Luisa Izzi

CDS-implied EDF Credit Measures and Fair-value Spreads

This chapter presents a framework that links default probabilities and credit spreads. This framework provides credit default swap-implied (CDS-implied) EDF (expected default frequency) credit measures that can be compared directly with equity-based EDF credit measures. The model also provides equity-based fair-value CDS spreads (FVS) which can be compared directly with observed CDS spreads.

Introduction: The Efficient Market Hypothesis and Basel III New Banking Regulations

The severity and amplitude of the recent global crisis provide between them convincing evidence that there is something fundamentally wrong with the prevailing theory on how financial markets work, and with the approach to market regulation that has accompanied it. Understanding what has happened and what should be done to avoid such a catastrophic crisis in the future will require a new way of thinking about how markets work.

Exposure at Default Valuation

The second risk parameter, required as input for the IRB approach to credit risk management and the calculation of regulatory capital, is represented by the ‘exposure at default’ (EAD), defined as the expected amount of a commitment (both in terms of on-/off-balance-sheet exposures) at the time of default of a customer.

Corporate SME and Retail PD models

The rating models represent relevant decision support tools within the processes of origination and monitoring of the credit granted by a bank to its customers.

Pricing in Liquid Markets

The degree of efficiency of the equity and fixed income markets is provided in the well-known joint behaviour of equity premiums and credit spreads on securities. The equity and fixed income markets have been profitable fields of research (both theoretical and practical) for a long time, while the CDS market has increased its liquidity — and therefore the need for proper tools for analysis — only in recent years. It has been shown that equity-implied volatility movements explain significantly spread movements (both theoretical and actual). Moreover, the theoretical spreads are a reliable proxy for forthcoming movement in actual spreads. This is a general result based on the seminal paper by Merton on structural default compared to CDS spreads. The Merton model derives a theoretical, implied credit spread, having as inputs, among others, equity-implied volatility, which can be compared with observable CDS spreads.

Validation of Internal Credit Models

International supervisory regulations require that the risk parameters estimated by banks — using an internal rating based (IRB) methodology — satisfy a series of (minimal) requirements to ensure their reliability and robustness over time.1

Sovereign and Banks’ Rating Models

In general, there are three main methodologies, summarized in Table 3.1, which can be used to develop a PD model: ■ good/bad analysis — illustrated in Chapter 2 and applied most notably to corporate SME and retail counterparts; ■ the pure expert ranking method — used, typically, for the development of large corporate models; and ■ the shadow rating approach — specific to the segments characterized by a limited number of defaults, such as the large corporate portfolio, but with the difference being that they are constituted by counterparts largely provided by an official rating assigned by an external agency (such as Standard & Poor’s, Moody’s, Fitch and so on).

Loss Given Default Estimation

Fundamental to credit risk management and the calculation of regulatory capital under the IRB approach is the ‘loss given default’ (LGD), which represents the loss experienced if a borrower defaults. In principle, supervisors do not require any specific technique for LGD estimation (or for estimating other IRB parameters); however, organizations will have to demonstrate that the methods they choose are appropriate to the institution’s activities and the portfolios to which they apply. The four main alternatives are ‘workout LGD’, ‘market LGD’, ‘implied market LGD’, and ‘implied historical LGD’. The latter two techniques are considered to be implicit because they are not based directly on the realized LGD of defaulted facilities; moreover, the implied historical LGD technique is allowed only for the retail exposure class.

Pricing in Nonliquid Markets

Default risk is the uncertainty surrounding a firm’s ability to honor its debts and obligations. Prior to default, there is no way to know for certain which firms will default and which will not, so assessments can only be made on the likelihood of default. As a consequence, firms generally pay a spread over the default-free rate of interest that is proportional to their default probability to compensate lenders for this uncertainty. Default is a deceptively rare event; however, there is considerable variation in default probabilities across firms.1 The loss suffered by a lender or counterparty in the event of default is usually significant and is determined largely by the details of the particular contract or obligation. For example, typical loss rates in the event of default for senior secured bonds, subordinated bonds and zero coupon bonds are 49, 68 and 81 percent, respectively. As in other rare events with high costs, default risk can only be managed effectively in a portfolio. In addition to knowing the default probability and loss given default, the portfolio management of default risk requires the measurement of default correlations. Correlations measure the degree to which the default risks of the various borrowers and counterparties in the portfolio are related.

Expert Judgment-based Rating Assignment Process

Broadly speaking, a rating is an assessment, for a given time horizon, of an obligor’s ability to honor its contractual obligations. External agency ratings, as well as banks’ judgmental rating grades, are usually ordinal measures of credit risk (as opposed to, for example, KMV Moody’s ‘expected default frequencies’ — EDFs), which have been determined by taking into account all relevant available information (both quantitative and qualitative). The ordinality allows for the ranking of obligors in terms of relative riskiness. To quantify obligors’ credit risk, probabilities of default are estimated for each rating category; the riskier a rating category is, the higher its PD estimate should be.