Forecasting and stress testing credit card default using dynamic models

We present discrete time survival models of borrower default for credit cards that include behavioural data about credit card holders and macroeconomic conditions across the credit card lifetime. We find that dynamic models which include these behavioural and macroeconomic variables provide statistically significant improvements in model fit, which translate into better forecasts of default at both account and portfolio levels when applied to an out-of-sample data set. By simulating extreme economic conditions, we show how these models can be used to stress test credit card portfolios.     

Bayesian loss given default estimation for European sovereign bonds

We develop and apply a Bayesian model for the loss rates given defaults (LGDs) of European Sovereigns. Financial institutions are in need of LGD forecasts under Pillar II of the regulatory Basel Accord and the downturn in LGD forecasts under Pillar I. Both are challenging for portfolios with a small number of observations such as sovereigns. Our approach comprises parameter risk and generates LGD forecasts under both regular and downturn conditions. With sovereign-specific rating information, we found that average LGD estimates vary between 0.46 and 0.64, while downturn estimates lay between 0.50 and 0.86.     

信息不对称下保险信用缺失分析及防范

当前,保险信用缺失已成为制约保险市场进一步发展的瓶颈。文中从投保人、保险人的逆向选择及道德风险三个方面阐述了保险信用缺失的原因,同时提出了健全我国保险信用体系的对策。     

Forecasting corporate default risk in China

Default risk prediction can not only provide forward-looking and timely risk measures for regulators and investors, but also improve the stability of the financial system. However, the determinants of corporate default risk in China have not been well-identified. An empirical analysis was conducted using a unique dataset of default events in the Chinese market to fill this gap. First, we demonstrated that the default probability estimated by a structural model, which is widely used in the literature, do not fully reveal the default risk of firms in China. Second, we classified default events into minor and major defaults for empirical analysis. We found that firms that survive minor defaults behave differently from other bankrupt firms. Our results suggest that the determinants of corporate default risk in China and the United States differ. We also found that a firm’s continued increase in cash holdings is one of the most important signs of default. Overall, our study significantly improves the accuracy of forecasting corporate default risk in China.     

Predicting loss given default (LGD) for residential mortgage loans: A two-stage model and empirical evidence for UK bank data

With the implementation of the Basel II regulatory framework, it became increasingly important for financial institutions to develop accurate loss models. This work investigates the loss given default (LGD) of mortgage loans using a large set of recovery data of residential mortgage defaults from a major UK bank. A Probability of Repossession Model and a Haircut Model are developed and then combined to give an expected loss percentage. We find that the Probability of Repossession Model should consist of more than just the commonly used loan-to-value ratio, and that the estimation of LGD benefits from the Haircut Model, which predicts the discount which the sale price of a repossessed property may undergo. This two-stage LGD model is shown to perform better than a single-stage LGD model (which models LGD directly from loan and collateral characteristics), as it achieves a better R² value and matches the distribution of the observed LGD more accurately.     

Predicting loss given default in leasing: A closer look at models and variable selection

Since the introduction of the Basel II Accord, and given its huge implications for credit risk management, the modeling and prediction of the loss given default (LGD) have become increasingly important tasks. Institutions which use their own LGD estimates can build either simpler or more complex methods. Simpler methods are easier to implement and more interpretable, but more complex methods promise higher prediction accuracies. Using a proprietary data set of 1,184 defaulted corporate leases in Germany, this study explores different parametric, semi-parametric and non-parametric approaches that attempt to predict the LGD. By conducting the analyses for different information sets, we study how the prediction accuracy changes depending on the set of information that is available. Furthermore, we use a variable importance measure to identify the input variables that have the greatest effects on the LGD prediction accuracy for each method. In this regard, we provide new insights on the characteristics of leasing LGDs. We find that (1) more sophisticated methods, especially the random forest, lead to remarkable increases in the prediction accuracy; (2) updating information improves the prediction accuracy considerably; and (3) the outstanding exposure at default, an internal rating, asset types and lessor industries turn out to be important drivers of accurate LGD predictions.     

Default risk and private student loans: Implications for higher education policies

In recent years, the proportion of students facing a binding constraint on government student loans has grown. This has led to substantially increased use of private loans as a supplementary source of finance for households׳ higher education investment. A critical aspect of the private market for student loans is that loan terms must reflect students׳ risk of default. College investment will therefore differ from a world in which government student loans, whose terms are not sensitive to credit risk, are expanded to no longer bind. Moreover, beyond simply crowding out private lending, expansions of the government student loan program will feed back into default risk on private loans. The goal of this paper is to provide a quantitative assessment of the likely effects of the private market for student loans on college enrollment. We build a model of college investment that reflects uninsured idiosyncratic risk and a well-defined life-cycle that is consistent with observed borrowing and default behavior across family income and college preparedness. We find that higher government borrowing limits increase college investment but lead to more default in the private market for student loans, while tuition subsides increase college investment and reduce default rates in the private market. Consequently, higher limits on government student loans have small negative welfare effects, while tuition subsidies increase aggregate welfare.     

Modern pandemic crises and default risk: Worldwide evidence

This article examines the relationship between modern health pandemic crises and financial stability. Specifically, it collects data on 250,223 firms in 43 countries (or regions) during five modern pandemic crises, SARS (2003), H1N1 (2009), MERS (2012), Ebola (2014), and Zika (2016), and finds that pandemic crises significantly increase the default risk of enterprises. Further analysis shows that formal and informal institutions acted as a “cushion” against the pandemic crisis. The earlier a country adopts IFRS, the more unimpeded access to information, and the more stable religious and ethnic relations within the country can reduce the negative impact of a pandemic on financial stability. This article addresses the hitherto inadequacy of COVID-related data. In addition, this article argues that governments should build sound state institutions to withstand macroeconomic shocks and highlights the heterogeneity of default risk for enterprises operating in countries with different institutions.     

The impact of central clearing on the market for single-name credit default swaps

In this article, we revisit the impact of the voluntary central clearing scheme on the CDS market. In order to address the endogeneity problem, we use a robust methodology that relies on dynamic propensity-score matching combined with generalized difference-in-differences. Our empirical findings show that central clearing results in a small increase in CDS spreads (ranging from 14 to 19 bps), while there is no evidence of an associated improvement in CDS market liquidity and trading activity or of a deterioration in the default risk of the underlying bond. These results suggest that the increase in CDS spreads can be mainly attributed to a reduction in CDS counterparty risk.     

Environmental,social and governance disclosure and default risk

We investigate whether environmental, social and governance (ESG) disclosure is related to default risk. Using a sample of US nonfinancial institutions from 2006 to 2017, we find that ESG disclosure is positively related to Merton's distance to default and is negatively related to the credit default swap spread, which suggests that firms with a higher ESG disclosure have lower default risk. Our analysis further indicates that the inverse effect of ESG disclosure on default risk is through increased profitability and reduced performance variability and cost of debt. We also document that the negative impact of ESG disclosure on default risk is existent only for mature and older firms. These results are important for all stakeholders of firms, including shareholders and bondholders to consider firm's ESG disclosure in conjunction with life cycle stage before making their investment decisions.     

Financial fragility,sovereign default risk and the limits to commercial bank bail-outs

We show that with intertwined weak banks and weak sovereigns, bank recapitalizations become much less effective. We construct a DSGE model with leverage constrained banks lending to firms and holding domestic government bonds. Bond prices reflect endogenously generated sovereign risk. This introduces a negative amplification cycle: after a credit crisis output losses increase more because higher interest rates trigger lower bond prices and subsequent losses at banks. This further tightens bank leverage constraints, and causes interest rates to rise further. Also bank recapitalizations are then much less effective. Recaps involve swaps of newly issued sovereign bonds for bank equity, the new debt increases sovereign debt discounts, leading to capital losses for the banks on their holdings of sovereign debt that (partially) offset the impact of the recapitalization. The favorable macroeconomic effects of bank recaps on the recovery after a financial crisis are correspondingly lower.     

A mixture model for credit card exposure at default using the GAMLSS framework

The Basel II and III Accords propose estimating the credit conversion factor (CCF) to model exposure at default (EAD) for credit cards and other forms of revolving credit. Alternatively, recent work has suggested it may be beneficial to predict the EAD directly, i.e.modelling the balance as a function of a series of risk drivers. In this paper, we propose a novel approach combining two ideas proposed in the literature and test its effectiveness using a large dataset of credit card defaults not previously used in the EAD literature. We predict EAD by fitting a regression model using the generalised additive model for location, scale, and shape (GAMLSS) framework. We conjecture that the EAD level and risk drivers of its mean and dispersion parameters could substantially differ between the debtors who hit the credit limit (i.e.“maxed out” their cards) prior to default and those who did not, and thus implement a mixture model conditioning on these two respective scenarios. In addition to identifying the most significant explanatory variables for each model component, our analysis suggests that predictive accuracy is improved, both by using GAMLSS (and its ability to incorporate non-linear effects) as well as by introducing the mixture component.     

Forecasting recovery rates on non-performing loans with machine learning

We compare the performance of a wide set of regression techniques and machine-learning algorithms for predicting recovery rates on non-performing loans, using a private database from a European debt collection agency. We find that rule-based algorithms such as Cubist, boosted trees, and random forests perform significantly better than other approaches. In addition to loan contract specificities, predictors that refer to the bank recovery process — prior to the portfolio’s sale to a debt collector — are also shown to enhance forecasting performance. These variables, derived from the time series of contacts to defaulted clients and client reimbursements to the bank, help all algorithms better identify debtors with different repayment ability and/or commitment, and in general those with different recovery potential.     

Sequential optimization three-way decision model with information gain for credit default risk evaluation

To categorize credit applications into defaulters or non-defaulters, most credit evaluation models have employed binary classification methods based on default probabilities. However, while some loan applications can be directly accepted or rejected, there are others on which immediate accurate credit status decisions cannot be made using existing information. To resolve these issues, this study developed an optimized sequential three-way decision model. First, an information gain objective function was built for the three-way decision, after which a genetic algorithm (GA) was applied to determine the optimal decision thresholds. Then, appropriate accept or reject decisions for some applicants were made using basic credit information, with the remaining applicants, whose credit status was difficult to determine, being divided into a boundary region (BND). Supplementary information was then added to reevaluate the credit applicants in the BND, and a sequential optimization process was employed to ensure more accurate predictions. Therefore, the model’s predictive abilities were improved and the information acquisition costs controlled. The empirical results demonstrated that the proposed model was able to outperform other benchmarking credit models based on performance indicators.     

Write-down bonds,credit risk and imperfect information

A structural model of pricing Write-Down (hereafter WD) bonds under imperfect information has been developed to investigate the effect of WD bonds issuance on credit risk. Information is not only delayed but also asymmetrically distributed between managers and outside investors. We derive analytical solutions for corporate securities prices and find the issuance of WD bonds could significantly improve firm value via reducing bankruptcy cost. Our numerical results further demonstrate that the WD bonds issuance increases corporate risk tolerance and reduces the risk of bankruptcy and credit spreads under imperfect information.     

Assessment of time-varying systemic risk in credit default swap indices: Simultaneity and contagiousness

This study assesses systemic risk inherent in credit default swap (CDS) indices using empirical and statistical analyses. We define systemic risk in two perspectives: the possibilities of simultaneous and contagious defaults, and then quantify them separately across benchmark models. To do so, we employ a Marshall-Olkin copula model to measure simultaneous default risk, and an interacting intensity-based model to capture contagious default risk. For an empirical test, we collect daily data for the iTraxx Europe CDS index and its tranche prices in the period from 2005 to 2014, and calibrate model parameters varying across time. In addition, we select forecasting models that have minimal prediction errors for the calibrated time series. Finally, we identify significant changes in each dynamic of systemic risk indicator before and after default and downgrade-related episodes that have occurred in the global financial crisis and European sovereign debt crisis.     

Measuring connectedness of euro area sovereign risk

We introduce a method for measuring the default risk connectedness of euro zone sovereign states using credit default swap (CDS) and bond data. The connectedness measure is based on an out-of-sample variance decomposition of model forecast errors. Due to its predictive nature, it can respond to crisis occurrences more quickly than common in-sample techniques. We determine the sovereign default risk connectedness using both CDS and bond data in order to obtain a more comprehensive picture of the system. We find evidence that there are several observable factors that drive the difference between CDS and bonds, but both data sources still contain specific information for connectedness spill-overs. In general, we can identify countries that impose risk on the system and the respective spill-over channels. Our empirical analysis covers the years 2009–2014, such that the recovery paths of countries exiting EU and IMF financial assistance schemes and the responses to the ECB’s unconventional policy measures can be analyzed.     

Estimation of the mean of the selected population under asymmetric loss function

This paper is concerned with the comparison of seven estimators of the mean of the selected population from two normal populations with unknown means and common known variance under an asymmetric loss namely the LINEX loss function. The proposed estimators are invariant under location transformation. The bias and risks of the seven estimators are computed and compared. The conclusion recommend the use of δ_P (σ) which is simple to use and it is minimax. Received: January 1999     

Corporate credit risk prediction under stochastic volatility and jumps

This paper examines the impact of allowing for stochastic volatility and jumps (SVJ) in a structural model on corporate credit risk prediction. The results from a simulation study verify the better performance of the SVJ model compared with the commonly used Merton model, and three sources are provided to explain the superiority. The empirical analysis on two real samples further ascertains the importance of recognizing the stochastic volatility and jumps by showing that the SVJ model decreases bias in spread prediction from the Merton model, and better explains the time variation in actual CDS spreads. The improvements are found particularly apparent in small firms or when the market is turbulent such as the recent financial crisis.     

Interconnectedness and systemic risk in the US CDS market

This study assesses systemic risk in the US credit default swap (CDS) market. First, this study estimates the bilateral exposures matrix using aggregate fair value data and theoretically analyze interconnectedness in the US CDS network using various network measures. Second, this study theoretically analyzes the contagious defaults. The default analysis shows the theoretical occurrence of many stand-alone defaults and one contagious default via the CDS network during the global financial crisis. A stress test based on a hypothetical severe stress scenario predicts almost no future contagious defaults. Thus, risk contagion via the CDS network is unlikely.