Rating Italy's Debt Hardly Seems Worth the Trouble
Today's looniest financial story is surely that Italy's state auditor, the Corte dei Conti, has opened an investigation into Standard & Poor's, Moody's and Fitch for downgrading Italy's debt over the past couple of years. Loony first of all for the reasoning:
Notifying S&P that it was considering legal action, the Corte dei Conti wrote: “S&P never in its ratings pointed out Italy’s history, art or landscape which, as universally recognised, are the basis of its economic strength.”
But also for the size of the claim: "Standard & Poor’s revealed on Tuesday it had been notified by Corte dei Conti that credit rating agencies may have acted illegally and opened themselves up to damages of €234bn." Two hundred and thirty-four billion euros! Obviously I typed that with my pinky on the corner of my mouth.
That is a silly amount of money! S&P's parent company, McGraw-Hill Financial, has total assets of $6 billion, so you could liquidate it dozens of times and still not pay off those (at this point pretty imaginary) damages.
But it's especially a lot of money compared to how much money S&P got paid for rating Italy's debt, which is, in round numbers, zero. S&P's business model is basically that issuers of bonds come to it and ask it to rate their bonds, and then it does, and it charges them money. There are conflicts of interest in this issuer-pays model, I guess, but at least it's a business model.
But for some issuers, S&P just does unsolicited ratings, which is considerably less of a business model as no one pays for those ratings. It's only a minority of issuers, but it's an important minority. France, Germany, Japan, Japan, France, Italy ... all get unsolicited ratings.1 S&P looks at public information, makes its best guess as to creditworthiness, and slaps a rating on them. For free, and without the sovereign's involvement.
Oh also the U.S.. The U.S. gets an unsolicited, unpaid, public-information-only rating from S&P. In 2011, that rating was reduced from AAA to AA-. The Justice Department later sued S&P for billions of dollars over some structured credit ratings, and S&P is arguing in court that the lawsuit was thinly veiled revenge for that downgrade. So that rating is working out great for S&P?
To review, the P&L for the unsolicited government ratings business is:
- Costs: millions (probably) in legal fees, some bad PR, a billion-dollar lawsuit, and a possible hundred-billion-euro sort-of-lawsuit-esque thing.
- Revenues: none.
Also, I mean, your mileage may vary, but there are those who think that sovereign ratings aren't very important, and that the market tends to ignore sovereign ratings changes, and that bonds tend to move in the opposite direction of what the ratings changes would imply. So it's not clear that those unsolicited ratings are really adding to the agencies' glory.
I dunno, hardly seems worth it? Obviously there are reasons for doing it.2 For one thing, if you're getting paid to rate a lot of sovereigns (and non-sovereigns) for money, it's sort of awkward to have big important gaps. Sure Albania has a B rating and Latvia is BBB+, but what everyone wants to know is, how does that compare to the U.S.? Or Germany? Or Italy, I guess. If you're rating everybody except the most important issuers, your whole ratings project looks a little thin. People tend to ask more about the U.S. and Italy than, you know, Albania.
Related Link: Rating the Raters - QuickTake
Also, the big issuers tend to have lots of other debt issuers in their countries. If you're going to rate U.S. municipalities or German banks or Italian corporates, I suppose it helps to have a rating on the sovereign as a reference point.
Also, everybody likes power and attention, and going around pronouncing on the fiscal future of the U.S. brings a certain cachet that you don't get from analyzing the terms of Belarus's latest bond offering, I guess.
Still, I wonder if all the lawsuits, and the changing rules around sovereign ratings, and the general hostility from the rated sovereigns, will alter that calculation. Cachet and power and completism are good reasons to do something that takes work and doesn't bring you any direct revenue, but I'm not sure they're good reasons to risk hundreds of billions of euros of lawsuits.
1 I was surprised to learn that it really is a minority: S&P rates over 100 sovereigns, of which only a few are unsolicited and unpaid. (They're labeled "(Unsolicited Ratings)" in that list I just linked.) The rest -- including some big ones such as Brazil and Canada -- have ratings agreements. That includes Greece, Spain and Portugal, all of whose ratings seem to be solicited and paid for.
2 I've occasionally suggested that one other purpose for unsolicited ratings is to be really bad so that the issuer will pay for a (better) solicited rating. If you are a cynic, you can read some structured credit ratings that way, but it's hard to imagine that it's true for sovereigns. The U.S. was not going to respond to a downgrade by paying S&P a ratings fee.