Paper: arXiv 2604.20877

Authors: Marco Pollanen

Abstract

A credit rating of AAA asserts near-certainty of repayment. This paper asks whether the pre-crisis information environment could have supported that assertion for structured products. Bayes’ theorem implies that any reliability target requires a minimum level of statistical discrimination between instruments that will repay and those that will not. At structured-finance base rates, a four-nines reliability target demands discrimination on the order of 10,000 to 1. A three-nines target demands 1,000 to 1. Nothing in the published credit-prediction literature provides an affirmative basis for believing that discrimination of this magnitude was achievable with the data available at rating time. Retrospectively, the realized system fell short of the four-nines benchmark by roughly 90,000-fold. The framework accommodates the historical feasibility of corporate AAA ratings, where high base rates and rich information produce low required discrimination. Illustrative calibrations for contemporary collateralized loan obligations suggest that material tension between the precision target and the information environment persists. The central implication is that the AAA precision claim itself likely exceeded what the available information could support.

Complexity vs Empirical Score

  • Math Complexity: 6.5/10
  • Empirical Rigor: 6.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper provides a novel, elegant Bayesian theoretical framework proving information-theoretic impossibility bounds for structured credit ratings. It backs these theoretical bounds with solid historical calibration data from pre-crisis CDOs and modern CLOs, offering strong clarity and conceptual insight.

Research Flowchart

  flowchart TD
    A["Research Question: Could pre-crisis data support AAA reliability for structured products?"] --> B["Apply Bayes' Theorem & Reliability Targets"]
    
    C["Inputs: Base Rates & Historical Data (Structured vs. Corporate)"] --> B
    
    B --> D["Calculate Required Statistical Discrimination (Likelihood Ratio)"]
    
    D --> E["Compare Required vs. Literature-Supported Discrimination"]
    
    E --> F1["Historical Corporate AAA: Achievable (High Base Rates, Rich Data)"]
    E --> F2["Structured AAA (Pre-Crisis): Failed (~90,000-fold Shortfall)"]
    E --> F3["Contemporary CLOs: Material Tension Persists"]

    style A fill:#f9f,stroke:#333,stroke-width:2px
    style F2 fill:#ff9999,stroke:#333,stroke-width:1px
    style F1 fill:#99ff99,stroke:#333,stroke-width:1px
    style F3 fill:#ffff99,stroke:#333,stroke-width:1px