Working Papers


Abstract: As financial advisors increasingly adopt model portfolios, the allocation of retail capital is shifting from advisors' discretionary judgment toward externally determined rules.  In this paper, I study whether this shift changes the incentives of fund managers competing for capital. I find that model portfolios reduce fund managers' incentives to compete on performance: model-implied demand is a significant driver of fund flows, generates a more stable capital base than discretionary investor demand, and buffers funds against performance-driven outflows. Consistent with this, fund managers who are more exposed to model portfolios take less risk in bad states and show weaker persistence in their performance. Moreover, they tend to cut fees around the time of a new model inclusion. Overall, these results suggest that model portfolio growth is partially shifting competition among fund managers from performance toward inclusion in recommendation platforms. In the aggregate, investors benefit from model adoption by being steered away from the worst-performing funds and by capturing a larger share of their funds' realized returns.

Recipient of the Swiss Finance Institute Best Paper Doctoral Award (2026)

Main presentations: DGF PhD Workshop (Hagen, 2025); SFI Research Days (Gerzensee 2026); 19th International Risk Management Conference - IRMC (Warsaw, 2026); HEC Paris Finance PhD Workshop (Jouy-en-Josas, 2026); 5th NSE PhD and Post-Docs Workshop (Naples, 2026); SASCA Conference (Venice, 2026); 32nd DGF Annual Meeting (WHU-HHU, Düsseldorf, 2026).


Abstract: We study how sophisticated investors form monetary policy expectations and how those expectations shape their portfolios and performance. Using a large corpus of hedge fund investor letters, we build a text-based measure of managers' expectations with a fine-tuned large language model. The expectations are consistent with a Taylor rule and respond to identified monetary policy shocks, especially to forward guidance. The aggregate consensus predicts future changes in the yield curve beyond market-implied rates, survey forecasts, and policy rules, and funds whose expectations are more accurate earn higher risk-adjusted returns. Consistent with anticipatory de-risking, a more hawkish consensus is associated with lower stated leverage, a less favorable view of emerging markets, and a stronger preference for the U.S. dollar, and with realized returns that carry less duration and less emerging-market risk. The evidence points to an expectations-based channel of monetary policy transmission through leveraged financial intermediaries. 



Abstract: We study whether corporate risk disclosure is informative about anticipated tail risk and how that information is reflected in asset prices. From the narrative risk-factor sections of 10-K filings, we build a text-based, firm-level measure of tail-risk disclosure and show that it captures independently measured tail-risk exposure. Firms with more extensive disclosure earn higher expected returns, consistent with investors demanding compensation for known downside exposures. Yet when disclosed risks materialize, these firms experience smaller and less persistent price declines than comparable non-disclosers, indicating that disclosure shifts adjustment from abrupt ex post repricing toward gradual ex ante pricing. We confirm this timing mechanism in two complementary event settings. When a previously disclosed risk materializes, prior disclosers experience smaller abnormal losses. When the perceived materiality of an existing risk increases without new firm-level information,firms that had disclosed the exposure are repriced first. The results establish risk disclosure as a measure of anticipated tail risk and show that transparent communication shapes when downside risk enters prices.


Abstract: The trading behavior of U.S. Senators has faced growing public scrutiny due to potential misuse of insider information. I document a novel empirical fact: stock trades by U.S. Senators predict abnormal returns in the same direction after disclosure exceeding 90 bps in a month, but not after the actual transaction. I reconcile this evidence by isolating a set of potentially speculative trades (multiple purchases and sales of the same stock), which frequently originate from industries under considerable governmental oversight and public controversy. Realized returns on speculative trades earn more than 9 bps daily. Since the speculative nature becomes apparent only ex-post, there is uncertainty on the trade type at the disclosure time. Consequently, the market may overreact or mistakenly replicate also uninformed trades. In accordance, the price impact after disclosure is only temporary. Nonetheless, a trading strategy timed on disclosure yields substantial financial gains, even while being agnostic about the nature of the underlying trade. Overall, this study suggests a re-evaluation of the effectiveness of public disclosure as a disciplining mechanism for political insiders. Indeed, politicians are still able to profit abnormally on many of their trades. At the same time, disclosure may unintentionally act as a catalyst in fostering noise trading. 

Main presentations: : USI Lugano, Bocconi-SFI PhD Workshop 2023, SFI Research Days 2024, Wolfe Research LLC (Invited, 2025).


Other Working Papers


Abstract: Obituaries are traditionally seen as expressions of grief and remembrance. We argue that they also have an underappreciated economic role: they are vehicles for strategic social and economic signaling. In this paper, we develop a simple theoretical framework in which paid obituaries serve as a form of self-promotion, especially when the deceased is a prominent public figure. We then test this hypothesis using data from Italy, exploiting the variation in mortality caused by the COVID-19 pandemic as a natural experiment. We show that higher mortality rates are associated with increases in per-capita obituaries, driven not by informational needs but by strategic advertising motives. Our results suggest that obituaries function as a marketplace for visibility and status, where social and economic incentives intersect.  



Abstract: The interest for the dynamics of natural disasters has significantly expanded during the last decades, from natural sciences to other fields, including economic research, due to the global increase in frequency and socio-economic impact of such events. Economic evaluation of natural disasters nevertheless constitutes a complex and multidisciplinary field. This note presents the Natural Disasters Database for Italy (NDDI), a data set obtained by combining different data sources and describing the evolution of natural disasters and their impact on public spending starting from the Second World War.

Media: Media: ItaliaOggi (09/08/19), OggiScienza, AUA-UnipolSai, Panorama Assicurativo, Giornale delle PMI, AgoraVox.