Chapter 11 - Estimating Parameters Required for Credit Risk Modeling

Introduction Three main variables affect the credit risk of a financial asset: (i) the probability of default (PD), (ii) the ‘loss given default’ (LGD), which is equal to one minus the recovery rate in the event of default (RR), and (iii) the exposure at default (EAD). While significant attention has been devoted by the credit risk literature on the estimation of the first component (PD), much less attention has been dedicated to the estimation of RR and to the relationship between PD and RR. This is mainly the consequence of two related factors. First, credit pricing models and risk management applications tend to focus on the systematic risk components of credit risk, as these are the only ones that attract risk-premia. Second, credit risk models traditionally assumed RR to be dependent on individual features (e.g. collateral or seniority) that do not respond to systematic factors, and therefore to be independent of PD. This traditional focus only on default analysis has been partly reversed by the recent increase in the number of studies dedicated to the subject of RR estimation and the relationship between the PD and RR (Fridson et al . 2000, Gupton et al . 2000, Altman et al . 2001, Altman et al . 2003, 2005, Frye 2000a, 2000b, 2000c, Hu and Perraudin 2002, Hamilton et al . 2001, Jarrow 2001 and Jokivuolle and Peura 2003). This is partly the consequence of the parallel increase in default rates and decrease of recovery rates registered during the 1999–2002 period.

<p>Credit risk refers to the probability of financial loss resulting from a borrower’s failure to repay a loan. Essentially, it encompasses the risk that a lender may not receive the owed principal and interest, leading to disrupted cash flows and increased collection costs. Lenders can mitigate credit risk by analyzing factors related to a borrower’s creditworthiness, such as their current debt load and income. In the last decade, many banks have started to make use of models to assess the risks of lending credit. The credit risk models are very complex and include algorithm-based methods of assessing credit risk. Such a model aims to help banks quantify, aggregate, and manage credit risk. Despite the method, the focus of credit risk assessment stays on credit quality and risk exposure. Strategies to reduce losses and manage risks are pertinent in credit risk management. However, banks have to organize and manage the lending function professionally and proactively and use advanced techniques to measure and manage risks. Credit risk management has become a hot topic due to the ongoing global economic crises, the rapid digital transformation, the recent technological innovations, and the growing use of artificial intelligence in banking. Regulators expect banks to have a clear and comprehensive understanding of their customers and their credit risk and to be transparent and capable in this area. As the Basel regulations change, banks will face more regulatory pressure. To meet the changing regulatory demands and to manage risk better, many banks are changing their credit risk practices. However, banks that see this as only a compliance issue are missing the point. The research findings indicate that the bank’s credit management practices have strongly influenced its profitability. In brief, manoeuvring through the complex risk terrain poses a significant hurdle for financial organizations. This requires ongoing adjustments and strategic choices to ensure both stability and profitability. Consequently, persistent academic research is essential to guide management, governments, and regulators.</p><p><strong> </strong></p><p><strong>JEL</strong>: G21, G32, G33, C53, E58, D81</p><p> </p><p><strong> Article visualizations:</strong></p><p><img src="/-counters-/soc/0771/a.php" alt="Hit counter" /></p>

The use of stress testing for risk monitoring has increased considerably over the last decade. Stress testing – a simulation technique used to assess the strength of a portfolio or a financial institution under unusual economic conditions – emerged as a powerful tool that was originally used in market risk. Its use has subsequently been extended into credit risk. To stress test a credit risk portfolio, practitioners focus on the key parameters that allow the risk of a credit portfolio to be assessed. These parameters, also known as Basel II parameters, are probability of default, loss given default, exposure at default and asset correlation. In this paper, using a time series approach (specifically, an error correction model), we focus on the probability of default parameter that is related to macroeconomic factors. Such an approach involves dealing with the nonstationarity of economic time series and cointegration issues. Hence, when the model is estimated, the probability of default can be simulated by measuring the effects of macroeconomic shocks applied to the model. In turn, these probabilities of default can be used to measure the impact on the probability of default under a given macroeconomic scenario, and then to improve the credit risk monitoring. The results of our study suggest that error correction models are well-suited to macroeconomic stress testing. Indeed, the fitting of the historical probability of default and the results under the stress scenarios considered here are satisfactory.

Decomposing the Smile: Systematic Credit Risk in Mortgage Portfolios

This paper develops rare-event simulation methods for the estimation of portfolio credit risk—the risk of losses to a portfolio resulting from defaults of assets in the portfolio. Portfolio credit risk is measured through probabilities of large losses, which are typically due to defaults of many obligors (sources of credit risk) to which a portfolio is exposed. An essential element of a portfolio view of credit risk is a model of dependence between these sources of credit risk: large losses occur rarely and are most likely to result from systematic risk factors that affect multiple obligors. As a consequence, estimating portfolio credit risk poses a challenge both because of the rare-event property of large losses and the dependence between defaults. To address this problem, we develop an importance sampling technique within the widely used Gaussian copula model of dependence. We focus on difficulties arising in multifactor models—that is, models in which multiple factors may be common to multiple obligors, resulting in complex dependence between defaults. Our importance sampling procedure shifts the mean of the common factor to increase the frequency of large losses. In multifactor models, different combinations of factor outcomes and defaults can produce large losses, so our method combines multiple importance sampling distributions, each associated with a shift in the mean of common factors. We characterize “optimal” mean shifts. Finding these points is both a combinatorial problem and a convex optimization problem, so we address computational aspects of this step as well. We establish asymptotic optimality results for our method, showing that—unlike standard simulation—it remains efficient as the event of interest becomes rarer.

The intersection of market and credit risk

In this paper we use a reduced form model for the analysis of Portfolio Credit Risk. For this purpose, we fit a Dynamic Factor model, DF, to a large dataset of default rates proxies and macro-variables for Italy. Multi step ahead density and probability forecasts are obtained by employing both the direct and indirect method of prediction together with stochastic simulation of the DF model. We, first, find that the direct method is the best performer regarding the out of sample projection of financial distressful events. In a second stage of the analysis, the direct method of forecasting through principal components is shown to provide the least sensitive measures of Portfolio Credit Risk to various multifactor model specifications. Finally, the simulation results suggest that the benefits in terms of credit risk diversification tend to diminish with an increasing number of factors, especially when using the indirect method of forecasting.

The quality of bank loan portfolios and the incurred credit risk loss is determined by the development of the economy of the country as well as the financial state of the debtors. With loans accounting for the largest share of the assets of Lithuanian banks, the credit risk is the main individual source of risk. With the current recession of the world economy and the respective situation in Lithuania, when the economy of the country started to recede at the end of 2008, it is obvious that in the future the processes of development of economy of Lithuania will surely be a major influence on the credit risk and results of activities of the bank system of the country, therefore evaluation of this effect under current conditions becomes critical. The importance of research of the influence of macroeconomic processes on the sorts of risks incurred by the bank sector is also proved by the abundance of theoretical research. Testing of the credit risk of the Lithuanian bank sector upon change of risk factors is carried out combining the models used in research performed by M. Sorge (2004), M. A. Segoviano Basurto, P. Padilla (2007) and other scientists. The research covers five phases: the first phase – modeling of macroeconomic scenarios performing an analysis of the current macroeconomic situation of the country, analyzing expert evaluations and establishing the main risk factors. During the second phase indicators describing the quality of bank loan portfolio are evaluated performing an evaluation of the current quality of loan portfolio and calculating the probability of default according to loan sectors. The third phase covers the establishment of macroeconomic indicators influencing the change of the probability of default and the net profit of banks by selecting macroeconomic indicators that can influence the change of the probability of default according to loan sectors and the net profit of banks, by eliminating the seasonal fluctuations of the selected indicators, by evaluating correlation between macroeconomic indicators and the probability of default as well as the net profit, by defining the parameters of explanatory macroeconomic variables, by employing the model of linear regression and verifying reliability. The fourth phase is the research of the influence of macroeconomic scenarios on the credit risk of the bank sector by providing a prognosis of explanatory macroeconomic variables under the formulated macroeconomic scenarios and evaluating the influence of the macroeconomic scenarios on the probability of default according to loan sectors and the net profit of banks. The research is completed with a prognosis of the reserves accumulated by banks for covering the probable losses on the basis of different scenarios: scenario 1– expensive loans, i.e. the rate of interest for loans in litas remains very high, scenario 2 – dropping real estate prices. Evaluation of the credit risk of the bank sector combining the models created by M. Sorge (2004) & M. A. Segoviano Basurto, P. Padilla (2007) helps to raise the most relevant questions of the macroeconomic state for the current period, and the integration of the analysis of the indicators of financial stability into the critical testing of the bank system and drawing up of macroeconomic prognosis helps to form an efficient future activity strategy for the bank sector or an individual financial institution. The results of the research showed that in case the macroeconomics of the country develops according to scenario 2 the Lithuanian bank system would face deficit of capital for covering the probable losses, therefore the banks of the country should accumulate more capital in order to be able to avoid such risk in the future.

Credit risk rating systems at large US banks

Chapter 21 - The Basics of Portfolio Credit Risk

Transition matrices are an important determinant in risk management and VAR calculations for credit portfolios. It is well known that rating migration behavior is not constant through time in that it shows cyclicality and significant change over the years. We investigate the effect of changes in migration matrices on credit portfolio risk in terms of expected loss and value-at-risk figures for illustrative loan portfolios. The estimates are based on historical transition matrices for different time horizons and a continuous-time simulation procedure. We further determine confidence sets for the probability of default (PD) in different rating classes by a bootstrapping methodology. Our findings are that there are substantial changes in VAR as well as in the width of estimated PD confidence intervals.

The main challenge of forecasting credit default risk in loan portfolios may be seen in forecasting the default probabilities and the default correlations. We derive a Merton-style threshold value model for the default probability which treats the asset value of a firm as unknown and uses a factor model instead. In addition, we demonstrate how default correlations can be easily modeled. The empirical analysis is based on a large data set of German firms provided by Deutsche Bundesbank. We find that default probabilities can be forecast given the values of risk drivers known at the point of time at which the forecast is made. In addition, correlations depend on the fit of the estimated default probabilities to the realized default rate for given points in time.

With the rapid development of social economy and information technology, the credit risk and financial risk of my country’s financial enterprises are also facing severe challenges. In financial enterprises, credit is related to the survival of the enterprise. As the business volume and scale of financial enterprises continue to expand, financial risks are correspondingly increased. Therefore, the research on financial enterprise credit and financial risks is of great significance. The research on the credit and financial risks of financial enterprises is helpful to help financial enterprises handle financial risks well and perform evasive operations on them. In addition, it can also enhance the credit awareness of enterprises and reduce the default rate in the financial industry. This paper studies and analyzes the financial enterprise credit and financial risk measurement based on the PSM model. First, it uses the literature method to study the PSM model, corporate credit, financial risk and other theoretical knowledge, and then establish a fuzzy neural network model for risk assessment. And the establishment of a PSM model to conduct a questionnaire survey experiment design, analyze the price sensitivity changes and acceptable price ranges under the PSM model, and get the optimal pricing of new financial products issued by financial companies. Finally, it analyzes the relationship between the default rate of corporate credit and internal finance. The conclusion is that when this financial product is priced at 45 yuan, the proportion of reserved recipients is the largest, reaching 66%; when the price is 75 yuan, the acceptable proportion is 23%, which is the acceptable number of people in the three price ranges. The proportion is the largest; if the price is 100 yuan, the unacceptable proportion is the largest, reaching 45%. This shows that the pricing of a new financial product is directly related to its sales. The reasonableness of the product pricing directly determines whether people are willing to pay for it and accept it.

Support vector machines (SVM) have been extensively used for classification problems in many areas such as gene, text and image recognition. However, SVM have been rarely used to estimate the probability of default (PD) in credit risk. In this paper, we advocate the application of SVM, rather than the popular logistic regression (LR) method, for the estimation of both corporate and retail PD. Our results indicate that most of the time SVM outperforms LR in terms of classification accuracy for the corporate and retail segments. We propose a new wrapper feature selection based on maximizing the distance of the support vectors from the separating hyperplane and apply it to identify the main PD drivers. We used three datasets to test the PD estimation, containing (1) retail obligors from Germany, (2) corporate obligors from Eastern Europe, and (3) corporate obligors from Poland. Total assets, total liabilities, and sales are identified as frequent default drivers for the corporate datasets, whereas current account status and duration of the current account are frequent default drivers for the retail dataset.

In this study, inspired by the Credit Portfolio View approach, we intend to develop an econometric credit risk model to estimate credit loss distributions of Turkish Banking System under baseline and stress macro scenarios, by substituting default rates with non-performing loan (NPL) ratios. Since customer number based historical default rates are not available for the whole Turkish banking system’s credit portfolio, we used NPL ratios as dependent variable instead of default rates, a common practice for many countries where historical default rates are not available. Although, there are many problems in using NPL ratios as default rates such as underestimating portfolio losses as a result of totally non-homogeneous total credit portfolios and transferring non-performing loans to asset management companies from banks’ balance sheets, our aim is to underline and limit some ignored problems using accounting based NPL ratios as default rates in macroeconomic credit risk modeling. Developed models confirm the strong statistical relationship between systematic component of credit risk and macroeconomic variables in Turkey. Stress test results also are compatible with the past experiences.