Research projects

Are you a Zombie Firm? An Early Warning System Based on Machine Learning Methods (with Thomas Heil and Franziska Peter) – Paper

Abstract: This paper develops an early warning system based on machine learning methods and logistic regressions to predict zombie firms. We use feature selection methods on large data sets of listed firms from Europe and the US to find the most important variables that separate a zombie firm from a recovered zombie. We find that beyond debt and income, taxes and equity are recurring features. Altogether, we document that differently to standard pre-selected variables, an ensemble of features related to the firm capital, financial, and industry structure are needed to predict zombie firms and recovered zombies.

Presentations: IFM Brown Bag 2020 (University of Bern), EFMA 2021 (University of Leeds), Swiss Society of Economics & Statistics Annual Congress 2021 (University of Zurich), EEA Women in Economics 2021, Bern Data Science Day 2021 (University of Bern), Big Data and Machine Learning in Finance Conference 2021 (Polytechnic of Milan), CEBRA Annual Meeting 2021 (MIT Golub Center for Finance and Policy), 28^th Finance Forum (Nova SBE), IFABS 2021 Oxford Conference (University of Oxford), Tepper Finance Brown Bag 2021 (Tepper School of Business, Carnegie Mellon University, Pittsburgh), Young Scholars Conference on Machine Learning in Economics & Finance 2021 (Federal Reserve Bank of Philadelphia), AEA/ASSA Annual Meeting 2022 (AEA Poster Session), 8^th Emerging Scholars in Banking and Finance Conference 2022 (Bayes Business School, London), OECD Seminar 2023 (OECD, Paris).

Media: ifo Institute, Frankfurter Allgemeine Zeitung

How Does Competition Affect Zombie Firms? (with Marc Brunner and Philip Valta) – Paper

Abstract: This paper analyzes the effects of product market competition on zombie firms in the US. We show that the asset-weighted share of zombie firms at the industry level decreases significantly with more competition. This decrease is mostly pronounced in industries characterized by low concentration and low margins. Moreover, neither the exit or default probability, nor the recovery likelihood are significantly affected by changes in competition. Finally, at the firm level, zombie firms grow more slowly, reduce their total assets and cash holdings, issue less equity, and obtain smaller loans with more competition. These findings suggest that zombie firms adapt to higher competition by reducing the size of their business.

Presentations: Gerzensee Alumni Conference 2021 (Gerzensee, Switzerland), Bank of England, Wadham College and Waseda University 2023 Workshop on “Financial Frictions, Zombie Firms and the Macroeconomy” (University of Oxford), Swiss National Bank/SIAW-University of St.Gallen 2023 Conference on “Corporate Distress as Financial Conditions Tighten” (Zurich, Switzerland), IBEFA Summer Meeting at 2024 Western Economic Association International (Seattle, USA).

When Companies Don’t Die: Analyzing Zombie Firms in a Low Interest Rate Environment
Swiss Review of International Economic Relations, Volume 73, Issue 1, pp. 67-85, 2023

Abstract: We examine whether low interest rates foster non-viable firms in Europe by analyzing two classes of firms: zombies and distressed. Controlling for the business cycle and recession periods, we find a significantly negative effect of short-term rates on the likelihood of being a zombie, while no effect for distressed firms is detected. A decrease in inflation and a lower state of the business cycle is associated with a rise in both zombies and distressed firms. Examining a non-conventional monetary policy program, we find no evidence of credit misallocation. Therefore, concurring monetary and macroeconomic phenomena likely explain the presence of non-viable firms, although with dissimilarities between zombies and distressed firms.

Media: The FinReg Blog, Duke Financial Economics Center

The Effects of Organized Crime on Distressed Firms – Paper

Abstract: This study captures the presence of organized crime groups in the Italian territories and examines their relationship to firms that are in a condition of financial distress. At the end of the nineteenth century the first criminal groups emerged in the South of Italy as a response to the demand for private protection by local landlords. Using firm-level data and detailed information on organized crime activities, this paper tests the hypothesis that organized crime groups support financially distressed companies by supplying them with private protection in the form of access to credit. The results show a higher concentration of distressed firms in regions with a high presence of organized crime groups.

Presentations: Economics of Crime Seminar 2020 (Jennifer Doleac Seminar Series), EEA Women in Economics 2020, European Public Choice Society Meeting 2018 (University of the Sacred Heart, Rome), 6^th Lindau Nobel Laureate Meeting (Lindau, Germany), German Economic Association Meeting 2017 (University of Vienna), Italian Doctoral Workshop (Collegio Carlo Alberto, Torino).

Conference discussions

Daniel Giamouridis, Chara Prassa, Debt Renegotiation, Default Risk and Risk-Shifting Incentives, European Financial Management Annual Meeting, University of Leeds, 2021.

Federico Huneeus, Joseph P. Kaboski, Mauricio Larrain, Sergio L. Schmuckler, Mario Vera, The Distribution of Crisis Credit and Firm Indebtedness, CEBRA Annual Meeting, MIT Golub Center for Finance and Policy, 2021.

Roberto Daluiso, Emanuele Nastasi, Andrea Pallavicini, Stefano Polo, Reinforcement Learning for Options on Target Volatility Funds, Big Data and Machine Learning in Finance Conference, Polytechnic of Milan, 2021.

Andreas Dibiasi, Klaus Abberger, Michael Siegenthaler, Jan-Egbert Sturm, The Effects of Policy Uncertainty on Investment: Evidence from the Unexpected Acceptance of a Far-Reaching Referendum in Switzerland, European Public Choice Society Meeting, University of Budapest, 2017.