A stochastic frontier approach to modelling financial constraints in firms: An application to India

We propose the use of stochastic frontier approach to modelling financial constraints of firms. The main advantage of the stochastic frontier approach over the stylised approaches that use pooled OLS or fixed effects panel regression models is that we can not only decide whether or not the average firm is financially constrained, but also estimate a measure of the degree of the constraint for each firm and for each time period, and also the marginal impact of firm characteristics on this measure. We then apply the stochastic frontier approach to a panel of Indian manufacturing firms, for the 1997–2006 period. In our application, we highlight and discuss the aforementioned advantages, while also demonstrating that the stochastic frontier approach generates regression estimates that are consistent with the stylised intuition found in the literature on financial constraint and the wider literature on the Indian credit/capital market.     

The term structure of banking crisis risk in the United States: A market data based compound option approach

We use a compound option-based structural credit risk model to estimate banking crisis risk for the United States based on market data on bank stocks on a daily frequency. We contribute to the literature by providing separate information on short-term, long-term and total crisis risk instead of a single-maturity risk measure usually inferred by Merton-type models or barrier models. We estimate the model by applying the Duan (1994) maximum-likelihood approach. A strongly increasing total crisis risk estimated from early July 2007 onwards is driven mainly by short-term crisis risk. Banks that defaulted or were overtaken during the crisis have a considerably higher crisis risk (especially higher long-term risk) than banks that survived the crisis.     

Consumer credit-risk models via machine-learning algorithms

We apply machine-learning techniques to construct nonlinear nonparametric forecasting models of consumer credit risk. By combining customer transactions and credit bureau data from January 2005 to April 2009 for a sample of a major commercial bank’s customers, we are able to construct out-of-sample forecasts that significantly improve the classification rates of credit-card-holder delinquencies and defaults, with linear regression R²’s of forecasted/realized delinquencies of 85%. Using conservative assumptions for the costs and benefits of cutting credit lines based on machine-learning forecasts, we estimate the cost savings to range from 6% to 25% of total losses. Moreover, the time-series patterns of estimated delinquency rates from this model over the course of the recent financial crisis suggest that aggregated consumer credit-risk analytics may have important applications in forecasting systemic risk.     

Affine Models for Credit Risk Analysis

Continuous-time affine models have been recently introducedin the theoretical financial literature on credit risk. Theyprovide a coherent modeling, rather easy to implement, but havenot yet encountered the expected success among practitionersand regulators. This is likely due to a lack of flexibilityof these models, which often implied poor fit, especially comparedto more ad hoc approaches proposed by the industry. The aimof this article is to explain that this lack of flexibilityis mainly due to the continuous-time assumption. We developa discrete-time affine analysis of credit risk, explain howdifferent types of factors can be introduced to capture separatelythe term structure of default correlation, default heterogeneity,correlation between default, and loss-given-default; we alsoexplain why the factor dynamics are less constrained in discretetime and are able to reproduce complicated cycle effects. Thesemodels are finally used to derive a credit-VaR and various decompositionsof the spreads for corporate bonds or first-to-default basket.     

Generalized parameter functions for option pricing

We extend the benchmark nonlinear deterministic volatility regression functions of Dumas et al. (1998) to provide a semi-parametric method where an enhancement of the implied parameter values is used in the parametric option pricing models. Besides volatility, skewness and kurtosis of the asset return distribution can also be enhanced. Empirical results, using closing prices of the S&P 500 index call options (in one day ahead out-of-sample pricing tests), strongly support our method that compares favorably with a model that admits stochastic volatility and random jumps. Moreover, it is found to be superior in various robustness tests. Our semi-parametric approach is an effective remedy to the curse of dimensionality presented in nonparametric estimation and its main advantage is that it delivers theoretically consistent option prices and hedging parameters. The economic significance of the approach is tested in terms of hedging, where the evaluation and estimation loss functions are aligned.     

Copula‐Based Formulas to Estimate Unexpected Credit Losses (The Future of Basel Accords?)

The model used to estimate the capital required to cover unexpected credit losses in financial institutions (Basel II) has some drawbacks that reduce its ability to capture potential joint extreme losses in downturns. This paper suggests an alternative approach based on Copula Theory to overcome such flaws. Similarly to Basel II, the suggested model assumes that defaults are driven by a latent variable which varies as a response to an unobserved factor. On the other hand, the use of copulas allows the identification of asymmetric dependence between defaults which has been registered in the literature. As an example, a specific copula family (Clayton) is adopted to represent the association between the latent variables and a formula to estimate potential unexpected losses at a certain level of confidence is derived. Simulations reveal that, in most of the cases, the alternative model outperforms Basel II for portfolios with right‐tail‐dependent probabilities of default (supposedly, a good representation for real loan portfolios).     

AGENT BANK BEHAVIOR IN BANK LOAN SYNDICATIONS

Using Shared National Credit (SNC) Program data from 1995 to 2000, we extend previous empirical work on bank loan syndications. First, we examine recent trends in the volume and examiner‐based credit quality of loans syndicated through the banking system. Second, we estimate a panel regression model to explain changes in an agent bank's retained share of a syndicated loan in terms of information asymmetries, loan credit quality, capital constraints, and loan age and maturity. We find that these variables are significant determinants of the proportion of a SNC loan retained by an agent bank for its portfolio over time.     

Checking for asymmetric default dependence in a credit card portfolio: A copula approach

Traditional credit risk models adopt the linear correlation as a measure of dependence and assume that credit losses are normally-distributed. However some studies have shown that credit losses are seldom normal and the linear correlation does not give accurate assessment for asymmetric data. Therefore it is possible that many credit models tend to misestimate the probability of joint extreme defaults.This paper employs Copula Theory to model the dependence across default rates in a credit card portfolio of a large UK bank and to estimate the likelihood of joint high default rates. Ten copula families are used as candidates to represent the dependence structure. The empirical analysis shows that, when compared to traditional models, estimations based on asymmetric copulas usually yield results closer to the ratio of simultaneous extreme losses observed in the credit card portfolio.Copulas have been applied to evaluate the dependence among corporate debts but this research is the first paper to give evidence of the outperformance of copula estimations in portfolios of consumer loans. Moreover we test some families of copulas that are not typically considered in credit risk studies and find out that three of them are suitable for representing dependence across credit card defaults.     

Estimating loss-given default through advanced credibility theory

The New Basel Accord allows internationally active banking organizations to calculate their credit risk capital requirements using an internal ratings based approach, subject to supervisory review. One of the modeling components is the loss-given default (LGD): it represents the credit loss for a bank when extreme events occur that influence the obligor ability to repay his debts to the bank. Among researchers and practitioners the use of statistical models such as linear regression, Tobit or decision trees is quite common in order to compute LGDs as a forecasting of historical losses. However, these statistical techniques do not seem to provide robust estimation and show low performance. These results could be driven by some factors that make differences in LGD, such as the presence and quality of collateral, timing of the business cycle, workout process management and M&A activity among banks. This paper evaluates an alternative method of modeling LGD using a technique based on advanced credibility theory typically used in actuarial modeling. This technique provides a statistical component to the credit and workout experts’ opinion embedded in the collateral and workout management process and improve the predictive power of forecasting. The model has been applied to an Italian Bank Retail portfolio represented by Overdrafts; the application of credibility theory provides a higher predictive power of LGD estimation and an out-of-time sample backtesting has shown a stable accuracy of estimates with respect to the traditional LGD model.     

Recovery rates of commercial lending: Empirical evidence for German companies

There are very few studies concerning the recovery rate of bank loans. Prediction models of recovery rates are increasing in importance because of the Basel II-framework, the impact on credit risk management, and the calculation of loan rates. In this study, we focus the analyses on the distribution of recovery rates and the impact of the quota of collateral, the creditworthiness of the borrower, the size of the company and the intensity of the client relationship on the recovery rate. All our hypotheses can be confirmed. A higher quota of collateral leads to a higher recovery rate, whereas the risk premium of the borrower and the size of the company is negatively related to the recovery rate. Borrowers with an intense client relationship with the bank exhibit a higher recovery rate.     

Multi-period credit default prediction with time-varying covariates

In credit default prediction models, the need to deal with time-varying covariates often arises. For instance, in the context of corporate default prediction a typical approach is to estimate a hazard model by regressing the hazard rate on time-varying covariates like balance sheet or stock market variables. If the prediction horizon covers multiple periods, this leads to the problem that the future evolution of these covariates is unknown. Consequently, some authors have proposed a framework that augments the prediction problem by covariate forecasting models. In this paper, we present simple alternatives for multi-period prediction that avoid the burden to specify and estimate a model for the covariate processes. In an application to North American public firms, we show that the proposed models deliver high out-of-sample predictive accuracy.     

The delivery option in credit default swaps

Under standard assumptions the reduced-form credit risk model is not capable of accurately pricing the two fundamental credit risk instruments – bonds and credit default swaps (CDS) – simultaneously. Using a data set of euro-denominated corporate bonds and CDS our paper quantifies this mispricing by calibrating such a model to bond data, and subsequently using it to price CDS, resulting in model CDS spreads up to 50% lower on average than observed in the market. An extended model is presented which includes the delivery option implicit in CDS contracts emerging since a basket of bonds is deliverable in default. By using a constant recovery rate standard models assume equal recoveries for all bonds and hence zero value for the delivery option. Contradicting this common assumption, case studies of Chapter 11 filings presented in the paper show that corporate bonds do not necessarily trade at equal levels following default. Our extension models the implied expected recovery rate of the cheapest-to-deliver bond and, applied to data, largely eliminates the mispricing. Calibrated recovery values lie between 8% and 47% for different obligors, exhibiting strong variation among rating classes and industries. A cross-sectional analysis reveals that the implied recovery parameter depends on proxies for the delivery option, primarily the number of available bonds and bond pricing errors. No evidence is found for a direct influence of the bid-ask spread, notional amount, coupon, or rating used as proxies for bond market liquidity.     

A compound option approach to model the interrelation between banking crises and country defaults: The case of Hungary 2008

We analyze the Hungarian financial crisis of 2008 in a stochastic framework that advances structural credit risk models for country defaults: by applying compound option theory we consider payments for bailing-out the banking sector together with debt service payments in a joint crisis model. We estimate the model parameters by applying the time series maximum-likelihood approach of Duan (1994) on yield spreads of Hungarian Bonds. We find that difficulties in acquiring funds for debt servicing in combination with high outstanding debt triggered the crisis, rather than problems in the domestic banking sector. The estimated crisis probabilities dramatically rise during 2008.     

Recovery rates,default probabilities,and the credit cycle

In recessions, the number of defaulting firms rises. On top of this, the average amount recovered on the bonds of defaulting firms tends to decrease. This paper proposes an econometric model in which this joint time-variation in default rates and recovery rate distributions is driven by an unobserved Markov chain, which we interpret as the “credit cycle”. This model is shown to fit better than models in which this joint time-variation is driven by observed macroeconomic variables. We use the model to quantitatively assess the importance of allowing for systematic time-variation in recovery rates, which is often ignored in risk management and pricing models.     

Irish credit unions: Investigating performance determinants and the opportunity cost of regulatory compliance

The study investigates how producer-specific environmental factors influence the performance of Irish credit unions. The empirical analysis uses a two-stage approach. The first stage measures efficiency by a data envelopment analysis (DEA) estimator, which explicitly incorporates the production of undesirable outputs such as bad loans in the modelling, and the second stage uses truncated regression to infer how various factors influence the (bias-corrected) estimated efficiency. A key finding of the analysis is that 68% of Irish credit unions do not incur an extra opportunity cost in meeting regulatory guidance on bad debt.     

Credit rating dynamics and Markov mixture models

Despite mounting evidence to the contrary, credit migration matrices, used in many credit risk and pricing applications, are typically assumed to be generated by a simple Markov process. Based on empirical evidence, we propose a parsimonious model that is a mixture of (two) Markov chains, where the mixing is on the speed of movement among credit ratings. We estimate this model using credit rating histories and show that the mixture model statistically dominates the simple Markov model and that the differences between two models can be economically meaningful. The non-Markov property of our model implies that the future distribution of a firm’s ratings depends not only on its current rating but also on its past rating history. Indeed we find that two firms with identical current credit ratings can have substantially different transition probability vectors. We also find that conditioning on the state of the business cycle or industry group does not remove the heterogeneity with respect to the rate of movement. We go on to compare the performance of mixture and Markov chain using out-of-sample predictions.     

Asymmetric effects of the business cycle on bank credit risk

Prior empirical research on the relation between credit risk and the business cycle has failed to properly investigate the presence of asymmetric effects. To fill this gap, we examine this relation both at the aggregate and the bank level exploiting a unique dataset on Italian banks’ borrowers’ default rates. We employ threshold regression models that allow to endogenously establish different regimes identified by the thresholds over/below which credit risk is more/less cyclical. We find that not only are the effects of the business cycle on credit risk more pronounced during downturns but cyclicality is also higher for those banks with riskier portfolios.     

Alternative bankruptcy prediction models using option-pricing theory

We examine the empirical properties of the theoretical Black–Scholes–Merton (BSM) bankruptcy model. We evaluate the predictive ability of various existing modifications of the BSM model and extend prior studies by estimating volatility directly from market-observable returns on firm value. We show that parsimonious models using our direct market-observable volatility estimate perform better than alternative, more sophisticated, models. Our findings suggest the adoption of simpler modelling approaches relying on market data when implementing the BSM model.     

THE INFORMATIONAL ROLE OF BANK LOAN RATINGS

We analyze the relatively new phenomenon of credit ratings on syndicated loans, asking first whether they convey information to the capital markets. Our event studies show that initial loan ratings and upgrades are not informative, but downgrades are. The market anticipates downgrades to some extent, however. We also examine whether public information reflecting borrower default characteristics explains cross‐sectional variation in loan ratings and find that ratings are only partially predictable. Our evidence suggests that loan and bond ratings are not determined by the same model. Finally, we estimate a credit spread model incorporating bank loan ratings and other factors reflecting default risk, information asymmetry, and agency problems. We find that ratings are related to loan rates, given the effect of other influences on yields, suggesting that ratings provide information not reflected in financial information. Ratings may capture idiosyncratic information about recovery rates, as each of the agencies claims, or information about default prospects not available to the market.     

Correlation expansions for CDO pricing

This paper develops numerical approximations for pricing collateralized debt obligations (CDOs) and other portfolio credit derivatives in the multifactor Normal Copula model. A key aspect of pricing portfolio credit derivatives is capturing dependence between the defaults of the elements of the portfolio. But, compared with an independent-obligor model, pricing in a model with correlated defaults is more challenging. Our approach strikes a balance by reducing the problem of pricing in a model with correlated defaults to calculations involving only independent defaults. We develop approximations based on power series expansions in a parameter that scales the underlying correlations. These expansions express a CDO tranche price in a multifactor model as a series of prices in independent-obligor models, which are easy to compute. The approach builds on a classical approximation for multivariate Gaussian probabilities; we introduce an alternative representation that greatly reduces the number of terms required to evaluate the coefficients in the expansion. We also apply this method to the underlying problem of computing joint probabilities of multivariate normal random variables for which the correlation matrix has a factor structure.