Estimating Default Probabilities Research Articles

A Study of the Potential Impact of Physical and Transition Climate Risks to European Utilities

In this working paper we provide a detailed descriptions of the impact to European Utilities on the impact of climate related risks.We include the impact of both transition and physical risks that could impact the companies over a period of time from 2020 to 2050. The approach utilises a long term simulation framework from an integrated assessment model that drives company investment choices over the simulation horizon. This approach gives rise to estimates of equity valuation, bond prices, credit ratings and default probabilities that can be used to assess long term credit exposures or potential shifts in current market valuation.

Do going concern opinions provide incremental information to predict corporate defaults?

Investors, regulators, and academics question the usefulness of going concern opinions (GCOs). We assess whether GCOs provide incremental information, relative to other predictors of corporate default. Our measure of incremental information is the additional predictive power that GCOs give to a default model. Using data from 1996 to 2015, initially we find no difference in predictive power between GCOs alone and a default model that includes financial ratios. However, there is an imperfect overlap between GCOs and other predictors. We show that GCOs increase the predictive power of several models that include ratios, market variables, probability of default estimates, and credit ratings. Using a model that includes ratios and market variables, GCOs increase the number of predicted defaults by 4.4%, without increasing Type II errors. Our findings suggest that GCOs summarize a complex set of conditions not captured by other predictors of default.

Implied Default Probabilities and Losses Given Default from Option Prices*

Abstract We propose a novel method of estimating default probabilities using equity options data. The resulting default probabilities are highly correlated with estimates of default probabilities extracted from CDS spreads, which assume constant losses given default. Additionally, the option-implied default probabilities are higher in bad economic times and for firms with poorer credit ratings and financial positions. A simple inferred measure of loss given default is related to underlying business conditions, and varies across sectors; the time series properties of this measure are similar after controlling for liquidity effects.

A hybrid model to estimate corporate default probabilities in China based on zero-price probability model and long short-term memory

ABSTRACT This article proposes a hybrid model by combining zero-price probability model with long short-term memory (ZPP-LSTM) to estimate corporate default probabilities. The ZPP-LSTM model enhances the time-series data forecast by introducing LSTM in ZPP model, which can better estimate the corporate default probabilities in the industry sensitive to an uncertain environment. The full samples of Chinese listed companies in construction and real estate industries are selected to evaluate the performance of ZPP-LSTM model. The results show that our proposed model outperforms other benchmark models in terms of the default probability estimation.

The Aggregate Consequences of Default Risk: Evidence from Firm-Level Data

This paper studies the implications of perceived default risk for aggregate output and productivity. Using a model of credit contracts with moral hazard, we show that a firm's probability of default is a sufficient statistic for capital allocation. The theoretical framework suggests an aggregate measure of the impact of credit market frictions based on firm-level probabilities of default which can be applied using data on firm-level employment and default risk. We obtain direct estimates of firm-level default probabilities using Standard and Poor's PD Model to capture the expectations that lenders were forming based on their historical information sets. We implement the method on the UK, an economy that was strongly exposed to the global financial crisis and where we can match default probabilities to administrative data on the population of 1.5 million firms per year. As expected, we find a strong correlation between default risk and a firm's future performance. We estimate that credit frictions (i) cause an output loss of around 28% per year on average; (ii) are much larger for firms with under 250 employees and (iii) that losses are overwhelmingly due to a lower overall capital stock rather than a misallocation of credit across firms with heterogeneous productivity. Further, we find that these losses accounted for over half of the productivity fall between 2008 and 2009, and persisted for smaller (although not larger) firms.

The Aggregate Consequences of Default Risk: Evidence from Firm-Level Data

This paper studies the implications of perceived default risk for aggregate output and productivity. Using a model of credit contracts with moral hazard, we show that a firm’s probability of default is a sufficient statistic for capital allocation. The theoretical framework suggests an aggregate measure of the impact of credit market frictions based on firm-level probabilities of default which can be applied using data on firm-level employment and default risk. We obtain direct estimates of firm-level default probabilities using Standard and Poor’s PD Model to capture the expectations that lenders were forming based on their historical information sets. We implement the method on the UK, an economy that was strongly exposed to the global financial crisis and where we can match default probabilities to administrative data on the population of 1.5 million firms per year. As expected, we find a strong correlation between default risk and a firm’s future performance. We estimate that credit frictions (i) cause an output loss of around 28% per year on average; (ii) are much larger for firms with under 250 employees and (iii) that losses are overwhelmingly due to a lower overall capital stock rather than a misallocation of credit across firms with heterogeneous productivity. Further, we find that these losses accounted for over half of the productivity fall between 2008 and 2009, and persisted for smaller (although not larger) firms

Default Probability Assessment for Project Finance Bank Loans and Basel Regulations: Searching for a New Paradigm

Syndicated loan data from India suggest that despite the pressure on margins and volumes, available “bankable” deals in infrastructure are still strongly contested by commercial lenders. With the implementation of Ind AS accounting norms (IFRS) for commercial banks, the expected credit loss–based provisions have to be set by the bank based on internally estimated probability of default (PD) for different loan portfolios. This change may lead to heavier risk-weighted capital requirements for banks, thus impacting the project finance business. Basel III norms and subsequent discussion papers propose a revised standardized approach doing away with internal modeling approaches and introduction of standardized output floors for specialized lending, including project finance. In this light, the authors present a cash flow simulation model to address the issue of PD estimation by simulating key risk factors. This method may be useful as each project and each sector is unique and so are the risks associated with it. Thus, the authors argue that the use of a simulation model will result in better assessment and monitoring of credit risk than conventional assessment methods, leading to lower default rates and therefore lower capital charge. The authors then suggest some new rules of engagement for project finance lenders to stay relevant in the changing regulatory scenario. TOPICS:Simulations, project finance, credit risk management Key Findings • Probability of default estimation and subsequent impact on credit risk capital is important for project finance lenders in India especially after emerging Basel III reforms and subsequent discussion papers on credit risk capital. (Informally called as Basel IV) • The authors suggest a cash flow simulation model developed using risk parameters “specific” to each project. The application of the model is shown on a road project and it calculates cumulative default risk probability of the project. • The article argues that this model will result in better assessment and monitoring of credit risk. Authors also suggest new rules for engagement for the project finance lenders in the emerging regulatory scenario.

Estimating the Probability of Default for No-Default and Low-Default Portfolios

SummaryThe paper proposes a sequential Bayesian updating approach to estimate default probabilities on rating grade level for no- and low-default portfolios. Bayesian sequential updating enables default probabilities to be obtained also for those rating grades for which no defaults have been observed. The advantage of this approach is that it preserves the rank order of rating grades in the case of no defaults. Rank preservation is not ensured when using an identical prior distribution across all rating grades. We discuss Bayesian sequential updating for the beta–binomial model and a model incorporating the asymptotic single-risk factor model of the Basel Accord. Practical aspects such as incorporating information from external sources and the margin of conservatism are addressed.

Extracting Option-Implied Probability of Default: A Novel Method

In this paper, we present a novel method to extract the risk-neutral probability of default from American put option prices. Under the assumptions of Carr and Wu (2011), we derive a closed form expression for American put options from which the probability of default can be inferred. Our empirical results based on four large institutions in the US during the 2007-2009 crisis show that directly applying the methodology of Carr and Wu (2011) can overestimate the probability of default. Our model also allows us to test the effect of positive equity recovery on estimated default probabilities.

The Systemic Risk Implications of Using Credit Ratings Versus Quantitative Measures to Limit Bond Portfolio Risk

Despite intense criticism, agency credit ratings are still widely used in regulation and risk management. One possible alternative is to replace them with quantitative default risk measures. For US data, I find that systemically relevant losses from corporate defaults are mostly smaller if risk-taking in portfolios is limited with the help of default probability estimates from the Credit Research Initiative rather than through Moody’s ratings. The results continue to hold when investors follow a regulatory arbitrage strategy that tilts portfolios toward issuers with high systematic risk. I further show that combining information from both measures can lead to a systemic risk profile that is more favorable than can be achieved by using only one.

Estimating the Probability of Default for No-Default and Low-Default Portfolios

This article proposes a sequential Bayesian updating approach to estimate default probabilities on rating grade level for no- and low-default portfolios. Bayesian sequential updating allows to obtain default probabilities also for those rating grades for which no defaults have been observed. The advantage of the proposed approach is that it preserves the rank order of rating grades in case of no defaults. Rank-preservation is not ensured when using an identical prior distribution across all rating grades. We discuss Bayesian sequential updating for the beta-binomial model and a model incorporating the asymptotic single risk factor model of the Basel Accord. Practical aspects such as incorporating information from external sources and the margin of conservatism are addressed.

Measuring Systemic Risk in the U.S. Banking System

This paper develops a novel measure of systemic risk that combines mapping technology and regression methods. Self-organizing maps (SOM) and lasso logistic regressions are employed to estimate default probabilities for individual U.S. commercial banks from 2001 to 2017. Subsequently, these probabilities are aggregated into a size-weighted measure of systemic risk dubbed SYSTEM. Empirical results show that, due primarily to large banks, volatility in systemic risk increased in 2005 followed by a very large spike from late 2006 to 2008 related to the financial crisis. Comparative tests to the popular systemic risk measure SRISK reveal that SYSTEM: (1) provided earlier warning signals of the impending 2008−2009 crisis; and (2) indicated relatively lower systemic risk after 2012. Further tests show that SYSTEM and SRISK are useful in predicting industry-wide nonperforming loans and numbers of bank failures. We conclude that micro- and macro-prudential measures of bank condition are useful in assessing and predicting systemic risk.

Adverse Selection in the Home Equity Line of Credit Market

Securitization has been widely assigned blame for contributing to the recent mortgage market meltdown and ensuing financial crisis. In this paper, we employ the Office of the Comptroller of the Currency (OCC) Home Equity database to develop estimates of default and prepayment probabilities for home equity lines of credit (HELOCs) originated during 2004–2008 and tracked through 2014. The results show that securitized HELOCs bear both higher default and prepayment risk compared to loans held in lender portfolios, consistent with adverse selection in the securitization process. Spreads on securitized HELOCs are higher as well, consistent with a “lemons spread.” The results are robust across multiple specifications, including models that address the endogenous decision to securitize, and support adoption of the credit risk retention rule for asset-backed securities.

Valoración de credit default swap aplicación del modelo de Jarrow y Turnbull en un bono de deuda privada en Colombia

The study presents the empirical application in reduced form of a model for the estimation of default probabilities, which offers benefits compared to other models because it can be used not only for the calculation of the probability of non-compliance but also for the valuation of the derivative of CDS credit from the buyer´s point of view. Using this model is appropriate when the market is characterized by low liquidity and insufficient statistical information. As the basis for the model´s application, the information on a five-year benchmark bond of the Ecopetrol oil company has been used, to show the procedure for the estimation of the CDS premium. This allowed making comparisons between the theoretical model and values available on the market. The result has been a very similar value between them, and it is shown that the model fits to estimate the annual base probability, the cumulative CDS default probability, and the CDS premium on private bonds. This information could serve as a reference for investors to manage default risk on private bonds in Colombia using CDS.

Do Stock Returns Really Decrease with Default Risk? New International Evidence

This study constructs a novel data set of bankruptcy filings for a large sample of non-U.S. firms in 14 developed markets and sheds new light on the cross-sectional relation between default risk and stock returns. Using the reduced-form approach of Campbell et al. (2008) to estimate default probabilities, we offer conclusive evidence supporting the existence of a significant positive default risk premium in international markets. This finding is robust to different portfolio weighting schemes, data filters, risk-adjusting approaches, and holding period definitions. Decomposing the default risk measure into its systematic and idiosyncratic components, we find that the former drives this positive relation. We also show that the default risk premium is more pronounced in countries where creditor protection is stronger and shareholder bargaining power is lower. The online appendix is available at https://doi.org/10.1287/mnsc.2016.2712 . This paper was accepted by Amit Seru, finance.

The Myth of the Credit Spread Puzzle

We ask whether a standard structural model (Black and Cox, 1976) is able to explain credit spreads on corporate bonds and, in contrast to much of the literature, we find that the model matches the level of investment grade spreads well. Model spreads for speculative grade debt are too low and we find that bond illiquidity contributes to this underpricing. Our analysis makes use of a new approach for calibrating the model to historical default rates that leads to much more precise estimates of investment grade default probabilities.

Estimating Default Risk of Bank Loans in Zimbabwe Using the Mover-Stayer Model

This paper estimates default probabilities of bank loans through the use of a mover stayer model, using the data of Zimbabwe. Management of credit risk is an element of financial engineering, which provides safeguard against institutions? financial failure. The study aimed at estimating default probabilities of bank loans through the use of a mover stayer model. In-order to achieve this, the study compared the predictive power of duration method and cohort method in forecasting default risk of bank loans, tested for presence of time homogeneity and determined the model with an upper hand between the Mover stayer model and the Markov Chain Model in gauging default risk. It was concluded that the cohort approach has an upper hand than the duration approach and that there was time inhomogeneity. There was also significant evidence that the mover-stayer model is a superior and effective way of estimating the risk of default. The effectiveness was shown through the process of back testing. The forecasted results for 2014 were in line with the actual 2014 default results hence the model called the Mover Stayer effectively and competently predicted the risk of default of bank advances and loans.

Adverse Selection in the Home Equity Line of Credit Market

Securitization has been widely assigned blame for contributing to the recent mortgage market meltdown and ensuing financial crisis. In this paper, we employ the OCC Home Equity database to develop estimates of default and prepayment probabilities for home equity lines of credit originated during 2004-2008 and tracked through 2014. Results show that securitized home equity loans bear both higher default and prepayment risk compared to loans held in portfolio by lenders, consistent with adverse selection in the securitization process. Spreads on securitized home equity lines are higher, as well, consistent with a “lemons spread”. Results are robust across multiple specifications, including models that address the endogenous decision to securitize, and support adoption of the credit risk retention rule for asset-backed securities.

Nonparametric Estimation for Non-Homogeneous Semi-Markov Processes: An Application to Credit Risk

We propose procedures for estimating the time-dependent transition matrices for the general class of finite nonhomogeneous continuous-time semi-Markov processes. We prove the existence and uniqueness of solutions for the system of Volterra integral equations defining the transition matrices, therefore showing that these empirical transition probabilities can be estimated from window censored event-history data. An implementation of the method is presented based on nonparametric estimators of the hazard rate functions in the general and separable cases. A Monte Carlo study is performed to assess the small sample behavior of the resulting estimators. We use these new estimators for dealing with a central issue in credit risk. We consider the problem of obtaining estimates of the historical corporate default and rating migration probabilities using a dataset on credit ratings from Standard & Poor's.

Estimation of Default Probabilities: Application of the Discriminant Analysis and the Structural Approach for Companies Listed on the BVC

This article aims to compare the calculated results of the structural approach (Internal Ratings-Based IRB) and the discriminant analysis (Z-score of Altman, 1968), based on data from companies listed on the BVC for the period from 02 January 2014 to December 31, 2014. The structural approach is directly linked to the economic reality of the company; the default takes place as soon as the market value of these assets falls below a certain threshold. The major constraint for this approach is the determination of the probabilities of default. This situation is overcome by using the Black & Scholes (1973) model, based on Monte Carlo simulations. While the Z-score method is a financial analysis technique of business failure predictions, which is based on financial and economic ratios.