Research

Publications and Working Papers

Published Papers

Smart Systemic-Risk Scores, February 2024, Journal of International Money and Finance, 140, 102968. [Companion Website and CASCaD Certification]  

Abstract: This paper proposes a new systemic-risk score to identify and regulate global systemically important banks (G-SIBs) by using an alternative weighting scheme based on volatility to aggregate all systemic-risk facets. Following a portfolio management approach, I equalize the risk contribution of each systemic-risk component to the cross-sectional volatility of the smart systemic-risk scores. The equally-weighted risk contribution (ERC) method appears to be a relevant alternative to the cap on the substitutability category. To discriminate between several systemic-risk scores, I modify and apply the axiomatic framework of Chen et al. (2013) to express supervisor preferences among systemic-risk scores. Such preferences are based on the expected value of the cross-sectional dispersion of systemic-risk scores over the years.

Pitfalls in Systemic-Risk Scoring (with Christophe Hurlin and Christophe Pérignon), April 2019, Journal of Financial Intermediation, 38, 19-44. [Companion Website]

Abstract: We identify several shortcomings in the systemic-risk scoring methodology currently used to identify and regulate Systemically Important Financial Institutions (SIFIs). Using newly-disclosed regulatory data for 119 US and international banks, we show that the current scoring methodology severely distorts the allocation of regulatory capital among banks. We then propose and implement a methodology that corrects for these shortcomings and increases incentives for banks to reduce their risk contributions. Unlike the current scores, our adjusted scores are mainly driven by risk indicators directly under the control of the regulated bank and not by factors that are exogenous to the bank, such as exchange rates or other banks' actions.

Where the Risks Lie: A Survey on Systemic Risk (with Jean-Edouard Colliard, Christophe Hurlin and Christophe Pérignon), March 2017, Review of Finance, 21(1), 109-152.

Abstract: We review the extensive literature on systemic risk and connect it to the current regulatory debate. While we take stock of the achievements of this rapidly growing field, we identify a gap between two main approaches. The first one studies different sources of systemic risk in isolation, uses confidential data, and inspires targeted but complex regulatory tools. The second approach uses market data to produce global measures which are not directly connected to any particular theory, but could support a more efficient regulation. Bridging this gap will require encompassing theoretical models and improved data disclosure.

Implied Risk Exposures (with Christophe Hurlin and Christophe Pérignon), October 2015, Review of Finance, 19(6), 2183-2222. [Companion Website]

Abstract: We show how to reverse-engineer banks’ risk disclosures, such as Value-at-Risk, to obtain an implied measure of their exposures to equity, interest rate, foreign exchange, and commodity risks. Factor Implied Risk Exposures (FIRE) are obtained by breaking down a change in risk disclosure into an exogenous volatility component and an endogenous risk-exposure component. In a study of large US and international banks, we show that (i) changes in risk exposures are negatively correlated with market volatility and (ii) changes in risk exposures are positively correlated across banks, which is consistent with banks exhibiting commonality in trading. 

Where is the System?, August 2014, International Economics, 138, 1-27.

Abstract: The aim of this paper is to determine the optimal size of the system (global, supranational or national) when measuring the systemic importance of a bank. Since 2011, the Basel Committee on Banking Supervision (BCBS) has tagged global systemically important banks (G-SIBs) and has imposed a higher regulatory capital of loss absorbency (HLA) requirement. However, the identification of G-SIBs may overlook banks with major domestic systemic importance, i.e. the domestic systemically important banks (D-SIBs). This paper describes how to adjust market-based systemic risk measures to identify D-SIBs. In an empirical analysis within the eurozone, I show that (1) the SRISK methodology produces similar rankings whatever the system used. However, (2) the SRISK values greatly vary across systems, which calls for imposing the higher of either D-SIB or G-SIB HLA requirements. Finally, (3) the ΔCoVaR methodology is extremely sensitive to the choice of the system.

Working Papers

Safe Distance To Systemic Risk (with Renzhi Liu), December 2024

Abstract: In this paper, we propose a new systemic risk indicator to measure the distance to the extreme losses of a financial system. Constructed from daily out-of-sample Value-at-Risk (VaR) exceptions across 95 large U.S. financial institutions from 2000 to 2023, our indicator calculates the shortfall in market value during these exceptions. By applying extreme value theory (EVT) to the maximum weekly shortfalls using a half-year rolling window, we effectively model the tail risk of the financial system. Our empirical analysis demonstrates that this indicator captures accurately significant financial crises, such as the Great Financial Crisis of 2008, the sovereign debt crisis of 2010, and the COVID-19 pandemic in 2020. Through quantile regression, we show that increases in our indicator significantly predict negative shocks to industrial production growth rates.

Managing a Lazy Investment: Being Actively Passive (with Jérémy Dudek and Indigo Jones), December 2024

Abstract: While passively managed financial instruments do not require intervention from their investors, some investors actively manage them anyway. To understand why and how, we use a novel micro-level dataset of 6,247 robo-advisor clients who made 9,250 changes to their investment portfolios between 2015 and 2022. Micro-level demographic and financial variables as well as macro-level market returns and volatility are factors in the decision to change one's passively managed portfolio. In addition, how these changes affected investors' returns are studied. A counterfactual test showed that on average accounts which adjusted their portfolio allocation outperformed identical hypothetical accounts in which no changes were made, but this result was not replicated in the field using a more constrained dataset including only realized gains (i.e., closed accounts).

Shortfall in Tax Revenue: Evaluating the Social Security Contribution Fraud in France  (with Denisa Banulescu-Radu and Christophe Hurlin), December 2024

Abstract: Social contribution fraud poses significant challenges with substantial economic implications, including reduced public revenue, increased inequality, and potential distortions in market competition. This study introduces a framework to estimate undetected fraud, defined as the potential tax adjustments that could have been imposed on firms with fraudulent or erroneous declarations if they had been inspected. Using an econometric model, we formalize the concept of tax revenue shortfall and derive a parametric estimator, validated through Monte Carlo simulations. Applying this method to a unique dataset from the Mutualité Sociale Agricole (MSA), which oversees the French agricultural social system, we quantify the associated fraud on social security contributions. Our findings indicate that undetected fraud among uncontrolled firms represents 3% to 7% of total collected social contributions, underscoring the need for improved inspection processes.

Bail-in Vs. Bailout: A Persuasion Game (with Maroua Riabi), January 2024. [Web Application]

Abstract: We propose a model with incomplete information where a distressed bank asks its creditor, a healthy bank, to reduce its debt. Given the information disclosed by the regulator about the asset quality of the distressed bank and its possible bailout by the government, the healthy bank can accept or not the bail-in operation. The role of the regulator is to select the optimal disclosure rule that reduces its expected loss function. We find that the full disclosure is desirable in some circumstances, especially in extreme periods, but not in others. For instance, when the bail-in cost is large, the optimal loss is reached thanks to a partial disclosure in normal times. In contrast, when the bailout cost is high, no disclosure minimizes the regulator's expected welfare losses in normal times.

Investing Through Economic Cycles with Ensemble Machine Learning Algorithms (with Thomas Raffinot), December 2018. 

Abstract: Ensemble machine learning algorithms (random forest and boosting) are applied to quickly and accurately detect economic turning points in the United States and in the Eurozone over the past three decades. The two key features of those algorithms are their abilities (i) to entertain a large number of predictors and (ii) to perform both variable selection and estimation simultaneously. The real-time ability to nowcast economic turning points is gauged by using investment strategies based on economic regimes induced by our models. When comparing predictive accuracy and profit measures, the model confidence set procedure is applied to avoid data snooping. We show that such investment strategies achieve impressive risk-adjusted returns: timing the market is thus possible.

A Theoretical and Empirical Comparison of Systemic Risk Measures (with Gilbert Colletaz, Christophe Hurlin and Christophe Pérignon), September 2014. [Companion Website]

Abstract: We derive several popular systemic risk measures in a common framework and show that they can be expressed as transformations of market risk measures (e.g. beta). We also derive conditions under which the different measures lead to similar rankings of systemically important financial institutions (SIFIs). In an empirical analysis of US financial institutions, we show that (1) different systemic risk measures identify different SIFIs and that (2) firm rankings based on systemic risk estimates mirror rankings obtained by sorting firms on market risk or liabilities. One-factor linear models explain most of the variability of the systemic risk estimates, which indicates that systemic risk measures fall short in capturing the multiple facets of systemic risk.

Work in Progress

Elicitability of Market-Based Systemic-Risk Measures (with O. Couperier, J. Leymarie and O. Scaillet), October 2022