nep-rmg New Economics Papers
on Risk Management
Issue of 2024‒03‒04
seventeen papers chosen by



  1. Provisions and Economic Capital for Credit Losses By Dorinel Bastide; St\'ephane Cr\'epey
  2. Frequency Volatility Connectedness and Portfolio Hedging of U.S. Energy Commodities By Evžen Kočenda; Michala Moravcová; Evžen Kocenda
  3. Improving Business Insurance Loss Models by Leveraging InsurTech Innovation By Zhiyu Quan; Changyue Hu; Panyi Dong; Emiliano A. Valdez
  4. Quantitative technologies and reflexivity: The role of tools and their layouts in the case of credit risk management By Céline Baud; Nathalie Lallemand-Stempak
  5. Credit Risk Meets Large Language Models: Building a Risk Indicator from Loan Descriptions in P2P Lending By Mario Sanz-Guerrero; Javier Arroyo
  6. Portfolio Selection Under Non-Gaussianity And Systemic Risk: A Machine Learning Based Forecasting Approach. By Weidong Lin; Abderrahim Taamouti
  7. Conclusions of the Fourth European Conference on Risk Perception, Behaviour, Management and Response - ECRP 2023 By Samuel Rufat; Iuliana Armas; Victor Santoni; Cosmina Albulescu; Karsten Uhing; Mariana de Brito; Paul Hudson
  8. The Non-linear Impact of Risk Tolerance on Entrepreneurial Profit and Business Survival By Melanie Koch; Lukas Menkhoff
  9. Balancing the Risk of Tipping: Early Warning Systems from Detection to Management By Florian Diekert; Daniel Heyen; Frikk Nesje; Soheil Shayegh
  10. Poverty is associated with both risk avoidance and risk taking: an empirical test of the desperation threshold model. By de Courson, Benoît; Frankenhuis, Willem; Nettle, Daniel
  11. Tail Copula Estimation for Heteroscedastic Extremes By Einmahl, John; Zhou, C.
  12. Second-Order Representations: A Bayesian Approach By Ozgur Evren
  13. A Cost Assessment of Tree Plantation Failure under Extreme Drought Events in France: What Role for Insurance? By Sandrine Brèteau-Amores; Marielle Brunette; Pablo Andrés-Domenech
  14. An Econometric Analysis of Volatility Discovery By Gustavo Fruet Dias; Fotis Papailias; Cristina Scherrer
  15. On the protective effects of European sustainable stocks during the Russian invasion of Ukraine By Kick, Andreas; Rottmann, Horst
  16. Putting eggs in one basket: insights from a correlation inequality By Pradeep Dubey; Siddhartha Sahi; Guanyang Wang
  17. Interplay between Cryptocurrency Transactions and Online Financial Forums By Ana Fern\'andez Vilas; Rebeca P. D\'iaz Redondo; Daniel Couto Cancela; Alejandro Torrado Pazos

  1. By: Dorinel Bastide (LaMME); St\'ephane Cr\'epey (LPSM)
    Abstract: Based on supermodularity ordering properties, we show that convex risk measures of credit losses are nondecreasing w.r.t. credit-credit and, in a wrong-way risk setup, credit-market, covariances of elliptically distributed latent factors. These results support the use of such setups for computing credit provisions and economic capital or for conducting stress test exercises and risk management analysis.
    Date: 2024–01
    URL: http://d.repec.org/n?u=RePEc:arx:papers:2401.07728&r=rmg
  2. By: Evžen Kočenda; Michala Moravcová; Evžen Kocenda
    Abstract: We analyze (frequency) connectedness and portfolio hedging among U.S. energy commodities from 1997 to 2023. We show that the total connectedness increased over time, likely due to the increasing financialization of energy commodities. It fluctuates with respect to (i) different investment horizons and (ii) different periods of distress. The early stage of the Russia-Ukraine war is associated with the highest systemic risk, followed by the Covid-19 pandemic and global financial crisis (GFC). In the frequency domain, the results imply that investors perceive the greatest risk at longer investment horizons, particularly during the three major distress periods. We also show that despite it is difficult and more costly to diversify an energy portfolio during distress periods, adding natural gas seems to bring non-marginal diversification benefits.
    Keywords: connectedness, volatility spillovers, frequency decomposition, portfolio weights and hedge ratios, energy commodities, distress
    JEL: C58 F65 G15 Q34 Q41
    Date: 2024
    URL: http://d.repec.org/n?u=RePEc:ces:ceswps:_10889&r=rmg
  3. By: Zhiyu Quan; Changyue Hu; Panyi Dong; Emiliano A. Valdez
    Abstract: Recent transformative and disruptive advancements in the insurance industry have embraced various InsurTech innovations. In particular, with the rapid progress in data science and computational capabilities, InsurTech is able to integrate a multitude of emerging data sources, shedding light on opportunities to enhance risk classification and claims management. This paper presents a groundbreaking effort as we combine real-life proprietary insurance claims information together with InsurTech data to enhance the loss model, a fundamental component of insurance companies' risk management. Our study further utilizes various machine learning techniques to quantify the predictive improvement of the InsurTech-enhanced loss model over that of the insurance in-house. The quantification process provides a deeper understanding of the value of the InsurTech innovation and advocates potential risk factors that are unexplored in traditional insurance loss modeling. This study represents a successful undertaking of an academic-industry collaboration, suggesting an inspiring path for future partnerships between industry and academic institutions.
    Date: 2024–01
    URL: http://d.repec.org/n?u=RePEc:arx:papers:2401.16723&r=rmg
  4. By: Céline Baud; Nathalie Lallemand-Stempak (GREGOR - Groupe de Recherche en Gestion des Organisations - UP1 - Université Paris 1 Panthéon-Sorbonne - IAE Paris - Sorbonne Business School)
    Abstract: The development of quantitative technologies is increasingly challenging professional practices and raises questions about whether and how organizations may foster plural and reflexive practices. In this paper, we outline the role played by tools and their layouts in this process. Tools can sustain the enactment of plural views, logics and evaluative principles. However, it is not clear why, in some cases, designing or using these tools triggers intractable conflicts instead of helping to sustain reflexivity in a "productive" way. To address this issue, we explore the case of a French bank that introduced in its credit management processes a new statistical approach of risk management, which conflicted with the professional approach that prevailed at the time. Relying on Boltanski's (2011) work on critique, we highlight how "productive" reflexivity emerges, not only from critique, but from a dynamic relationship between critique, confirmation and practical action. This framework allows us to bring a fresh look at the layouts identified in the literature as able to sustain pluralism by exposing their differences regarding whether and how they may contribute to trigger reflexivity. We especially show that, when quantitative technologies are involved, the creation of compromising accounts may prompt dynamics of escalating conflict, while combinations may help organising a pluralism of modes of evaluation that nurtures reflexivity without inhibiting action. Moreover, our study shows how, in credit risk management, quantitative technologies can be implemented, even in the most operational processes, without bringing about an unreflexive "illusion" of control.
    Keywords: Quantitative technologies, Reflexivity, Risk management, Tools, Compromises, Combinations
    Date: 2024–06
    URL: http://d.repec.org/n?u=RePEc:hal:journl:hal-04419872&r=rmg
  5. By: Mario Sanz-Guerrero; Javier Arroyo
    Abstract: Peer-to-peer (P2P) lending has emerged as a distinctive financing mechanism, linking borrowers with lenders through online platforms. However, P2P lending faces the challenge of information asymmetry, as lenders often lack sufficient data to assess the creditworthiness of borrowers. This paper proposes a novel approach to address this issue by leveraging the textual descriptions provided by borrowers during the loan application process. Our methodology involves processing these textual descriptions using a Large Language Model (LLM), a powerful tool capable of discerning patterns and semantics within the text. Transfer learning is applied to adapt the LLM to the specific task at hand. Our results derived from the analysis of the Lending Club dataset show that the risk score generated by BERT, a widely used LLM, significantly improves the performance of credit risk classifiers. However, the inherent opacity of LLM-based systems, coupled with uncertainties about potential biases, underscores critical considerations for regulatory frameworks and engenders trust-related concerns among end-users, opening new avenues for future research in the dynamic landscape of P2P lending and artificial intelligence.
    Date: 2024–01
    URL: http://d.repec.org/n?u=RePEc:arx:papers:2401.16458&r=rmg
  6. By: Weidong Lin; Abderrahim Taamouti
    Abstract: The Sharpe-ratio-maximizing portfolio becomes questionable under non-Gaussian returns, and it rules out, by construction, systemic risk, which can negatively affect its out-of-sample performance. In the present work, we develop a new performance ratio that simultaneously addresses these two problems when building optimal portfolios. To robustify the portfolio optimization and better represent extreme market scenarios, we simulate a large number of returns via a Monte Carlo method. This is done by Örst obtaining probabilistic return forecasts through a distributional machine learning approach in a big data setting, and then combining them with a Ötted copula to generate return scenarios. Based on a large-scale comparative analysis conducted on the US market, the backtesting results demonstrate the superiority of our proposed portfolio selection approach against several popular benchmark strategies in terms of both proÖtability and minimizing systemic risk. This outperformance is robust to the inclusion of transaction costs.
    Keywords: Portfolio optimization; probability forecasting; quantile regression neural network; extreme scenarios; big data.
    Date: 2023–08
    URL: http://d.repec.org/n?u=RePEc:liv:livedp:202310&r=rmg
  7. By: Samuel Rufat (CY - CY Cergy Paris Université, PLACES - EA 4113 - PLACES - Laboratoire de géographie et d'aménagement - CY - CY Cergy Paris Université); Iuliana Armas (UniBuc - University of Bucharest); Victor Santoni (MRTE - EA 4112 - Laboratoire Mobilités, Réseaux, Territoires, Environnements - CY - CY Cergy Paris Université); Cosmina Albulescu (UniBuc - University of Bucharest); Karsten Uhing (Fraunhofer IML - Fraunhofer Institute for Material Flow and Logistics - Fraunhofer-Gesellschaft - Fraunhofer); Mariana de Brito (UFZ - Helmholtz Zentrum für Umweltforschung = Helmholtz Centre for Environmental Research); Paul Hudson (University of York [York, UK])
    Abstract: The Fourth ECRP conference in June 2023 in Bucharest, Romania, has gathered again our two communities, the Risk Perception and Behaviour Survey of Surveyors (Risk-SoS) and the H2020-DRS-01 Cluster on risk perception and adaptive behaviour (a grouping of several Horizon Europe projects). The ECRP conference cycle aims to contribute to improve the ability of researchers in the field to work together and build cumulative knowledge, fostering scientific communication and collaborative learning, ultimately leading to joint research publications and projects. This cycle emerged in response to the challenges posed by the current fragmentation of the studies of risk perceptions and how perceptions influence behaviour. It remains unclear why people fail to act adaptively to reduce future losses, even when there is ever richer information available on a wide range of hazards (flood, drought, earthquakes, etc). The current collection of seemingly independent case studies hinders comparability and transferability across scales and contexts and hampers recommendations for policy and risk management. Another challenge derives from the lack of a robust theoretical base and the apparent path dependency of design choices routinely based on previous research, consolidating the predominance of socio-psychological theories and methodological individualism, which are often non-contextual. A greater diversity of theoretical frameworks could lead to increased attention to socioecological processes and the socio-cultural context of risk, which might be critical for case studies cross-validation.
    Keywords: Mitigation, Disaster, Response, Disaster management, Hazard, Risk perception, Behaviour, Adaptation, Risk
    Date: 2023
    URL: http://d.repec.org/n?u=RePEc:hal:journl:hal-04401500&r=rmg
  8. By: Melanie Koch; Lukas Menkhoff
    Abstract: Entrepreneurs tend to be risk tolerant but is more risk tolerance always better? In a sample of about 2, 100 small businesses, we find an inverted U-shaped relation between risk tolerance and profitability. This relationship holds in a simple bilateral regression and also when we control for a large set of individual and business characteristics. Apparently, one major transmission goes from risk tolerance via investments to profits. This is quite robust as it applies for past investments as well as planned investments. Considering business survival, we show, first, that less profitable businesses leave the market while moderately risk tolerant entrepreneurs survive more often. Second, the high risk-low profit part of the U-shaped relation seems to disappear among businesses being four years and older, indicating that such inferior risk-profit combinations disappear over time. These findings are important for the concept of business readiness trainings as the motivation (and ability) to take risks should potentially be accompanied by some warning that too much risk taking can be detrimental to long-term business success.
    Keywords: risk tolerance; entrepreneurs; profits; investments
    JEL: D22 D81 L26 M21
    Date: 2024
    URL: http://d.repec.org/n?u=RePEc:diw:diwwpp:dp2067&r=rmg
  9. By: Florian Diekert; Daniel Heyen; Frikk Nesje; Soheil Shayegh
    Abstract: Early warning signals (EWS) of imminent regime shifts can be identified through the observation of a system’s behavior under increasing stress and before crossing a tipping point. Despite many advances in the detection of EWS in recent years, EWS are yet to find direct application in management. Here, we focus on operationalizing the EWS information in an early warning system consisting of a tipping indicator (e.g., autocorrelation), whose value increases as the system approaches the tipping point, and a trigger value, beyond which an EWS is sent. We demonstrate how such an early warning system allows managers to balance the risk of tipping by providing information for updating their belief about the location of the tipping point. In particular, deployment of an early warning system results in taking more cautious early steps while it encourages more risk taking behavior in later stages if no EWS is sent. We uncover a tension between better information about the location of the tipping point and increased risk of crossing it as a result of EWS. Our framework complements the emerging EWS knowledge in the natural sciences with a better understanding of how, when, and why EWS improve management.
    Keywords: catastrophic regime shifts, tipping points, early warning signals, learning, optimal ecosystem management
    JEL: C61 D83 Q54
    Date: 2024
    URL: http://d.repec.org/n?u=RePEc:ces:ceswps:_10892&r=rmg
  10. By: de Courson, Benoît; Frankenhuis, Willem; Nettle, Daniel (Centre Nationale de la Recherche Scientifique)
    Abstract: In situations of poverty, do people take more or less risk? Some theories state that poverty makes people 'vulnerable': they cannot buffer against losses, and therefore avoid risk. Yet, other theories state the opposite: poverty makes people 'desperate': they have little left to lose, and therefore take risks. Each theory has some support: most studies find a negative association between resources and risk taking, but risky behaviors such as crime are more common in deprived populations. Here, we test the 'desperation threshold' model, which integrates both hypotheses. The model assumes that people attempt to stay above a critical level of resources, representing their 'basic needs'. Just above the threshold, people have too much to lose, and should avoid risk. Below it, they have little to lose, and should take risks. We conducted preregistered tests of this prediction using longitudinal data of 472 adults over the age of 25 in France and the UK, who completed a survey once a month for 12 months. We examined whether risk taking first increased and then decreased as a function of objective and subjective financial resources. Results supported this prediction for subjective resources, but not for objective resources. Next, we tested whether risk taking varies more among people who have fewer resources. We find strong evidence for both more extreme risk avoidance and more extreme risk taking in this group. We rule out alternative explanations related to question comprehension and measurement error, and discuss implications of our findings for welfare states, poverty, and crime.
    Date: 2024–02–09
    URL: http://d.repec.org/n?u=RePEc:osf:socarx:gqjkm&r=rmg
  11. By: Einmahl, John (Tilburg University, Center For Economic Research); Zhou, C. (Tilburg University, Center For Economic Research)
    Keywords: Extreme value statistics; functional limit theorems; non-identical distributions; tail empirical process; tail dependence
    Date: 2024
    URL: http://d.repec.org/n?u=RePEc:tiu:tiucen:6bcb09c5-8b19-48b8-9320-b80e0d9db36b&r=rmg
  12. By: Ozgur Evren (New Economic School)
    Abstract: For choice problems under ambiguity, I provide a behavioral characterization of a decision maker who holds a second-order belief and updates it in a Bayesian fashion in response to new information concerning the true distribution of the states. The model features a unique second-order belief that can be elicited from choice data and is quite comprehensive in terms of ambiguity attitudes and risk preferences. Special versions, such as the smooth ambiguity model or the recursive non-expected utilitymodel, are easily characterized by additional assumptions on compound-risk preferences. Thereby, the model provides a testing ground to compare and contrast these well-known representations as well as alternative specifications that may be of interest. To illustrate potential benefits of alternative specifications, I provide a detailed analysis of a rank-dependent extension of the smooth ambiguity model.
    Keywords: Ambiguity Aversion and Seeking, Ellsberg Paradox, Second-Order Belief, Probabilistic Sophistication, Bayesian Updating, Compound Risk JEL Classifications: D81, D83
    Date: 2024–01
    URL: http://d.repec.org/n?u=RePEc:abo:neswpt:w0291&r=rmg
  13. By: Sandrine Brèteau-Amores (BETA - Bureau d'Économie Théorique et Appliquée - AgroParisTech - UNISTRA - Université de Strasbourg - Université de Haute-Alsace (UHA) - Université de Haute-Alsace (UHA) Mulhouse - Colmar - UL - Université de Lorraine - CNRS - Centre National de la Recherche Scientifique - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement); Marielle Brunette (BETA - Bureau d'Économie Théorique et Appliquée - AgroParisTech - UNISTRA - Université de Strasbourg - Université de Haute-Alsace (UHA) - Université de Haute-Alsace (UHA) Mulhouse - Colmar - UL - Université de Lorraine - CNRS - Centre National de la Recherche Scientifique - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement); Pablo Andrés-Domenech (BETA - Bureau d'Économie Théorique et Appliquée - AgroParisTech - UNISTRA - Université de Strasbourg - Université de Haute-Alsace (UHA) - Université de Haute-Alsace (UHA) Mulhouse - Colmar - UL - Université de Lorraine - CNRS - Centre National de la Recherche Scientifique - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement)
    Abstract: Research Highlights: We analyze the costs of plantation failure and evaluate the distribution of replantation costs and risk sharing between the forestry company and the forest owner in France. Background and Objectives: Due to the lack of a clear definition of drought, forestry companies are increasingly considered as liable for plantation failure, increasing their costs and leading to financial instability. In this context, this paper aims to address the following questions. In the case of plantation failure, is it less costly to replant, not replant, or restart the whole plantation? What is the impact of changing the liability scheme between the company and the forest owner in terms of replantation costs and risk sharing? Materials and Methods: We performed a cost assessment of different itineraries of plantations as a function of different mortality rates. The breakdown of the replantation costs between the company and the forest owner was also investigated. Results: No replanting is the least expensive option for the forest owner, followed by replanting and then by starting the whole plantation anew. Reducing the company's liability is an interesting option to reduce its exposure to risk. Conclusions: Modifications of the company's liability allows for the inclusion of private insurance contracts against plantation failure.
    Keywords: Forest, Regeneration, Plantation, Drought, Insurance, Costs
    Date: 2023–02
    URL: http://d.repec.org/n?u=RePEc:hal:journl:hal-03998594&r=rmg
  14. By: Gustavo Fruet Dias (School of Economics, University of East Anglia); Fotis Papailias (King’s Business School, King’s College London); Cristina Scherrer (Department of Finance, London School of Economics)
    Abstract: We investigate information processing in the stochastic process driving stock’s volatility (volatility discovery). We apply fractionally cointegration techniques to de-compose the estimates of the market-speciï¬ c integrated variances into an estimate of the common integrated variance of the efficient price and a transitory component. The market weights on the common integrated variance of the efficient price are the volatility discovery measures. We relate the volatility discovery measure to the price discovery framework and formally show their roles on the identiï¬ cation of the inte-grated variance of the efficient price. We establish the limiting distribution of the volatility discovery measures by resorting to both long span and in-ï¬ ll asymptotics. The empirical application is in line with our theoretical results, as it reveals that trading venues incorporate new information into the stochastic volatility process in an individual manner and that the volatility discovery analysis identiï¬ es a distinct information process than that based on the price discovery analysis.
    Keywords: long memory, fractionally cointegrated vector autoregressive model, realized measures, market microstructure, price discovery, high-frequency data, double asymptotics
    Date: 2023–12
    URL: http://d.repec.org/n?u=RePEc:uea:ueaeco:2024-01&r=rmg
  15. By: Kick, Andreas; Rottmann, Horst
    Abstract: Sustainable investments remain popular, attracting investors and researchers alike. Especially the tail-risk properties seem to differ between sustainable stocks and common stocks. Empirically, this can be observed in particular during extreme events. On February 24, 2022 Russian forces invaded Ukraine, thereby marking the beginning of a major historical event. Using standard event study methodology, we analyze if and how Refinitiv's environmental, social, and governance (ESG) ratings, as well as carbon dioxide (CO2) intensity, influence cumulative abnormal returns during different event windows. We find that the abnormal returns of companies with high ecological scores exhibit a protective effect in the pre- and post-event windows. However, this effect did not materialize in all observed event windows. Therefore, our results do not fully support the hypothesis of an 'ESG hedge' against such extreme events.
    Keywords: abnormal returns, war, Ukraine, ESG, Russia
    JEL: G11 G14 M14
    Date: 2024
    URL: http://d.repec.org/n?u=RePEc:zbw:hawdps:283004&r=rmg
  16. By: Pradeep Dubey; Siddhartha Sahi; Guanyang Wang
    Abstract: We give examples of situations – stochastic production, military tactics, corporate merger – where it is beneficial to concentrate risk rather than to diversify it, i.e., to put all eggs in one basket. The examples admit a dual interpretation: as optimal strategies of a single player (the “principal†) or, alternatively, as dominant strategies in a non-cooperative game with multiple players (the “agents†). The key mathematical result can be formulated in terms of a convolution structure on the set of increasing functions on a Boolean lattice (the lattice of subsets of a finite set). This result generalizes the well-known Harris inequality from statistical physics and discrete mathematics; we give a simple self-contained proof of this result, and then prove a further generalization based on game-theoretic ideas.
    Date: 2024
    URL: http://d.repec.org/n?u=RePEc:nys:sunysb:24-02&r=rmg
  17. By: Ana Fern\'andez Vilas; Rebeca P. D\'iaz Redondo; Daniel Couto Cancela; Alejandro Torrado Pazos
    Abstract: Cryptocurrencies are a type of digital money meant to provide security and anonymity while using cryptography techniques. Although cryptocurrencies represent a breakthrough and provide some important benefits, their usage poses some risks that are a result of the lack of supervising institutions and transparency. Because disinformation and volatility is discouraging for personal investors, cryptocurrencies emerged hand-in-hand with the proliferation of online users' communities and forums as places to share information that can alleviate users' mistrust. This research focuses on the study of the interplay between these cryptocurrency forums and fluctuations in cryptocurrency values. In particular, the most popular cryptocurrency Bitcoin (BTC) and a related active discussion community, Bitcointalk, are analyzed. This study shows that the activity of Bitcointalk forum keeps a direct relationship with the trend in the values of BTC, therefore analysis of this interaction would be a perfect base to support personal investments in a non-regulated market and, to confirm whether cryptocurrency forums show evidences to detect abnormal behaviors in BTC values as well as to predict or estimate these values. The experiment highlights that forum data can explain specific events in the financial field. It also underlines the relevance of quotes (regular mechanism to response a post) at periods: (1) when there is a high concentration of posts around certain topics; (2) when peaks in the BTC price are observed; and, (3) when the BTC price gradually shifts downwards and users intend to sell.
    Date: 2023–11
    URL: http://d.repec.org/n?u=RePEc:arx:papers:2401.10238&r=rmg

General information on the NEP project can be found at https://nep.repec.org. For comments please write to the director of NEP, Marco Novarese at <director@nep.repec.org>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.