Global Credit Review

Volume 2

Risk Management Institute, Singapore

• Publisher: WSPC
• Year: 2012
• Print ISBN: 9789814412636
• eBook ISBN: 9789814412650
• Topic: Biological Sciences
• Subtopic: Science General
• Index: Biological Sciences

This annual publication provides an overview of the most important developments in global credit markets and the regulatory landscape. It covers theoretical and empirical research on credit ratings and credit risk, and reports on recent findings and evolutions of the Risk Management Institute's Credit Research Initiative. The ultimate objective of this publication is to advance the state of research and development in the critical area of credit risk and rating systems. With a distinctive focus on topics related to credit markets and credit risk, this publication will be useful to finance professionals, policy makers and academics with an interest in credit markets.

Contents:

• Credit Markets: Retrospect and Prospect (David Rowe)
• An Improved Regulatory Framework for Credit Rating Agencies? (James Weston)
• Stress Testing (Noel D'Cruz and Davide Crippa)
• Mega-Banks Self-Insurance with Cocos: A Work in Progress (George von Furstenberg)
• What are the Driving Factors behind the Rise of Spreads and CDS of Euro-Sovereign Bonds? (Emmanuel Mamatzakis and Panos Remoundos)
• Measuring Distance-to-Default for Financial and Non-Financial Firms (Jin-Chuan Duan and Tao Wang)
• NUS-RMI Credit Research Initiative Technical report (RMI staff)
• A Lead-Lag Investigation of RMI PD and CRA Ratings (RMI staff)

Readership: Finance professionals, policy makers and academics with an interest in credit markets.

NUS-RMI Credit Research Initiative Technical Report Version: 2012 update 2

INTRODUCTION

This document describes the implementation of the system which the NUS Risk Management Institute’s Credit Research Initiative uses to produce probabilities of default (PDs). As of this version of the Technical Report, these PDs cover exchange listed firms in 44 economies in Asia, Asia-Pacific, North America, Europe and Latin America. Currently, RMI covers over 35,000 listed companies. Of these, over 28,000 firms have sufficient data to release daily updated PDs. The full list of firms is freely available to users who can give evidence of their professional qualifications to ensure that they will not misuse the data. General users who do not request global access are restricted to a list of 2,300 firms. The individual company PD data along with aggregate PDs at the economy and sector level can be accessed at http://rmicri.org

The primary goal of this initiative is to drive research and development in the critical area of credit rating systems. As such, a transparent methodology is essential to this initiative. Having the details of the methodology available to everybody means that there is a base from which suggestions and improvements can be made. The objective of this Technical Report is to provide a full exposition of the CRI system. Readers of this document who have access to the necessary data and who have a sufficient level of technical expertise will be able to implement a similar system on their own. For a full exposition of the conceptual framework of the CRI, see Duan and Van Laere (2012).

The system used by the CRI will evolve as new innovations and enhancements are applied. The most substantial changes to the 2011 technical report and operational implementation of our model are (1) the default definition which now excludes covenant breachesand some default corporate actions that are specific to Taiwan (e.g., bounced checks); (2) priority of financial statements and treatment of net income, with the latter now being included on a quarterly basis when available; (3) treatment of stale market capitalization prices; (4) regrouping of economies for calibration purposes; (5) increased coverage to include Latin America and the rest of the eurozone countries; and (6) treatment of relative size. This version of the technical report provides an update on the operational implementation of the CRI and includes all changes to the system that had been implemented by July 2012. The latest version of the Technical Report is available via the web portal and will include any changes to the system that have been implemented since the publication of this version.

The remainder of this Technical Report is organized as follows. The next section describes the quantitative model that is currently used to compute PDs from the CRI. The model was first described in Duan et al. (2012). The description includes calibration procedures, which are performed on a monthly basis, and individual firm PD computations, which are performed on a daily basis.

Section 2 describes the input variables of the model as well as the data used to produce the variables for input into the model. This model uses both input variables that are common to all firms in an economy and input variables that are firm-specific. Another critical component when calibrating a probability of default estimation system is the default data, and this is also described in this section.

While Section 1 provides a broader description of the model, Section 3 describes the implementation details that are necessary to apply given real world issues of, for example, bad or missing data. The specific technical details needed to develop an operational system are also given, including details on the monthly calibration, daily computation of individual firm PDs and aggregation of the individual firm PDs. Distance-to-default (DTD) in a Merton-type model is one of the firm-specific variables. The calculation for DTD is not the standard one, and has been modified to allow a meaningful computation of the DTD for financial firms. While most academic studies on default prediction exclude financial firms from consideration, it is important to include them given that the financial sector is a critical component in every economy. The calculation for DTD is detailed in this section.

Section 4 shows an empirical analysis for those economies that are currently covered. While the analysis shows excellent results in several economies, there is room for improvement in a few others. This is because, at the CRI’s current stage of development, the economies all use the variables used in the academic study of US firms in Duan et al. (2012). Future development within the CRI will deal with variable selection specific to different economies, and the performance is then expected to improve. Variable selection and other planned developments are discussed in Section 5.

I. MODEL DESCRIPTION

The quantitative model that is currently being used by the CRI is a forward intensity model that was introduced in Duan et al. (2012). This model allows probability of default forecasts to be made at a range of horizons. In the current CRI implementation of this model, PDs are made from a horizon of one month up to a horizon of two years. In other words, for every firm, the probability of that firm defaulting within one month, three months, six months, one year, eighteen months and two years is given. The ability to assess credit quality for different horizons is a useful tool for risk management, credit portfolio management, policy setting and regulatory purposes, since short- and long-term credit risk profiles can differ greatly depending on a firm’s liquidity, debt structures and other factors.

The forward intensity model is a reduced form model in which the probability of default is computed as a function of different input variables. These can be firm-specific or common to all firms within an economy. The other category of default prediction model is the structural model, whereby the corporate structure of a firm is modeled in order to assess the firm’s probability of default.

A similar reduced form model by Duffie et al. (2007) relied on modeling the time series dynamics of the input variables in order to make PD forecasts for different horizons. However, there is little consensus on assumptions for the dynamics of variables such as accounting ratios, and the model output will be highly dependent on these assumptions. In addition, the time series dynamics will be of very high dimension. For example, with the two common variables and two firm-specific variables that Duffie et al. (2007) use, a sample of 10,000 firms gives a dimension of the state variables of 20,002.

Given the complexity in modeling the dynamics of variables such as accounting ratios, this model will be diffcult to implement if different forecast horizons are required. The key innovation of the forward intensity model is that PD for different horizons can be consistently and effciently computed based only on the value of the input variables at the time the prediction is made. Thus, the model specification becomes far more tractable.

Fully specifying a reduced form model includes the specification of the function that computes a PD from the input variables. This function is parameterized, and finding appropriate parameter values is called calibrating the model. The forward intensity model can be calibrated by maximizing a pseudo-likelihood function. The calibration is carried out by economy and all firms within an economy will use the same parameter values along with each firm’s variables in order to compute the firm’s PD.

Subsection 1.1 will describe the modeling framework, including the way PDs are computed based on a set of parameter values for the economy and a set of input variables for a firm. Subsection 1.2 explains how the model can be calibrated.

1.1. Modeling Framework

While the model can be formulated in a continuous time framework, as done in Duan et al. (2012), an operational implementation will require discretization in time. Since the model is more easily understood in discrete time, the following exposition of the model will begin in a discrete time framework.

Variables for default prediction can have vastly different update frequencies. Financial statement data is updated only once a quarter or even once a year, while market data like stock prices are available at frequencies of seconds. A way of compromising between these two extremes is to have a fundamental time period

t of one month in the modeling framework. As will be seen later, this does not preclude updating the PD forecasts on a daily basis. This is important since, for example, large daily changes in a firm’s stock price can signal changes in credit quality even when there is no change in financial statement data.

Thus, for the purposes of calibration and subsequently for computing time series of PD, the input variables at the end of each month will be kept for each firm. The input variables associated with the i^th firm at the end of the n^th month (at time t=n

t) is denoted by X_i(n). This is a vector consisting of two parts: X_i(n)=(W(n), U_i(n)). Here, W(n) is a vector of variables at the end of month n that is common to all firms in the economy and U_i(n) is a vector of variables specific to firm i.

In the forward intensity model, a firm’s default is signaled by a jump in a Poisson process. The probability of a jump in the Poisson process is determined by the intensity of the Poisson process. The forward intensity model draws an explicit dependence of intensities a...

Table of contents

• Cover
• Halftitle
• Copyright
• Message from the Editor
• Contents
• Credit Markets: Retrospect and Prospect
• An Improved Regulatory Framework for Credit Rating Agencies?
• Stress Testing
• Mega-Banks Self-Insurance with Cocos: A Work in Progress
• What are the Driving Factors behind the Rise of Spreads and CDS of 79 Euro-Sovereign Bonds?
• Measuring Distance-to-Default for Financial and Non-Financial Firms
• NUS-RMI Credit Research Initiative Technical report
• A Lead-Lag Investigation of RMI PD and CRA Ratings