S&P Has No Special Magic for Judging U.S. Creditworthiness

On Friday afternoon, as everyone on the planet has heard by now, Standard and Poor’s downgraded the United States Government to AA+, worried about the political “brinksmanship” in Washington and the fact that “the differences between political parties have proven to be extraordinarily difficult to bridge.” I meet a lot of people who have the same opinion. Does S&P know anything they don’t?

S&P, like Moody’s and Fitch, certainly could not have acquired expertise through a long history of rating sovereign issuers. As recently as 1974, the only two countries S&P rated were the United States and Canada. By 1985, the number of rated countries was up to 14. For perspective, S&P has been rating a large universe of corporate and municipal securities since 1922.

Nor could S&P have learned how to predict sovereign defaults by calibrating models to reams of data. Since S&P entered this business, not a single developed country has defaulted. After Argentina defaulted 10 years ago, sovereign defaults have been limited to Moldova, Uruguay, Dominican Republic, Belize and Ecuador. 

Perhaps the process S&P uses is so distinguished that it commands our respect. You can reject that hypothesis by skimming "Sovereign Government Rating Methodology and Assumptions" on S&P’s website. You will find a hodgepodge of pretentious, untested and arbitrary indicators that S&P amalgamates through secret formulas and secret committees. For instance, a country gets a score from 1 to 6 that ranks its political effectiveness. Imagine the arrogance of S&P staffers sitting in a room evaluating “the strength and stability of the [U.S.] government's institutions, and the effectiveness of its policy-making.”

Finally, sovereign ratings might deserve the status the market has accorded them because rating agencies performed effectively in other areas of their business. Undeniably, corporate ratings have an impressive track record. From 1983 to 2010, the chance an issuer rated Aaa by Moody’s would default at any time over the subsequent 10 years was 0.19%; lower rated corporates have generally defaulted more frequently: Aa1: 0.16%; Aa2: 0.72%; Aa3: 0.50%; A1: 1.47%; A2: 2.64%; A3: 2.50%; Baa1: 2.47%; Baa2: 5.10%; Baa3: 7.74%; Ba1: 13.97%; Ba2: 15.24%; Ba3: 32.31%; B1: 36.69%;B2: 40.74%; B3: 56.15%. The results for S&P and Fitch are similar.  (All three rating agencies are careful to report the number of rated issuers divided by the number of defaulting issuers. The dollar-weighted results would be skewed by mishaps like Lehman and Wamu. This would show, unhelpfully, that default rates for A1 are higher than double-B.)  

But does this expertise extend to issuers beyond corporates? The disgraceful experience with structured finance gives a strong piece of evidence that it does not. Worse, we can draw no comfort from the failure of all three agencies to foretell the strains in peripheral Europe. Ireland was rated Aa2 by Moody’s and AA by S&P in August 2010 and triple-A by both in March 2009. Or how about this press release headline: “Moody's changes outlook on Greece’s A1 bond ratings to positive.” That was January 2007. Or this one: “Moody’s: Investor Fears Over Greek Government Liquidity Misplaced.” That was December 2009. Even if we could convince ourselves that rating agencies had insight into emerging countries, it’s a big stretch to suppose this translates to superpowers.

Rating agencies have inexplicably taken hold of the global psyche. Here’s a modest proposal to fight back.

First, from now on, let’s call them “ratings companies” rather than “agencies.”  (Their official designation in exchange for submitting to a tiny amount of SEC oversight is “Nationally recognized statistical rating organizations.”)

Part two of my proposal is to remind ourselves as often as possible that rating big countries is thoroughly ridiculous. This is not to say that big countries can’t default—of course they can, and someday even the U.S. might. But we can all look at the same budget numbers and assess the political prospects as well as S&P. Like forecasting stock market indices, no one has any special magic. 

The first sentence of S&P's “rationale” for Friday’s downgrade illustrates the absurdity of this endeavor. The rating company frets that the debate in Congress on spending and revenues “will remain a contentious and fitful process.” Yet Congress has been contentious and fitful for some time now — it’s a good thing S&P didn’t witness the debate on war with Britain in 1812 — and nonetheless for 235 years the United States has muddled through.

Richard Robb is chief executive of Christofferson, Robb & Co., an investment management firm, and professor of professional practice of international finance at Columbia University’s School of International and Public Affairs.