Skip to main content

Moody's Slightly Reduces Overrating Of Banks

  • Author:
  • Updated:

Are we supposed to care about these downgrades? I like Glenn Schorr at Nomura, emphasis mine:

We think the net financial impact of these downgrades will be manageable as 1) potential collateral calls are small percentages of these firms’ liquidity pools; 2) counterparties have been preparing for this for some time and ratings downgrades have been an issue for the last 2+ years (there was little impact on Citi and BAC when they were downgraded back in September of 2011); 3) ratings are a relative game: given that Moody’s downgraded all capital markets firms, no single-firm is an outlier, so we don’t expect to see one company uniquely impacted. Yes, we get that counterparties looking to do long-dated derivatives might prefer a single-A rated entity, but as Basel III is implemented and more derivatives move to central clearinghouses, counterparty ratings should become less meaningful and clients will adapt (and not do all their business with JPM and GS).

It would be a serious misinterpretation of credit ratings to think of them as a global rank ordering of risks in the world. "A-rated things are of course safer than BBB-rated things," you say, and get punched in the face repeatedly by life. A-rated things are not safer than BBB-rated things. A-rated RMBS CDOs were not safer than BBB-rated corporates, A-rated corporates are not safer than BBB-rated municipalities, and A-rated banks are it goes without saying not safer than BBB-rated software companies. Nobody really suggests otherwise - if they did, this graph would be a huge embarrassment to Moody's:

What that chart says is that, roughly speaking, over the last thirty years, single-A banks are about as likely to default as single-B companies that are not banks.* And the market is not dumb: bank credit has for a long time traded wider than corporate credit for any given rating. It is a cliché to say that credit ratings are a severely lagging indicator but in this case I'm not sure they're even that. This is less a belated recognition that, for instance, Citi is as risky as BofA (though it is that), and more a nominal nudge of global bank ratings in the direction of other ratings. And a pretty slight nudge at that: to pick some Baa2 names at random, I see BofA 5-year holdco CDS at 276bps yesterday afternoon, Citi at 255, CBS Corporation at 113, Dominion Resources at 57, Kraft Foods at 57, McKesson at 38, Motorola Solutions at 98 ... you get the idea.** The market is not pricing Citi like the A3-rated company it was yesterday, and it is not pricing it like the Baa2-rated company it is today.

It is, though, pricing it like a Baa2-rated bank, since what matters is not the rating but the ordering within the field: A-rated RMBS CDOs really are usually safer than BBB-rated RMBS CDOs, A-rated corporates are usually safer than BBB-rated corporates, and A-rated banks are safer than BBB-rated banks though the predictive value there is less clear. And Moody's did very little to change the ordering within investment banks. This means that fears about so-and-so losing derivatives business to such-and-such, because counterparties really care about having highly rated counterparties, are probably overblown: you can fondle Microsoft, but it won't write you an interest rate swap. You're more or less stuck with a bank, and if banks stop being Aa-ish things and start being Baa-ish things, then you'll be buying your swaps from Baa-ish things.*** Those swaps will be less profitable now that they're more likely to be collateralized - my dopey math had Morgan Stanley lose about $800mm in DVA on a 2-notch downgrade, which happened - but with derivatives lurching toward central clearing anyway that is not a distinguishing feature of these Moody's downgrades.

If you treat these ratings seriously as an indication of changing credit quality then you might be confused at the timing of these downgrades. A better way to look at them is as a redenomination: Moody's recognized that global bank ratings are inflated relative to most other things that they rate, whether measured by market prices or leverage ratios or default probabilities, and took down their ratings a bit to be more in line with everything else. As with any change of currency, there are some friction costs: derivative contracts, investment charters, etc., all are based on the old denominations, so there will be some pain as they catch up to the current reality. But that pain is probably more a transition cost than a real shift in the world. Not that there hasn't been a shift: investors and counterparties probably do recognize that global capital markets banks are not as safe as they once thought, and that they have to give more thought to managing and diversifying their credit risk to those banks. They just didn't figure that out from reading a Moody's downgrade in June 2012.

[Update: counterpoint! I don't really buy that the interbank market is irreversibly keyed to ratings though I take the point that it is near-term keyed to ratings (but consider governmental efforts to delink ratings from capital regulation etc.). Friction costs. Or maybe not.]

Moody’s downgrades firms with global capital markets operations [Moody's via Credit Writedowns]
Bank Investors Dismiss Moody’s Cuts As Years Too Late [Bloomberg]
Moody's Downgrade of Banks ‘Absurd,’ Says Dick Bove [CNBC]

* More detail: the chart is from Figure 6, Panel A of this paper, which we've discussed before. Basically read it as "what percent of defaults within a year, across each asset class, come from this rating or lower?" So the gray solid line for structured products means (eyeballing) that something like 50-60% of defaults in structured products come from paper rated Ba2 or below, and 40-50% come from paper rated Ba1 or better (normalized for number of issues in each bucket so don't take those numbers too seriously). The "right" answer is probably something like the corporate line (the dottiest dots), where there are a lot of defaults for very junky rated paper and very few for high investment grade rated paper and a smooth transition from the one to the other; the "wrong" answer is the randomly-assigned straight line, where Aaa paper is exactly as likely to default as Caa paper. Sample is 1980 through 2010.

** Source is CMA mid prices as of 6/21 via Bloomberg. "At random" is an overstatement; I had my reasons for picking BofA and Citi but the rest came from names in CDX.IG.NA.9, which has been in the news a bit, and I make no claims to representativeness.

*** This is an approximate statement. The thing where Morgan Stanley is structurally more exposed to holdco ratings than its competitors are probably matters, etc. etc., whatever. Also if you look at CDS levels for the big banks they do not particularly closely track Moody's pre- or post-yesterday; Baa1 Morgan Stanley is in the high 300s, A3 Goldman is wider than Baa2 BAC and Citi, etc. Baby steps.