What Is Credit Analysis?

Credit analysis is a systematic investigation and research on the repayment ability of borrowers by commercial banks. The purpose is to prevent the risks that banks may encounter in the process of issuing loans, and to ensure the safety and timely return of bank operating funds. It includes two aspects: Credit risk assessment. That is, through a systematic analysis of the borrower's quality, business ability, loan use, loan amount, source of loan repayment funds, loan repayment time, mortgage quality, etc., the decision is made on whether to loan or not, how much and the conditions of the loan. In order to control the size and structure of the entire bank's loans. Financial analysis. That is, through analysis of the borrower's balance sheet, profit and loss statement, statement of changes in financial position (or cash flow statement) and other schedules, it accurately grasps the borrower's liquidity ratio (including current ratio, quick ratio, cash ratio, Receivables turnover ratio, deposit and loan turnover ratio), leverage ratio (including debt-to-capital ratio, interest expense yield, fixed fee yield, net fixed asset to net capital ratio, dividend payment ratio, etc.), profitability ratio (including Sales growth rate, return on assets, return on shareholders, etc.), based on which to prepare and forecast its financial statements, forecast its possible financial situation and operating risks, and specifically assess its credit risk, and then determine the amount of loans and repayment time. ^[1]

Credit analysis

Credit analysis is a systematic analysis of the debtor's moral character, capital strength, repayment ability, guarantee, and environmental conditions to determine whether to grant a loan and the corresponding loan conditions.

Since the 1980s, affected by the debt crisis, banks in various countries have generally attached importance to the management and prevention of credit risk. Engineering thinking and technology have gradually been applied to the field of credit risk management, resulting in a series of successful quantitative credit risk management. model. Modern credit risk measurement models are divided into three types according to their measured risk levels: one is a measurement model for a single counterparty or issuer, the other is a measurement model at the asset portfolio level, and the third is a measurement model for derivatives.

Counterparty or issuer-level measurement models

(1) Risk measurement model based on option pricing technology.

Merton found that the bank paid the discount d for a loan with a face value of d equal to the payment for a put option with an exercise price d. Therefore, the value of a risky loan is equivalent to the value of a non-default risk loan with a face value of d plus a short put. The call value of a loan depends on five variables, namely the market value of corporate assets, the volatility of the market value of corporate assets, the face value of discounted loans, the remaining term of the loan, and the risk-free interest rate.

Based on the market value of an enterprise and its unobservability of volatility, in 1995 the American KMV company developed the KMV model, which is also known as the expected default frequency model (edf). The model uses the market value and The structural relationship between the market value of assets is used to calculate the market value of corporate assets; the structural relationship between the volatility of corporate assets and the volatility of corporate equity is used to calculate the volatility of corporate assets, and the statistics are at a certain standard deviation level The percentage of companies that go bankrupt within one year is a measure of the probability of default for companies with the same standard deviation.

This model is one of the most widely used credit risk models in practice. The theoretical basis of this model is similar in many respects to the option pricing methods of Black-Scholes (1973), Merton (1974), and Hull and White (1995). The basic idea is that when the value of a company drops to a certain level, the company defaults on its debt. According to relevant analysis, KMV found that the most frequent demarcation point for default is when the value of the company is equal to current liabilities ± 50% of long-term liabilities. With the company's expected value at the future moment and the default point at this time, it can be determined how much the company's value drops when the default point is reached. To reach the point of default, the value of the asset must be reduced by a multiple of the standard deviation of the asset value, which is called the default distance. Default distance = (expected value of assets-default point) / expected value of assets × volatility of asset value. This method has a sufficient theoretical basis and is especially suitable for the credit risk of listed companies.

The advantage of the KMV model is that it associates default with company characteristics rather than the company's initial credit rating, making it more sensitive to changes in the quality of debtors; at the same time, it uses stock prices to measure the expected default probability of listed companies, and therefore market information It can also be reflected in the model, making it have a certain degree of forward-looking, and the model's predictive ability is strong; and because the variables used by the model are market-driven and show greater time variability, the holding period of The choice is more flexible than the credit metric model.

(2) Credit measurement model based on value at risk var.

var refers to the normal market conditions and a given level of confidence, used to evaluate and measure the maximum value loss that financial assets may suffer in a certain period of time. When calculating the var of market risk for financial instruments, the key input variables are the current market price and volatility of financial assets. Due to the lack of liquidity of loans, the market value and volatility of loans cannot be observed.

JP Morgan (1997) developed a credit metrics system that addresses the valuation and risk calculation of non-tradable assets such as loans and private placements. This method calculates the market value of a loan and its volatility based on the borrower's credit rating, credit transfer matrix, default loan recovery rate, and credit risk spreads in the bond market. It infers the var of individual loans or portfolios, and thus provides a Trading assets are valued and credit risk assessed.

The advantage of the credit measurement model is that for the first time, it integrates credit rating transfer, default rate, default recovery rate, and default correlation into a unified framework to measure credit risk. This model is applicable to the risk measurement of credit portfolios such as commercial credit, bonds, loans, loan commitments, letters of credit, and market instruments (swap, forward, etc.). However, the model has the following problems in application: the default rate is directly taken from the average value of historical data, but empirical research shows that the default rate has a direct relationship with macroeconomic conditions, and is not fixed. The research shows that the actual distribution is mostly thick-tailed; the hypothesis that the correlation between corporate asset returns is equal to the correlation between corporate securities returns remains to be verified.

(3) Credit risk + system based on actuarial calculations.

Credit Suisse First Boston (CSFB, 1997) developed the leading idea of the credit risk plus (creditrisk +) system from insurance actuarial science, that is, the loss depends on the frequency of the disaster and the degree of loss or damage caused by the disaster. Analyze the reasons for default, and the model is only for default risk and does not involve transfer risk, which is especially suitable for credit risk analysis of loan portfolios containing a large number of small and medium-sized loans.

This method is based on the assumption that the default of any single loan in the loan portfolio is random; the probability of default for each loan is independent, so this method assumes that the probability distribution of the default of the single loan in the loan portfolio follows the Possion distribution. The advantage of the credit risk additional model is that it only requires limited input data, basically only the loan default rate, default rate volatility and risk exposure of each group in the loan portfolio, so loan losses are easy to calculate.

(4) Creditportfolio View system based on macro simulation.

This credit portfolio view system was developed by Mckinsey (Wilson, 1997), and it is a macroeconomic simulation system of default risk. Because business cycle factors affect the probability of default, McKinsey & Company incorporates cyclical factors into the measurement model. The system processes cyclical factors based on credit metrics, and combines the rating transfer matrix with economic growth rates, unemployment rates, The relationship between macroeconomic variables such as interest rates, exchange rates, and government expenditures is methodified, and the Monte Carlo method is used to simulate the "shock" of cyclical factors to determine the change in rating transfer probability, and to analyze changes in the macroeconomic situation and credit default probability and transfer probability And then analyze the credit risk of borrowers in different industries or sectors with different credit levels.

The advantage of this model is that it incorporates various macro factors that affect the probability of default and changes in credit rating into its own system, and gives a specific loss distribution, which can characterize the uncertainty of the recovery rate and the risk due to national risks. Loss; all market exposures are market-marked, which is more suitable for credit risk measurement of a single debtor and a group of debtors. It is mainly applicable to the credit risk measurement of speculative debtors who are sensitive to changes in macroeconomic factors.

Econometric model of portfolio level

Modern asset portfolio theory (MPT) shows that the proper use of the correlation between assets can effectively reduce risks and improve the risk-return status of asset portfolios. However, poorly liquid loan and bond portfolios have problems with non-normal returns, unobservable returns and correlation coefficients, etc., which makes asset portfolio theory simply not applicable to these portfolios. The non-normality of returns makes the portfolio theory based on the two moments (mean and variance) can be better described only by increasing the skewness and kurtosis. The lack of historical price and transaction data makes it extremely difficult to calculate returns, variances, and covariances and correlation coefficients between returns using historical time series data. Credit risk measurement models at the portfolio level have been developed by overcoming these problems. Such models can be broadly divided into two categories: one is seeking to calculate the total risk-return alternating relationship of the portfolio, such as KMV's asset portfolio management model; the other is the var calculation that focuses on the risk dimension and the combination, such as Creditmetrics portfolio model.

Credit risk measurement model of derivatives

Derivatives can be divided into interest rate derivatives and credit derivatives. The former can be divided into symmetrical derivatives according to their risk-return characteristics, mainly referring to forwards, futures and swaps, while options are asymmetric derivatives, and their risk-return characteristics show a typical nonlinearity. The latter mainly uses the decomposition and combination technology to change the overall risk characteristics of assets, such as credit swaps, credit options, and credit forwards.

There are many differences between the credit risk of derivatives and on-balance sheet business. First, the non-default value of the contract must be negative for the counterparty; second, the counterparty must be in financial distress; and again, at any level of default probability, the settlement of derivative instruments generally adopts a rollover method. Losses are often lower than the default losses on loans of the same amount; in the end, banks and other financial institutions use many other mechanisms to reduce the probability and loss of default. In view of this, researchers have successively proposed many measurement models, but mainly focus on two types of derivatives: swaps and options.

The pricing of credit derivatives is a difficult issue in the field of credit risk management research. At present, there are mainly three types of methods for pricing credit derivatives in academia and practice: pricing based on insurance theory, pricing based on replication technology, and pricing based on stochastic models. In the pricing method based on insurance theory, the insurance company bears the credit risk of the insured, so it must get a certain premium as compensation. This pricing method is a statistical method based on the historical default database of insurance companies. It has a narrow application range and can only provide insurance for credit derivatives with historical default data. However, the pricing based on replication technology needs to determine the value of all positions in the portfolio one by one. For complex derivatives, this technology is difficult to achieve. Stochastic model-based pricing is the current mainstream direction, of which strength models and hybrid models are widely used.