CRISP Research

Credit Rating and Investors Services Philippines, Inc. (CRISP) aims to contribute a valuable local perspective to the study of the probability of default and credit ratings within the Philippine context. 

CRISP’s research typically focuses on adapting global rating methodologies to local market conditions, analyzing default rates among Philippine corporate issuers, and calibrating rating scales to reflect the country’s unique risk environment.  Their studies often highlight the importance of macroeconomic factors, regulatory developments, and sectoral characteristics that may influence default probabilities in emerging markets such as the Philippines.  By providing empirical data and contextual analysis, CRISP’s work enhances understanding of default risk and supports the development of sound credit risk management practices among domestic investors and issuers.

The accurate measurement of probability of default (PD) is central to effective credit risk management in the global financial system.  Credit ratings, as assigned by agencies such as Moody’s Investors Service (Moody’s), Standard & Poor’s (S&P), and Fitch Ratings (Fitch), are widely used benchmarks for assessing the creditworthiness of issuers and debt instruments.  Understanding how these agencies estimate PD and how their ratings relate to default risk is essential for finance professionals, regulators, and academics.  This paper examines the methodologies used by Moody’s, S&P, and Fitch to estimate PD, explores the relationship between credit ratings and default risk, and provides a comparative analysis of the agencies’ approaches.  The discussion aims to provide a comprehensive and objective overview to inform both academic study and practical application.

Overview of Credit Ratings

Definitions and Purpose

Credit ratings are opinions provided by rating agencies regarding the relative credit risk of an entity or financial instrument.  These ratings reflect the agencies’ assessment of an issuer’s ability and willingness to meet its financial obligations in full and on time.  Ratings are not guarantees of default or loss but serve as standardized measures to facilitate market transparency and comparability.

Rating Scales

The three major rating agencies—Moody’s, S&P, and Fitch—employ similar but distinct alphanumeric scales to indicate credit quality:

• Moody’s: Ranges from Aaa (highest quality) to C (lowest), with intermediate grades such as Aa, A, Baa, Ba, B, Caa, and Ca.  Modifiers “1,” “2,” and “3” further refine grades.
• S&P and Fitch: Both use AAA (highest) to D (default), with categories such as AA, A, BBB, BB, B, CCC, CC, and C. “+” and “−” modifiers are used for further granularity.

Investment grade ratings are typically Baa3/BBB− or higher, while ratings below this threshold are considered speculative or non-investment grade.

The Role of Moody’s, S&P, and Fitch

Moody’s, S&P, and Fitch are globally recognized as Nationally Recognized Statistical Rating Organizations (NRSROs).  Their ratings inform investment decisions, regulatory capital requirements, pricing of debt instruments, and risk modeling across the financial industry.

Probability of Default (PD): Conceptual Framework

Probability of default (PD) is the likelihood that a borrower will fail to meet its debt obligations within a specified time horizon, typically one year.  PD is a fundamental component of credit risk quantification and is used in portfolio risk assessments, capital allocation, and regulatory compliance (e.g., Basel III capital requirements).

The estimation of PD is critical for pricing credit products, managing loan loss provisions, and informing risk-based decision-making. PD can be expressed as a percentage, reflecting the expected default rate for a given rating or risk segment.

Methodologies for Measuring Probability of Default

Statistical Models

Rating agencies and financial institutions employ various statistical models to estimate PD. Key methodologies include:

• Logistic Regression and Discriminant Analysis: Statistical techniques that predict default probability based on financial ratios, macroeconomic variables, and issuer characteristics.
• Survival Analysis: Models time-to-event data, estimating the probability that a default occurs within a given timeframe.
• Machine Learning Approaches: Increasingly, agencies and institutions explore advanced algorithms (e.g., decision trees, neural networks) to refine PD predictions, though transparency remains a concern.

Historical Default Rates

A core approach to PD estimation involves analyzing realized default rates over time for each rating category.  Rating agencies maintain extensive databases of historical defaults, allowing them to compute average annual default rates by rating grade.  These empirical rates are utilized as reference points for associating ratings with probability of default (PD), and are routinely revised to incorporate the latest available data.

Credit Ratings and Probability of Default

Mapping Ratings to PD

Each rating agency publishes historical default studies that map credit ratings to observed PDs.  For example, issuers rated AAA/Aaa by S&P, Fitch, or Moody’s exhibit extremely low historical default rates, while those rated B or lower show significantly higher probabilities.  Mapping involves aggregating default data by rating and calculating empirical default frequencies, usually over one-, three-, five-, and ten-year horizons.

Empirical Evidence and Theoretical Underpinnings

Empirical studies consistently demonstrate a monotonic relationship between credit ratings and default risk: as ratings decline, observed PD rises.  This relationship validates the predictive power of credit ratings and supports their use in risk-based decision frameworks.  Theoretically, the mapping of ratings to PD is underpinned by the rating agencies’ internal methodologies, which combine quantitative models, expert judgment, and qualitative factors.

Comparative Analysis of Agency Approaches

Moody’s

Moody’s methodology emphasizes a combination of quantitative analysis (financial ratios, leverage, and liquidity) and qualitative assessment (governance and industry outlook). Moody’s publishes annual default studies detailing historical default rates by rating class.  The agency’s PD estimates are based on observed default frequencies, adjusted for rating migrations and macroeconomic factors.

Standard & Poor’s (S&P)

S&P’s approach similarly blends quantitative modeling with qualitative analysis, focusing on business risk, financial risk, and management quality.  S&P provides comprehensive default and transition studies, which form the basis of its PD estimates.  S&P’s historical data set is one of the most extensive, enabling robust empirical mapping of ratings to PD.

Fitch Ratings

Fitch employs a framework that integrates sector-specific analysis, financial modeling, and scenario-based stress testing.  Fitch’s default studies and transition matrices inform its PD estimates, with an emphasis on transparency in methodology and regular updates to reflect market developments.

Key Differences and Similarities

• All three agencies rely on historical default data and transition matrices to estimate PD and calibrate their rating scales.
• Each agency applies proprietary models and expert judgment, resulting in occasional rating differences for the same issuer or instrument.
• Methodological transparency varies, with Fitch often emphasizing disclosure, while Moody’s and S&P balance proprietary processes with periodic methodological updates.
• Differences in default definitions, data coverage, and rating philosophies may lead to variations in published PDs for equivalent rating categories.

Credit Rating Scales and Associated Default Rates

Credit ratings are designed to reflect the relative likelihood that an issuer or instrument will default on its obligations. While Moody’s, S&P, and Fitch maintain similar hierarchical scales, each employs distinct nomenclature and rating modifiers. The table below synthesizes typical rating categories and maps them to observed one-year average default rates, as reported in agency research and academic literature.

Moody’s	S&P / Fitch	Description	Empirical One-Year Default Rate (%)
Aaa	AAA	Prime, highest quality	0.00–0.02
Aa	AA	High quality	0.02–0.03
A	A	Upper-medium grade	0.05–0.10
Baa	BBB	Lower-medium grade (investment grade threshold)	0.20–0.30
Ba	BB	Speculative, non-investment grade	0.90–1.20
B	B	Highly speculative	3.00–5.00
Caa–C	CCC–C	Default imminent or in default	10.00–30.00
C	D	Actual default	100.00

Historical Defaults and Credit Rating Assignment

Empirical default studies conducted by Moody’s, S&P, and Fitch consistently demonstrate a positive correlation between declining credit ratings and increasing observed default rates. Over multi-decade data sets, issuers assigned top-tier ratings (Aaa/AAA) exhibit exceedingly low one-year default frequencies, while those assigned speculative or distressed ratings (Ba/BB or lower) display substantially elevated probabilities. These statistics validate the practical utility of credit ratings as proxies for PD, informing risk-weighted asset calculations and pricing models for bonds, loans, and derivatives.

The agencies publish annual transition and default studies, which detail not only default rates, but also rating migrations and recoveries. Such longitudinal analyses enable market participants to calibrate risk models and regulatory capital requirements, in accordance with both historical experience and prevailing market conditions.

Implications for Financial Practice and Regulation

The mapping of credit ratings to observed default rates has profound implications for credit risk management, portfolio construction, and regulatory compliance. Financial institutions rely on agency ratings and historical PD metrics for the estimation of expected losses, credit valuation adjustments, and the allocation of economic capital. Regulators frequently reference agency data in the formulation of capital adequacy standards, such as those promulgated under Basel III. A rigorous understanding of the relationship between ratings and default risk therefore remains vital for sound financial practice and academic inquiry alike.

Conclusion

In summary, credit ratings issued by Moody’s, S&P, and Fitch Ratings provide structured, empirically validated estimates of the probability of default.  The historical linkage between ratings and observed default frequencies underscores their importance in risk management, financial regulation, and investment decision-making.  Ongoing research may further refine these relationships as data availability and analytic techniques advance.

Governance and Oversight

• The Risk Management Committee shall oversee all PD measurement processes, including approval of methodologies and review of periodic validation reports.
• Any model amendments, overrides, or exceptions must be appropriately documented and justified, subject to approval by senior management.
• The Internal Audit function will perform annual reviews of the PD measurement framework to verify ongoing compliance and operational effectiveness.

Review and Updates

This policy will be subject to annual review or as necessary to address changes in regulatory requirements, market conditions, or organizational strategy. Any revisions will be communicated to all relevant stakeholders.