Credit Ratings: What They Are, and What They Are Not

That sentence is the whole article, and it is worth stating at the top because ratings occupy an unusual position in finance: they are opinions with the force of infrastructure. Institutional mandates are written around them, regulatory capital rules have referenced them, index eligibility depends on them, and — as the previous article described — a single notch can determine who is legally permitted to own a bond. So a reader needs two things: to understand what the scales mean and how the industry works, and to hold clearly in mind what ratings systematically do not tell you. The 2008 record is part of this story, and this article reports it rather than skirting it.

The scales, the agencies, and what a rating actually assesses

Three agencies dominate globally, with a number of smaller and regional firms alongside them, and their scales are similar but not identical. In the most familiar notation, the ladder runs from AAA (or Aaa) at the top through AA, A, and BBB — the investment-grade range — and then BB, B, CCC, CC, C, down to ratings denoting default. Within most categories sit notches (AA+, AA, AA− and so on) allowing finer gradation, and the BBB−/Baa3 line is the investment-grade boundary from the previous article. Two elements accompany the letter and are routinely ignored: the outlook (positive, stable, negative — the agency's indication of likely direction over a medium horizon) and watch or review status (signalling a near-term reassessment, often tied to a specific event such as an acquisition). A downgrade is usually preceded by one or both, so the letter alone discards forward-looking information. Now the crucial definitional point: a rating assesses creditworthiness, and nothing else. Specifically, an agency rating is an opinion on the likelihood of default and, in some scales, expected loss given default — it is not an assessment of whether a bond's price will fall, whether its yield is adequate compensation, whether it is liquid, or whether it suits any investor. A AAA-rated thirty-year bond can lose a third of its market value on a rate move with its rating entirely unchanged, because interest-rate risk is simply not what ratings measure. Conflating the two is the most common misreading, and it is the reason this pillar separates credit risk from rate risk so insistently. Note also that ratings apply at different levels — issuer ratings versus issue-specific ratings — and the latter can differ sharply from the former depending on seniority and security, so "the company is rated A" does not tell you the rating of the particular subordinated bond you are looking at.

The business model, its conflict, and the 2008 record

Ratings are produced by commercial firms, and how they are paid matters. The dominant arrangement is the issuer-pays model: the borrower seeking the rating pays the agency that assigns it. The conflict of interest is structural and obvious — the agency's customer is the entity being judged, and issuers can in principle shop between agencies for a favourable assessment. The industry's defences are real and worth stating fairly: reputational capital is an agency's entire asset, analytical staff are separated from commercial staff, methodologies are published, and rating committees rather than individuals decide. The subscriber-pays alternative exists but has never displaced the dominant model, in part because ratings function as public goods once published. Then there is the record. In the run-up to 2008, agencies assigned top ratings to large volumes of structured mortgage securities that subsequently suffered catastrophic losses, and the subsequent official investigations examined both the models used and the incentive structures that produced them; the episode was central to the crisis narrative the market-history pillar covers, and it produced substantial regulatory response — registration and oversight regimes for rating agencies in the US and EU, requirements around methodology disclosure and conflict management, and a deliberate policy effort to reduce mechanical regulatory reliance on ratings. This portal reports that history because a reader who does not know it will over-trust the letter. It also reports the fair counterweight: ratings on ordinary corporate and sovereign bonds have a long track record that broadly orders default risk correctly — higher-rated issuers default less often than lower-rated ones, consistently, across decades and cycles. Both things are true. Ratings are useful, imperfect, occasionally very wrong, and structurally conflicted, and the appropriate response is to use them as one input rather than either dismissing them or delegating judgment to them.

How to use a rating properly

Six practical disciplines follow, all of them modest. Read the whole rating, not the letter: agency, scale, outlook, watch status, issuer-level versus issue-level, and the date it was last affirmed — a rating is a point-in-time opinion that may be stale. Expect ratings to lag. Market prices generally move before ratings change; a bond trading at a yield implying serious distress while still rated investment grade is not a mispricing to exploit but a signal that the market has updated and the agency has not yet. Know that ratings are used mechanically by others, which creates price effects independent of the underlying credit — the fallen-angel forced-selling dynamic from the previous article. Remember they say nothing about rate risk, liquidity, or valuation. Prefer the underlying reasoning where available: agencies publish rationales, and the reasons for a rating are more informative than the letter. And treat ratings as a floor for your own reading, not a substitute for it — the credit questions from the issuers article (cash-flow stability, debt burden, refinancing schedule, covenants) remain the reader's to consider. One closing note on scope: sovereign ratings follow the same scales but assess a different kind of borrower, with the own-currency considerations that article raised, and they are politically sensitive in ways corporate ratings are not — sovereign downgrades have prompted public disputes between governments and agencies, which is itself a reason to treat them as contestable opinions. And ESG and other supplementary scores issued by the same firms are separate products assessing different things entirely; they are not credit ratings, and conflating them is a growing source of confusion. This portal takes no view on any rating, agency, or issuer.