nep-rmg New Economics Papers
on Risk Management
Issue of 2023‒01‒16
twenty-six papers chosen by
Stan Miles
Thompson Rivers University

  1. Optimal Systemic Risk Bailout: A PGO Approach Based on Neural Network By Shuhua Xiao; Jiali Ma; Li Xia; Shushang Zhu
  2. Fire Sales and Ex Ante Valuation of Systemic Risk: A Financial Equilibrium Networks Approach By Spiros Bougheas; Adam Hal Spencer
  3. Financing Constraints and Risk Management: Evidence From Micro-Level Insurance Data By Bustos, Emil; Engist, Oliver; Martinsson, Gustav; Thomann, Christian
  4. Modelling spatial correlation between earthquake insured losses in New Zealand: a mixed-effects analysis By F. Marta L. Di Lascio; Ilan Noy; Selene Perazzini
  5. Measuring adequately the benefit of diversification in the extreme quantiles: An inquiry into covariation on the brink of catastrophe By Pierre-Charles Pradier; Guillaume Rideau; Sakina Rrguiti
  6. Benchmarking Machine Learning Models to Predict Corporate Bankruptcy By Emmanuel Alanis; Sudheer Chava; Agam Shah
  7. Saddle-Point Approach to Large-Time Volatility Smile By Chun Yat Yeung; Ali Hirsa
  8. Limiting sequential decompositions and applications in finance By Gero Junike; Hauke Stier; Marcus C. Christiansen
  9. Hedging the Risks of MENA Stock Markets with Gold: Evidence from the Spectral Approach By Awatef Ourir; Elie Bouri; Essahbi Essaadi
  10. Toxic Liquidation Spirals : Evidence from the bad debt incurred by AAVE By Jakub Warmuz; Amit Chaudhary; Daniele Pinna
  11. Financial Connectedness and Risk Transmission Among MENA Countries: Evidence from Connectedness Network and Clustering Analysis By Mehmet Balcilar; Shawkat Hammoudeh
  12. Deep Quadratic Hedging By Alessandro Gnoatto; Silvia Lavagnini; Athena Picarelli
  13. Beyond Surrogate Modeling: Learning the Local Volatility Via Shape Constraints By Marc Chataigner; Areski Cousin; St\'ephane Cr\'epey; Matthew Dixon; Djibril Gueye
  14. Financial integration and international risk spillovers By Dongwon Lee
  15. Decentralized lending and its users: Insights from Compound By Kanis Saengchote
  16. On Efficient and Accurate Calibration to FX Market Skew by a Fully Parameterized Local Volatility Model By Dongli Wu; Bufan Zhang; Xiao Lin
  17. The Strategic Allocation to Style-Integrated Portfolios of Commodity Futures By Hossein Rad; Rand Kwong Yew Low; Joelle Miffre; Robert Faff
  18. The global financial cycle and macroeconomic tail risks By Beutel, Johannes; Emter, Lorenz; Metiu, Norbert; Prieto, Esteban; Schüler, Yves
  19. Physical and transition risk premiums in euro area corporate bond markets By Joost Bats; Giovanna Bua; Daniel Kapp
  20. Superstar Returns By Francisco Amaral; Martin Dohmen; Sebastian Kohl; Moritz Schularick
  21. Jacobi Stochastic Volatility factor for the Libor Market Model By Pierre-Edouard Arrouy; Bernard Lapeyre; Sophian Mehalla; Alexandre Boumezoued
  22. Sovereign Default and Liquidity: The Case for a World Safe Asset By François Le Grand; Xavier Ragot
  23. Optimal investment under partial observations and robust VaR-type constraint By Nicole B\"auerle; An Chen
  24. Consumer Bankrupcty, Mortgage Default and Labor Supply By Wenli Li; Costas Meghir; Florian Oswald
  25. Fallen Angel Bonds Investment and Bankruptcy Predictions Using Manual Models and Automated Machine Learning By Harrison Mateika; Juannan Jia; Linda Lillard; Noah Cronbaugh; Will Shin
  26. Consumer credit in the age of AI: Beyond anti-discrimination law By Langenbucher, Katja

  1. By: Shuhua Xiao; Jiali Ma; Li Xia; Shushang Zhu
    Abstract: The bailout strategy is crucial to cushion the massive loss caused by systemic risk in the financial system. There is no closed-form formulation of the optimal bailout problem, making solving it difficult. In this paper, we regard the issue of the optimal bailout (capital injection) as a black-box optimization problem, where the black box is characterized as a fixed-point system that follows the E-N framework for measuring the systemic risk of the financial system. We propose the so-called ``Prediction-Gradient-Optimization'' (PGO) framework to solve it, where the ``Prediction'' means that the objective function without a closed-form is approximated and predicted by a neural network, the ``Gradient'' is calculated based on the former approximation, and the ``Optimization'' procedure is further implemented within a gradient projection algorithm to solve the problem. Comprehensive numerical simulations demonstrate that the proposed approach is promising for systemic risk management.
    Date: 2022–12
  2. By: Spiros Bougheas; Adam Hal Spencer
    Abstract: We introduce endogenous fire sales into a simple network model. For any given initial distribution of shocks across the network, we develop a clearing algorithm to solve for the financial equilibrium. We then utilise the results to perform ex ante risk assessment and derive risk premia for every balance sheet item where liabilities are differentiated according to priority rights. We find that risk premia reflect both idiosyncratic risk and risk of contagion (network risk). Moreover, we show that network risk magnifies the gap between the risk premia of equity and debt. We also perform comparative statics, showing that changes to the distribution of shocks and network structure can have substantial effects on the level of systemic losses.
    Keywords: networks, fire sales, systemic risk premia, risk assessment
    JEL: G33 G32 D85
    Date: 2022
  3. By: Bustos, Emil (Research Institute of Industrial Economics (IFN)); Engist, Oliver (Stockholm School of Economics); Martinsson, Gustav (Royal Institute of Technology); Thomann, Christian (Royal Institute of Technology)
    Abstract: We study the impact of financing constraints on corporate risk management. Using data on credit scores matched with unique information on firm level commercial insurance purchases, we find that financing constraints lead to higher insurance spending. We adopt a regression discontinuity design and show that financially constrained firms spend 5–14% more on insurance than otherwise similar unconstrained firms. Our findings add new insights to the longstanding empirical puzzle of whether financially constrained firms engage more in risk management. Furthermore, our results shed light on risk management in smaller, mostly private firms.
    Keywords: Financing Constraints; Risk Management; Insurance Demand; Credit Scores
    JEL: D22 D25 G22 G32
    Date: 2022–12–20
  4. By: F. Marta L. Di Lascio (Free University of Bozen-Bolzano, Italy); Ilan Noy (School of Economics and Finance, Victoria University of Wellington, New Zealand); Selene Perazzini (DMS StatLab, Department of Economics and Management, University of Brescia, Italy)
    Abstract: Earthquake insurance is a critical risk management strategy that contributes to improving recovery and thus greater resilience of individuals. Insurance companies construct premiums without taking into account spatial correlations between insured assets. This leads to potentially underestimating the risk, and therefore the exceedance probability curve. We here propose a mixed-effects model to estimate losses per ward that is able to account for heteroscedasticity and spatial correlation between insured losses. Given the significant impact of earthquakes in New Zealand due to its particular geographical and demographic characteristics, the government has established a public insurance company that collects information about the insured buildings and any claims lodged. We thus develop a two-level variance component model that is based on earthquake losses observed in New Zealand between 2000 and 2021. The proposed model aims at capturing the variability at both the ward and territorial authority levels and includes independent variables, such as seismic hazard indicators, the number of usual residents, and the average dwelling value in the ward. Our model is able to detect spatial correlation in the losses at the ward level thus increasing its predictive power and making it possible to assess the effect of spatially correlated claims that may be considerable on the tail of loss distribution.
    Keywords: Earthquake losses, Insurance, Mixed-effects model, Spatial correlation, Variance component model.
    JEL: C10 C21
    Date: 2022–12
  5. By: Pierre-Charles Pradier; Guillaume Rideau (BPCE - BPCE); Sakina Rrguiti (UP1 - Université Paris 1 Panthéon-Sorbonne, CES - Centre d'économie de la Sorbonne - UP1 - Université Paris 1 Panthéon-Sorbonne - CNRS - Centre National de la Recherche Scientifique, BPCE - BPCE)
    Abstract: The aim of this work is to better understand the nature of covariation in the vicinity of extremes on financial data and assess whether the usual assumptions and covariation measures fits the actual data. For simplicity, we consider pairs of random variables. In order to identify the shape of the covariation all along the distribution, and particularly as the extreme quantiles are approached, we describe the contribution of each of the variables from a random couple to the quantiles of the weighted sum of these variables. This approach makes sense since it can be interpreted in terms of Value-at-Risk in a financial institution: the VaR of the sum of variables may represent the capital requiremet for a diversified conglomerate, while the sum of VaR of the variables would correspond to the capital requirements for the components of the conglomerate, without taking diversification into account. The ratio of these two quantities appears as a good measure of both the benefit of diversification and the decorrelation of variables. We thus compare the values of quantiles and ratio taken from a representative dataset to the values obtained from various simulations relying on the usual assumptions. The result of this comparison is that the usual assumptions do not correctly model the covariation of the real-word data. In particular, the usual assumptions tend to exaggerate the correlation in the vicinity of extreme loss while the benefit of diversification is uniform across distribution. Additional simulations and modelling assumptions may be required to assess the generality of this result.
    Keywords: Financial conglomerates, Diversification, Value-at-Risk, Capital requirement
    Date: 2022–11
  6. By: Emmanuel Alanis; Sudheer Chava; Agam Shah
    Abstract: Using a comprehensive sample of 2, 585 bankruptcies from 1990 to 2019, we benchmark the performance of various machine learning models in predicting financial distress of publicly traded U.S. firms. We find that gradient boosted trees outperform other models in one-year-ahead forecasts. Variable permutation tests show that excess stock returns, idiosyncratic risk, and relative size are the more important variables for predictions. Textual features derived from corporate filings do not improve performance materially. In a credit competition model that accounts for the asymmetric cost of default misclassification, the survival random forest is able to capture large dollar profits.
    Date: 2022–12
  7. By: Chun Yat Yeung; Ali Hirsa
    Abstract: We extend upon the saddle-point equation presented in [1] to derive large-time model-implied volatility smiles, providing its theoretical foundation and studying its applications in classical models. As long as characteristic function fulfills a L\'evy-type scaling behavior in large time, the approach allows us to study analytically the large-time smile behaviors under specific models, and moreover, to reach a very wide class of arbitrage-free model-inspired parametrizations, in the same manner as stochastic-volatility-inspired (SVI).
    Date: 2022–12
  8. By: Gero Junike; Hauke Stier; Marcus C. Christiansen
    Abstract: The sequential updating (SU) decomposition is a well-known technique to obtain a profit and loss (P&L) attribution, e.g. of a bond portfolio, by dividing the time horizon into n subintervals and only vary one risk factor, e.g. FX, IR, CS or calendar time, in each subinterval. We show that the SU decomposition converges for large n if the P&L attribution can be expressed by a smooth function of the risk factors. We consider the average SU decomposition, which does not depend on the order or labeling of the risk factors. Sufficient conditions are given to reduce the computational complexity significantly when calculating the average SU decomposition.
    Date: 2022–12
  9. By: Awatef Ourir (University of Jendouba); Elie Bouri (Lebanese American University); Essahbi Essaadi (University of Manouba)
    Abstract: This paper contributes to the old debate on the dynamic correlation between gold and stock markets by considering a spectral approach within the framework of portfolio hedging. Specifically, we consider eight MENA stock markets (Tunisia, Egypt, Morocco, Jordan, the United Arab Emirates, Saudi Arabia, Qatar, and Oman) and examine the optimal composition between gold and the stock market index, with a minimum portfolio risk and a high expected return. Based on the spectral approach, we propose seven portfolio structures and evaluate them through a comparison with the conventional DCC-GARCH method. The main results show that the spectral-based approach outperforms the DCC-GARCH method. In fact, the optimal gold-stock composition depends on the spectral density of each stock market index, where a stock market index with a stable spectral density requires more investments in gold than a stock market index with an unstable spectral density.
    Date: 2021–11–20
  10. By: Jakub Warmuz; Amit Chaudhary; Daniele Pinna
    Abstract: On November 22nd 2022, the lending platform AAVE v2 (on Ethereum) incurred bad debt resulting from a major liquidation event involving a single user who had borrowed close to \$40M of CRV tokens using USDC as collateral. This incident has prompted the Aave community to consider changes to its liquidation threshold, and limitations on the number of illiquid coins that can be borrowed on the platform. In this paper, we argue that the bad debt incurred by AAVE was not due to excess volatility in CRV/USDC price activity on that day, but rather a fundamental flaw in the liquidation logic which triggered a toxic liquidation spiral on the platform. We note that this flaw, which is shared by a number of major DeFi lending markets, can be easily overcome with simple changes to the incentives driving liquidations. We claim that halting all liquidations once a user's loan-to-value (LTV) ratio surpasses a certain threshold value can prevent future toxic liquidation spirals and offer substantial improvement in the bad debt that a lending market can expect to incur. Furthermore, we strongly argue that protocols should enact dynamic liquidation incentives and closing factor policies moving forward for optimal management of protocol risk.
    Date: 2022–12
  11. By: Mehmet Balcilar (Eastern Mediterranean UniversityAuthor-Name: Ahmed Elsayed; Zagazig University); Shawkat Hammoudeh (University of Economics HCMC)
    Abstract: This study examines the financial connectedness and risk transmission among MENA economies by accounting for financial connectedness in the short and long run as well dependency under extreme market conditions and network graph analysis. To this end, Composite Financial Stress Indices are constructed for 11 MENA countries. In addition, a battery of econometric models is applied including the standard spillover approach, the frequency domain method, the quantile connectedness technique, and connectedness networks analysis. Using daily data over the period from June 30, 2006 to June 30, 2021, the empirical results show a positive and strong association between financial stress co-movements and spillovers in those MENA countries, particularly during the long run and high extreme stress periods. Furthermore, the five Gulf countries are strongly financially connected among themselves than with the other countries. Contrary, to Tunisia, Saudi Arabia is the main financial stress and risk transmitter to other MENA economies whereas, the North African countries are relatively mild receivers of risk. Finally, the more open countries in terms of capital controls, particularly Kuwait, Oman, Qatar, and UAE seem to play a more central role in financial connectedness and risk spillovers
    Date: 2022–11–20
  12. By: Alessandro Gnoatto; Silvia Lavagnini; Athena Picarelli
    Abstract: We present a novel computational approach for quadratic hedging in a high-dimensional incomplete market. This covers both mean-variance hedging and local risk minimization. In the first case, the solution is linked to a system of BSDEs, one of which being a backward stochastic Riccati equation (BSRE); in the second case, the solution is related to the F\"olmer-Schweizer decomposition and is also linked to a BSDE. We apply (recursively) a deep neural network-based BSDE solver. Thanks to this approach, we solve high-dimensional quadratic hedging problems, providing the entire hedging strategies paths, which, in alternative, would require to solve high dimensional PDEs. We test our approach with a classical Heston model and with a multi-dimensional generalization of it.
    Date: 2022–12
  13. By: Marc Chataigner; Areski Cousin; St\'ephane Cr\'epey; Matthew Dixon; Djibril Gueye
    Abstract: We explore the abilities of two machine learning approaches for no-arbitrage interpolation of European vanilla option prices, which jointly yield the corresponding local volatility surface: a finite dimensional Gaussian process (GP) regression approach under no-arbitrage constraints based on prices, and a neural net (NN) approach with penalization of arbitrages based on implied volatilities. We demonstrate the performance of these approaches relative to the SSVI industry standard. The GP approach is proven arbitrage-free, whereas arbitrages are only penalized under the SSVI and NN approaches. The GP approach obtains the best out-of-sample calibration error and provides uncertainty quantification.The NN approach yields a smoother local volatility and a better backtesting performance, as its training criterion incorporates a local volatility regularization term.
    Date: 2022–12
  14. By: Dongwon Lee (Department of Economics, University of California Riverside)
    Date: 2023–01
  15. By: Kanis Saengchote
    Abstract: Permissionless blockchains offer an information environment where users can interact privately without fear of censorship. Financial services can be programmatically coded via smart contracts to automate transactions without the need for human intervention or knowing user identity. This new paradigm is known as decentralized finance (DeFi). We investigate Compound (a leading DeFi lending protocol) to show how it works in this novel information environment, who its users are, and what factors determine their participation. On-chain transaction data shows that loan durations are short (31 days on average), and many users borrow to support leveraged investment strategies (yield farming). We show that systemic risk in DeFi arises from concentration and interconnection, and how traditional risk management practices can be challenging for DeFi.
    Date: 2022–12
  16. By: Dongli Wu; Bufan Zhang; Xiao Lin
    Abstract: In this paper we have studied a fully parameterized local volatility model for pricing the American option and Asian option. This model, after implemented by a grid or Monte-Carlo numerical method, can be calibrated to the FX market skew volatilities efficiently and accurately. Thus, the model will deliver the reliable prices to the exotic options in the daily trading activities.
    Date: 2022–11
  17. By: Hossein Rad (University of Queensland [Brisbane]); Rand Kwong Yew Low (Bond University [Gold Coast]); Joelle Miffre (Audencia Business School); Robert Faff (Bond University [Gold Coast])
    Abstract: Our study lies at the intersection of the literature on the diversification benefits of commodity futures and the literature on style integration. It augments the traditional asset mix of investors with a long-short portfolio that integrates the styles that matter to the pricing of commodity futures. Treating the style-integrated portfolio of commodities as part of the strategic mix of investors is found to enhance out-of-sample performance and reduce crash risk compared to the alternatives considered thus far. The conclusion holds across traditional asset mix, portfolio allocation methods, integration strategies, and sub-periods. The diversification benefits of style integration also persist, albeit lower, in a long-only setting.
    Keywords: Commodity futures, Style integration, Strategic asset allocation, Diversification
    Date: 2022–12
  18. By: Beutel, Johannes; Emter, Lorenz; Metiu, Norbert; Prieto, Esteban; Schüler, Yves
    Abstract: We study the link between the global financial cycle and macroeconomic tail risks using quantile vector autoregressions. Contractionary shocks to financial conditions and monetary policy in the United States cause elevated downside risks to growth around the world. By tightening financial conditions globally, these shocks affect the left tail of the conditional output growth distribution more strongly than the center of the distribution. This effect is particularly pronounced for countries with less flexible exchange rate arrangements, higher foreign currency exposures, and higher levels of private sector leverage, suggesting that exchange rate policies and macroprudential policies can mitigate downside risks to growth.
    Keywords: Financial shocks,Monetary policy,Global financial cycle,Growth-at-Risk,International spillovers,Quantile VAR
    JEL: C32 E23 E32 E44 F44
    Date: 2022
  19. By: Joost Bats; Giovanna Bua; Daniel Kapp
    Abstract: We study climate risk premiums in euro area corporate bond markets. As gauges of climate risk, we distinguish between physical and transition risks using textual analysis. Our findings show that, since the Paris agreement, physical risk is significantly priced in corporate bonds with longer-term maturities. Physical risk is also priced in bonds with shorter-term maturities, but the premium is smaller and less significant. The estimated physical risk premium reflects investors demanding higher future returns on bonds that underperform during adverse physical risk shocks. Our findings also point to a sizable transition risk premium, although the transition risk estimates are insignificant.
    Keywords: Climate risk; physical risk; transition risk; corporate bonds
    JEL: G12 Q51 Q54
    Date: 2023–01
  20. By: Francisco Amaral (University of Bonn); Martin Dohmen (University of Bonn); Sebastian Kohl (Max Planck Institute for the Study of Societies - Max-Planck-Gesellschaft); Moritz Schularick (University of Bonn, ECON - Département d'économie (Sciences Po) - Sciences Po - Sciences Po - CNRS - Centre National de la Recherche Scientifique, Federal Reserve Bank of New York)
    Abstract: We study long-term returns on residential real estate in twenty-seven "superstar" cities in fifteen countries over 150 years. We find that total returns in superstar cities are close to 100 basis points lower per year than in the rest of the country. House prices tend to grow faster in the superstars, but rent returns are substantially greater outside the big agglomerations, resulting in higher long-run total returns. The excess returns outside the superstars can be rationalized as a compensation for risk, especially for higher covariance with income growth and lower liquidity. Superstar real estate is comparatively safe.
    Keywords: Housing returns, Housing risk, Superstar cities, Regional housing markets
    Date: 2021–12
  21. By: Pierre-Edouard Arrouy (Recherche et Développement, Milliman Paris - Milliman France); Bernard Lapeyre (CERMICS - Centre d'Enseignement et de Recherche en Mathématiques et Calcul Scientifique - ENPC - École des Ponts ParisTech, MATHRISK - Mathematical Risk Handling - UPEM - Université Paris-Est Marne-la-Vallée - ENPC - École des Ponts ParisTech - Inria de Paris - Inria - Institut National de Recherche en Informatique et en Automatique); Sophian Mehalla (CERMICS - Centre d'Enseignement et de Recherche en Mathématiques et Calcul Scientifique - ENPC - École des Ponts ParisTech, MATHRISK - Mathematical Risk Handling - UPEM - Université Paris-Est Marne-la-Vallée - ENPC - École des Ponts ParisTech - Inria de Paris - Inria - Institut National de Recherche en Informatique et en Automatique, Recherche et Développement, Milliman Paris - Milliman France); Alexandre Boumezoued (Recherche et Développement, Milliman Paris - Milliman France)
    Abstract: We propose a new method to efficiently price swap rates derivatives under the LIBOR Market Model with Stochastic Volatility and Displaced Diffusion (DDSVLMM). This method uses polynomial processes combined with Gram-Charlier expansion techniques. The standard pricing method for this model relies on dynamics freezing to recover an Heston-type model for which analytical formulas are available. This approach is time consuming and efficient approximations based on Gram-Charlier expansions have been recently proposed. In this article, we first discuss the fact that for a class of stochastic volatility model, including the Heston one, the classical sufficient condition ensuring the convergence of the Gram-Charlier series can not be satisfied. Then, we propose an approximating model based on Jacobi process for which we can prove the stability of the Gram-Charlier expansion. For this approximation, we have been able to prove a strong convergence toward the original model; moreover, we give an estimate of the convergence rate. We also prove a new result on the convergence of the Gram-Charlier series when the volatility factor is not bounded from below. We finally illustrate our convergence results with numerical examples.
    Keywords: Stochastic Volatility,Jacobi dynamics,Polynomial processes,Gram-Charlier expansions,LIBOR Market Model
    Date: 2022–09
  22. By: François Le Grand (EM - emlyon business school, ETH Zürich - Eidgenössische Technische Hochschule - Swiss Federal Institute of Technology [Zürich]); Xavier Ragot (ECON - Département d'économie (Sciences Po) - Sciences Po - Sciences Po - CNRS - Centre National de la Recherche Scientifique, CNRS - Centre National de la Recherche Scientifique, OFCE - Observatoire français des conjonctures économiques (Sciences Po) - Sciences Po - Sciences Po)
    Abstract: This paper presents a positive and normative study of a world financial market when sovereign countries can default on their debt. We construct a tractable model that enables us to study sovereign default in general equilibrium. The amount of safe assets is thus endogenous and determined by international risk-sharing. We characterize the equilibrium structure and we show that the market equilibrium can generate multiple equilibria. In addition, the market equilibrium is not constrained-efficient because countries do not fully internalize the value of their debt being used as liquidity. We prove that a world fund issuing a safe asset increases aggregate welfare. The fund's relationship with the IMF's Special Drawing Rights is discussed.
    Keywords: Sovereign Default,Safe Asset,International Liquidity
    Date: 2021–07
  23. By: Nicole B\"auerle; An Chen
    Abstract: The present paper extends the literature on utility maximization by combining the framework of partial information and (robust) regulatory constraints. Partial information is characterized by the fact that the stock price itself is observable to the optimizing financial institution, but the outcome of the market price of risk $\theta$ is unknown to the institution. The regulator builds the same or a different belief about the market price of risk as the financial institution. The solution to our optimization problem takes the same form as in the full information case: the optimal wealth can be expressed as a decreasing function of the state price density, and the regulatory threshold is ensured in the intermediate economic states. The main difference lies in the terminal state price density depending on the entire evolution of the estimated market price of risk $\hat{\theta}(s)$. The subjective evaluation of the regulatory constraint influences the width of the ensured region.
    Date: 2022–12
  24. By: Wenli Li (Federal Reserve Bank Philadelphia); Costas Meghir (Yale University [New Haven], CEPR - Center for Economic Policy Research - CEPR); Florian Oswald (ECON - Département d'économie (Sciences Po) - Sciences Po - Sciences Po - CNRS - Centre National de la Recherche Scientifique)
    Abstract: We specify and estimate a lifecycle model of consumption, housing demand and labor supply in an environment where individuals may file for bankruptcy or default on their mortgage. Uncertainty in the model is driven by house price shocks, {education specific} productivity shocks, and catastrophic consumption events, while bankruptcy is governed by the basic institutional framework in the US as implied by Chapter 7 and Chapter 13. The model is estimated using micro data on credit reports and mortgages combined with data from the American Community Survey. We use the model to understand the relative importance of the two chapters (7 and 13) for each of our two education groups that differ in both preferences and wage profiles. We also provide an evaluation of the BACPCA reform. Our paper demonstrates importance of distributional effects of Bankruptcy policy.
    Keywords: Lifecycle, Bankruptcy, Mortgage Default, Housing, Labor Supply, Consumption, Education, Insurance, Moral hazard
    Date: 2022–03–17
  25. By: Harrison Mateika; Juannan Jia; Linda Lillard; Noah Cronbaugh; Will Shin
    Abstract: The primary aim of this research was to find a model that best predicts which fallen angel bonds would either potentially rise up back to investment grade bonds and which ones would fall into bankruptcy. To implement the solution, we thought that the ideal method would be to create an optimal machine learning model that could predict bankruptcies. Among the many machine learning models out there we decided to pick four classification methods: logistic regression, KNN, SVM, and NN. We also utilized an automated methods of Google Cloud's machine learning. The results of our model comparisons showed that the models did not predict bankruptcies very well on the original data set with the exception of Google Cloud's machine learning having a high precision score. However, our over-sampled and feature selection data set did perform very well. This could likely be due to the model being over-fitted to match the narrative of the over-sampled data (as in, it does not accurately predict data outside of this data set quite well). Therefore, we were not able to create a model that we are confident that would predict bankruptcies. However, we were able to find value out of this project in two key ways. The first is that Google Cloud's machine learning model in every metric and in every data set either outperformed or performed on par with the other models. The second is that we found that utilizing feature selection did not reduce predictive power that much. This means that we can reduce the amount of data to collect for future experimentation regarding predicting bankruptcies.
    Date: 2022–12
  26. By: Langenbucher, Katja
    Abstract: Search costs for lenders when evaluating potential borrowers are driven by the quality of the underwriting model and by access to data. Both have undergone radical change over the last years, due to the advent of big data and machine learning. For some, this holds the promise of inclusion and better access to finance. Invisible prime applicants perform better under AI than under traditional metrics. Broader data and more refined models help to detect them without triggering prohibitive costs. However, not all applicants profit to the same extent. Historic training data shape algorithms, biases distort results, and data as well as model quality are not always assured. Against this background, an intense debate over algorithmic discrimination has developed. This paper takes a first step towards developing principles of fair lending in the age of AI. It submits that there are fundamental difficulties in fitting algorithmic discrimination into the traditional regime of antidiscrimination laws. Received doctrine with its focus on causation is in many cases ill-equipped to deal with algorithmic decision-making under both, disparate treatment, and disparate impact doctrine.0F 1 The paper concludes with a suggestion to reorient the discussion and with the attempt to outline contours of fair lending law in the age of AI.
    Keywords: credit scoring methodology,AI enabled credit scoring,AI borrower classification,responsible lending,credit scoring regulation,financial privacy,statistical discrimination
    JEL: C18 C32 K12 K23 K33 K40 J14 O31 O33
    Date: 2022

This nep-rmg issue is ©2023 by Stan Miles. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. 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.