Sovereign credit rating reviews

What a rating is, and what it is not 

A credit rating is an opinion about the likelihood that a borrower will pay what it owes, on time and in full, expressed on an ordinal letter scale. Three agencies dominate the sovereign market globally, and several regional agencies operate alongside them. Each publishes its own methodology, its own scale and its own definitions of what each grade means.

Three things a rating is not are worth stating plainly. It is not a measure of the value of a bond, since a correctly priced low-rated bond and a correctly priced high-rated bond are both correctly priced. It is not a probability, since the scale is ordinal and the agencies publish historical default rates by grade as evidence rather than as a definition. And in most jurisdictions it is legally an opinion, protected as such, which is a position the agencies have defended in litigation and which shapes how their output is worded.

A sovereign rating also carries a feature no corporate rating has. A government that borrows in its own currency and controls the issuance of that currency can always produce the money to repay in nominal terms, so a default in local currency is a choice rather than an incapacity. Agencies therefore rate local currency and foreign currency debt separately, and the distinction between capacity and willingness is explicit in the sovereign methodologies in a way it is not elsewhere.

The scale, the outlook and the watch 

The letter grade is only the first of three pieces of information an agency publishes about a sovereign. The outlook indicates the likely direction of the rating over a medium horizon and is described as positive, stable or negative. A credit watch, or review for downgrade or upgrade, is a shorter and more urgent designation applied when a specific event has put the rating in question and a resolution is expected within a defined period.

This layering means a scheduled review has more possible outcomes than a reader might assume, and most of them do not change the letter. An affirmation with an unchanged outlook, an affirmation with a revised outlook, a placement on watch, a resolution of an existing watch, and an actual change of grade are all outcomes of the same calendar entry. Outlook changes are considerably more common than rating changes, and analysts treat a shift from stable to negative as the substantive information in most reviews.

The sovereign rating additionally functions as a reference point for the borrowers inside a country. Under the older convention a company could not be rated above the sovereign of its domicile, on the reasoning that a government in distress can impose capital controls or taxation that reach any domestic issuer. That ceiling has been relaxed and is now applied with exceptions rather than absolutely, but the relationship persists in practice, so a sovereign action frequently carries consequences for banks, utilities and large corporates that were not themselves under review.

Why one boundary matters more than the others 

One step on the scale is different in kind from the others: the boundary between investment grade and the speculative grades below it. The difference is not analytical, since the two adjacent grades describe similar credit quality. It is mechanical. Investment mandates, regulatory capital rules, central bank collateral frameworks and bond index membership criteria are frequently written with that boundary in them, so a downgrade across it can force sales by holders whose rules do not permit them to hold the instrument, regardless of their own view.

Index rules commonly require agreement between agencies before a bond leaves an index, so the sequencing of decisions between the agencies is itself watched. A first downgrade across the boundary and a second one carry very different mechanical consequences even though each is one notch.

Cadence and publication 

In the European Union, agencies are required to publish an annual calendar of the dates on which they intend to review each rated sovereign, and to publish outside the hours in which European venues are open. Elsewhere the practice varies, but publication after the close is the general convention. Most sovereigns are scheduled for two reviews a year with each agency.

A scheduled date does not oblige an agency to publish. An agency may pass over a review date without announcement if it sees no reason to act, and the absence of a publication on a scheduled date is itself commonly read as an affirmation. Conversely, agencies may act outside the calendar when circumstances require, and such unscheduled actions are permitted under the same rules that impose the calendar.

The agencies also publish rating reports, criteria documents and periodic commentary between reviews. A commentary is not a rating action, and the distinction is not always preserved in secondary reporting.

The business model, stated plainly 

The agencies are paid by the issuers they rate under the prevailing model, which is a conflict of interest that regulators, academics and the agencies themselves have discussed for decades. Regulation introduced after the financial crisis addressed it through disclosure requirements, separation of analytical and commercial functions, rotation rules in some jurisdictions and supervision of methodologies, rather than by replacing the model.

Two further criticisms recur in the literature and are worth knowing. Ratings have been observed to be procyclical, moving after conditions have already deteriorated and thereby reinforcing the move rather than anticipating it. And the reliance on ratings written into regulation and mandates gives a private opinion a quasi-regulatory function, which is the mechanism that turns a one-notch judgement into forced selling. Regulators have worked to reduce hard-wired references to ratings in rules for precisely that reason, with partial success.