The Rating Is a Regulation

A credit rating is an opinion. A firm reads an issuer's accounts, forms a view about whether the debt will be repaid, and publishes a letter grade.

As an opinion it would be unremarkable. What made it structural is that rules began to reference it, so the grade stopped being advice and became a legal fact.

The provision governing recognised rating organisations formalised which firms' opinions count, which is the moment a private judgement acquired regulatory force.

Once a rule says an institution may hold only investment grade paper, the firm assigning the grade is deciding what that institution may own.

That is a delegation of regulatory authority to a private party, and it happened without ever being described as one.

The competitive structure explains why reform proposals keep failing at the same point.

The payment model is the part everybody knows and nobody has replaced. The issuer of the debt pays the firm that rates it.

Recognition is a regulatory status, so entry requires approval, and approval requires a track record that a new entrant cannot accumulate without being used.

The conflict is obvious and the alternatives are worse in specific ways, which is why it survives. Investor-paid ratings become private goods and stop informing the market.

That produces a small number of firms whose position is protected by the same rules that give their opinions legal force.

The regulations address the conflict through disclosure, separation and conduct rules, all of which are sensible and none of which changes who writes the cheque.

It is a licensed oligopoly created as a side effect of a rule about prudence, and nobody involved set out to create it.

The mechanical consequences are more interesting than the conflict and get far less attention.

Which is why the sharpest criticism is not about the conflict of interest at all. It is that the regulation manufactured the market structure it now has to supervise.

A downgrade does not merely inform. It triggers, because contracts and rules are written against thresholds, so a single ratings action forces sales by institutions that had no view of their own.

Forced selling into a falling market is precisely the dynamic that turns a credit problem into a liquidity problem, and it is created by the referencing rather than by the rating.

So the instrument that was supposed to reduce risk by informing investors instead synchronises their behaviour, which increases it.

The register carries the rulemakings that have tried to reduce this reliance, and the striking feature is how hard removing a reference turns out to be.

A rule referencing a rating is easy to write and enforce. A rule requiring the regulated institution to form its own credit judgement is neither.

Supervisors would then be second-guessing thousands of internal assessments instead of reading one letter, and no supervisor has the capacity for that.

Which is the honest reason the references persist. They are not there because anyone trusts the ratings; they are there because they are administrable.

The published statements record the sovereign side, where the same dynamic applies to governments and the political sensitivity is far higher.

A government downgraded pays more to borrow, and the downgrade rests on a judgement about future political behaviour that no methodology can make objective.

The firms are candid that sovereign ratings involve judgement of that kind. The rules referencing them treat the output as though it were a measurement.

That gap between what the rating claims to be and what the regulation assumes it is remains the unresolved centre of the whole arrangement.

The reform that would work is the unglamorous one: keep the ratings as information and remove them from the rules, accepting that supervision becomes harder and more expensive.

Nobody has been willing to pay that price, so a private opinion continues to carry the force of law, and everyone involved knows it and describes it as something else.