. We show that the introduction of a market for transaction data reduces trading fees and increases volume and welfare. Caps on data fees can increase data quality but lower welfare.
Alternative Data, Firm Ownership and Stock Price Fragility. First Draft: September 2026.
Abstract: We study how technological innovations in information production affect firms’ ownership structures. Focusing on the expansion of commercial alternative data, we develop a novel text-based measure of firm-level alternative data coverage. We show that greater alternative data coverage reallocates ownership away from long-term investors and toward hedge funds, which trade more actively and shorten their holding horizons. While this ownership reallocation reduces stock return comovement with the market and industry, consistent with greater incorporation of firm-specific information into prices, it also increases stock price fragility: During market downturns, hedge funds reduce their holdings more aggressively, making highly covered firms’ stock prices more sensitive to market-wide shocks. Overall, our findings show that ownership reallocation is one channel through which changes in firms’ information environment can increase stock price fragility.
Old working papers
“Linkage Principle, Multidimensional Signals and Blind Auctions”, with Stefano Lovo, 2004 (draft on SSRN)
“Price formation and order placement strategies in a dynamic order driven markets”, 1995 (draft)
Thierry Foucault 