The higher the market value of the assets, the greater the nominal value of the debt, the volatility of the asset s market value and the maturity of the debt. It is further assumed that the bank that granted the loan can cover the credit risk associated with it through the purchase of a put option on the value of the asset, with the same maturity of the loan and exercise price equal to the nominal value or repayment of the debt F in this way, the bank has a guaranteed payoff equal to F, regardless of whether the value of the assets is higher or lower than that of the debt.
It can therefore be concluded that the probability of default is equivalent to the probability of exercising the put option with which the bank hedged its credit risk. Using the formulas for calculating the probability of exercise of an option130, one is then able to calculate the probability of insolvency of the company. However, Merton s model presents clear limits linked to the simplifying hypothesis of a single form of liabilities with repayment of principal and interest in a single maturity solution, as well as the fact that it does not consider the risk of migration but only that of insolvency and, finally, to the fact that some input variables of the model are often not directly observable in the market.
Assuming that the distribution of asset returns is normal, from the distance from the default one can trace the probability of insolvency. However, the Km model is not based on this hypothesis of the normality of the distribution and determines the probability of insolvency by analyzing on a historical basis the percentage of failed companies according to different levels of DD using a very large sample of companies, the frequency with which companies characterized by the same DD have become insolvent during a certain period expected default frequency EDF in this way a sort of rating system is constructed that allows associating to every new company investigated, on the basis of its DD, a certain effective frequency of insolvency which represents the probability of insolvency of the same. For continue: https://queueapps.weebly.com/
The methods discussed so far allow us to estimate the probability of insolvency however, in order to define the expected loss, it is necessary to supplement this estimate with that of the loss rate in case of insolvency or, alternatively, the recovery rate. It is a function of various factors related to the severity of the state of insolvency, the degree of liquidity of the assets held by the company, the presence or absence of guarantees, the time required for recovery but also to the productive sector of the company, to its legal location and general economic performance. An extremely large sample would therefore be needed to construct recovery rate estimates that jointly consider all these characteristics for this reason, typically, it is based on the characteristics of the single transaction or, alternatively, on those connected to the sector to which it belongs.
As for the estimate of the probability of insolvency, also that of the recovery rate can be made in different ways. In fact, it can be based on internal evaluations of individual banks, conducted on the basis of the historical experience of the loan portfolio or on the use of data obtained from the bond market134, even if the second option is not adequate to estimate the recovery rates of traditional loans. Banking, being the bonds on which this approach is based, issued by large companies, radically different from those that represent the main share of the credit portfolio of a commercial bank.
In estimating the recovery rate it is important to take into account that it is influenced not only by the percentage of recoverable credit, but also from the financial cost related to the time elapsed from the moment of insolvency to that of the actual recovery, as well as from the administrative costs incurred by the bank for recovery135. All these factors represent click link https://danielladickinson.cabanova.com/ random variables and are at the origin of the risk of recovery. The unexpected loss is connected to the possibility that the estimated random variables to determine the expected loss expected exposure, probability of insolvency and loss rate in case of insolvency may prove, a posteriori, higher than the bank s forecast.
It is therefore necessary to estimate the potential variation that this loss can undergo with respect to expectations, within a specific time horizon137 and with a given confidence level138, with reference to the entire portfolio of credit exposures, to get to estimate the value at risk of the same. The different models proposed for estimating portfolio risk are all based on the determination of the maximum potential loss that a portfolio of exposures can undergo over the defined time horizon and with a specific level of confidence, with the same logic of the VaR models developed. For market risks. The best known model is that proposed by the JP Morgan bank and called CreditMetricsTM.
It is based on the analysis of the migration phenomenon, in the deterioration of the creditworthiness of the counter parties, and it is proposed to estimate the distribution at maturity of the changes that the market value of a portfolio of credit exposures suffers, generally within the one year. In particular, the model under examination consists of six distinct phases, the first five relating to the single exposure and the last relating to the estimate of portfolio risk.
The first phase consists in estimating the value of each exposure which in the simplest case corresponds to the value of the loan granted each of them is then associated to a rating class internal or external and the probability of migration from one class to another is estimated also considering the insolvency as one of the credit states towards which it is possible to migrate, thus building a transition matrix139 relative to the reference time period. The third phase consists of the estimate of the recovery rate, for which the model proposes to use the market values of the bonds after the insolvency of the related issuers, although it is still possible to use internal bank data deriving from the historical analysis of the portfolio.
It can therefore be concluded that the probability of default is equivalent to the probability of exercising the put option with which the bank hedged its credit risk. Using the formulas for calculating the probability of exercise of an option130, one is then able to calculate the probability of insolvency of the company. However, Merton s model presents clear limits linked to the simplifying hypothesis of a single form of liabilities with repayment of principal and interest in a single maturity solution, as well as the fact that it does not consider the risk of migration but only that of insolvency and, finally, to the fact that some input variables of the model are often not directly observable in the market.
Assuming that the distribution of asset returns is normal, from the distance from the default one can trace the probability of insolvency. However, the Km model is not based on this hypothesis of the normality of the distribution and determines the probability of insolvency by analyzing on a historical basis the percentage of failed companies according to different levels of DD using a very large sample of companies, the frequency with which companies characterized by the same DD have become insolvent during a certain period expected default frequency EDF in this way a sort of rating system is constructed that allows associating to every new company investigated, on the basis of its DD, a certain effective frequency of insolvency which represents the probability of insolvency of the same. For continue: https://queueapps.weebly.com/
The methods discussed so far allow us to estimate the probability of insolvency however, in order to define the expected loss, it is necessary to supplement this estimate with that of the loss rate in case of insolvency or, alternatively, the recovery rate. It is a function of various factors related to the severity of the state of insolvency, the degree of liquidity of the assets held by the company, the presence or absence of guarantees, the time required for recovery but also to the productive sector of the company, to its legal location and general economic performance. An extremely large sample would therefore be needed to construct recovery rate estimates that jointly consider all these characteristics for this reason, typically, it is based on the characteristics of the single transaction or, alternatively, on those connected to the sector to which it belongs.
As for the estimate of the probability of insolvency, also that of the recovery rate can be made in different ways. In fact, it can be based on internal evaluations of individual banks, conducted on the basis of the historical experience of the loan portfolio or on the use of data obtained from the bond market134, even if the second option is not adequate to estimate the recovery rates of traditional loans. Banking, being the bonds on which this approach is based, issued by large companies, radically different from those that represent the main share of the credit portfolio of a commercial bank.
In estimating the recovery rate it is important to take into account that it is influenced not only by the percentage of recoverable credit, but also from the financial cost related to the time elapsed from the moment of insolvency to that of the actual recovery, as well as from the administrative costs incurred by the bank for recovery135. All these factors represent click link https://danielladickinson.cabanova.com/ random variables and are at the origin of the risk of recovery. The unexpected loss is connected to the possibility that the estimated random variables to determine the expected loss expected exposure, probability of insolvency and loss rate in case of insolvency may prove, a posteriori, higher than the bank s forecast.
It is therefore necessary to estimate the potential variation that this loss can undergo with respect to expectations, within a specific time horizon137 and with a given confidence level138, with reference to the entire portfolio of credit exposures, to get to estimate the value at risk of the same. The different models proposed for estimating portfolio risk are all based on the determination of the maximum potential loss that a portfolio of exposures can undergo over the defined time horizon and with a specific level of confidence, with the same logic of the VaR models developed. For market risks. The best known model is that proposed by the JP Morgan bank and called CreditMetricsTM.
It is based on the analysis of the migration phenomenon, in the deterioration of the creditworthiness of the counter parties, and it is proposed to estimate the distribution at maturity of the changes that the market value of a portfolio of credit exposures suffers, generally within the one year. In particular, the model under examination consists of six distinct phases, the first five relating to the single exposure and the last relating to the estimate of portfolio risk.
The first phase consists in estimating the value of each exposure which in the simplest case corresponds to the value of the loan granted each of them is then associated to a rating class internal or external and the probability of migration from one class to another is estimated also considering the insolvency as one of the credit states towards which it is possible to migrate, thus building a transition matrix139 relative to the reference time period. The third phase consists of the estimate of the recovery rate, for which the model proposes to use the market values of the bonds after the insolvency of the related issuers, although it is still possible to use internal bank data deriving from the historical analysis of the portfolio.
No comments:
Post a Comment