Notes from the Vault

Gerald P. Dwyer
July/August 2011

  • Standard & Poor's (S&P) downgrade of the long-term credit rating of the United States has been criticized because the risk of default on U.S. government debt in dollars is zero.
  • U.S. Treasury debt is nominally risk free but not really risk free.
  • While S&P does include a monetary score in its assessment of creditworthiness, a higher inflation rate will earn a lower score, not a higher one.

Critics react to the U.S. downgrade
Standard & Poor's downgraded the credit rating of the United States from AAA to AA+ on August 5. That move generated a great deal of commentary. For instance, Warren Buffett told Fox Business Network the downgrade

"doesn't make sense. … Think about it. The U.S., to my knowledge, owes no money in currency other than the U.S. dollar, which it can print at will. Now if you're talking about inflation, that's a different question." (Claman 2011).

On Meet the Press, Alan Greenspan made a similar observation, saying,

"The United States can pay any debt it has because we can always print money to do that. So there is zero probability of default." (Allen 2011).

Similar comments were made by other distinguished commentators, for example, by the well-known mutual fund manager Bill Miller of Legg Mason (2011).

From one point of view, these observations by Buffett and Greenspan are conclusive. The U.S. government promises to pay its debt in dollar bills and it prints dollar bills at virtually no cost. How could it default? From another point of view, Buffett's and Greenspan's remarks seem like the height of goofiness. Their argument implies that any government is a good credit risk if the government can inflate away its debt by printing money. Indeed, John Plender (2011) noted the incongruity of this argument in the Financial Times with an article titled, "Bond vigilantes focus on ability to print money."1

Why nominally risk-free debt is not really risk-free debt
The incongruity is apparent because financing the debt by printing money could have dramatic effects on inflation. If just the interest on U.S. government debt were financed by increasing the monetary base, the effects would be substantial. In fiscal year 2010, federal outlays for interest on U.S. debt were $196 billion. The monetary base at the end of the fiscal year in September 2010 was about $2 trillion. If a 10 percent increase in the monetary base eventually translated into a similar percentage increase in the money supply and inflation, inflation would go up 10 percent or more. That inflation would amount to a substantial decrease in the value of the dollars paid to bondholders. While not a default in a legal sense, that inflation definitely would make the bondholders' receipts less valuable in terms of what they can buy.

On the flip side, it is easy to see the logic behind Buffett's and Greenspan's remarks. Credit ratings measure credit risk: the risk of payments not being made on a debt. U.S. Treasury securities promise to pay dollars.2 It is hard to imagine a series of events in which the federal government would not pay the dollars it promised.3 If worse comes to worst, the U.S. Bureau of Engraving and Printing can print more bills and make the payments. As a result, U.S. Treasury securities are risk free in terms of dollars. Economists put this another way: U.S. Treasury securities are risk free in nominal terms. In other words, U.S. Treasury debt is nominally risk free.

U.S. Treasury debt is not really risk free, though. Bondholders will be able to buy less than they expected if inflation turns out to be higher than expected. U.S. Treasury debt promises to pay dollars and makes no guarantee about how much those dollars will buy. Higher inflation means that the promised nominal payments will buy less than if inflation were lower. In terms of what bondholders care about—consumption of goods and services—they would receive less for their investment with higher inflation. In terms of what bondholders can buy, U.S. Treasury securities are not risk free. This can also be said other ways. U.S. Treasury securities are not risk free in real terms—in terms of what payments on the securities buy. In sum, U.S. Treasury securities are not really risk free even though they are nominally risk free.

It is hard to see how inflation risk could be considered a credit risk. Inflation risk is a risk, just not a credit risk. Credit risk is the risk of not being paid (S&P 2011). A review of documents at S&P's website does not suggest that S&P directly considers inflation risk when giving countries a credit rating. Nor does it attempt to forecast the future inflation-adjusted value of payments on bonds. If that's true, then, how can S&P give anything other than a AAA credit rating to any country that has all its debt in its own currency? And the United States is not the only country that effectively has all of its debt in its own currency and a credit rating lower than AAA. Japan also has all its debt in its own currency and a credit rating below AAA, in fact, AA-.

S&P's criteria for credit ratings
So, why did S&P downgrade U.S. Treasury securities?
S&P's explanation of the downgrade (2011b) indicated that the "government's medium-term debt dynamics" have not been stabilized and "the effectiveness, stability, and predictability of American policymaking and political challenges have weakened at a time of ongoing fiscal and economic challenges." These challenges, according to S&P, include an increase of net general government debt

"from an estimated 74% of GDP [gross domestic product] by the end of 2011 to 79% in 2015 and 85% by 2021. Even the projected 2015 ratio of sovereign indebtedness is high in relation to those of peer credits and, as noted, would continue to rise under the act's revised policy settings" (S&P 2011b, p. 4).

As S&P noted earlier in the year in a general discussion of sovereign credit ratings, S&P's "analysis may encompass political risks, including the effectiveness and predictability of policymaking and institutions. …" (2011c, p. 6). Whether or not the U.S. government is a riskier borrower in terms of ultimately paying dollar bills on the debt, S&P apparently is being consistent in applying its standards for determining credit ratings.

How do the observations made by Buffett and Greenspan fit into S&P's evaluation? A "monetary score" is part of S&P's credit rating. Perhaps not surprisingly, S&P's criteria do not say that a country is rated AAA if a central bank can print money and pay the debt. Instead, a sovereign government receives a monetary score based on, among other things, "the credibility of monetary policy, as measured by inflation trends" (S&P 2011d, p. 31). S&P's description of this point makes it clear that a lower inflation rate is associated with a higher monetary score. Using S&P's actual criteria, inflating away debt generates a lower credit rating, not a higher one. This implies that the ability to print money and make nominal payments is not a sufficient criterion for receiving an outstanding credit rating.

S&P's criteria for credit ratings are at least partly driven by its evaluation of the process determining fiscal policy. S&P concludes, as have other observers, that "the effectiveness, stability, and predictability of American policymaking" is less than ideal.

Despite the problems with U.S. fiscal policy, comments by Buffett, Greenspan, and others indicate that not everyone agrees with S&P's view of the U.S. government as having less than sterling credit. The two other large credit rating agencies have not followed S&P and downgraded U.S. debt. In addition, some customers of S&P that hold U.S. government securities disagree with S&P's AA+ rating for those securities in their portfolios and have stopped using S&P to rate their portfolios (Aneiro and Natarajan, 2011). S&P's criteria are subjective and partly determined by an evaluation of the processes determining fiscal and monetary policy. If S&P is wrong about the correct credit rating for the U.S. government, it is the criteria that are incorrect.

Gerald Dwyer is the director of the Center for Financial Innovation and Stability at the Atlanta Fed. Thomas Cunningham provided helpful comments on an earlier draft. The views expressed here are the author's and not necessarily those of the Federal Reserve Bank of Atlanta or the Federal Reserve System.


1 Plender's article discusses the ongoing sovereign debt crisis in Europe and the concern about countries' debt precisely because the governments promise to pay euros and cannot print them.

2 The small amount of Treasury Inflation Protected Securities has nominal payments that increase with the price level, but they are nowhere near large enough to affect the point. On its website S&P indicates, "Credit ratings are forward-looking opinions about credit risk" (S&P 2011).

3 The probability of a delay in payment due to the increase in the debt limit had little or nothing to do with the ratings downgrade. The downgrade occurred after the increase in the U.S. federal government's debt limit and the probability of any such delay is more than a year in the future.

Allen, Patrick. 2011. No chance of default, US can print money: Greenspan. CNBC. August 7.

Aneiro, Michael, and Prabha Natarajan. 2011. Municipalities abandon S&P after ratings downgrade. Wall Street Journal. August 17.

Claman, Liz. 2011. Buffett to FBN: S&P downgrade "doesn't make sense." Fox Business Network. August 5.

Miller, Bill. 2011. A precipitate, wrong and dangerous decision. Financial Times. August 8.

Plender, John. 2011. Bond vigilantes focus on ability to print money. Financial Times. August 16.

Standard & Poor's. 2011. Credit ratings definitions and FAQs.

Standard & Poor's. 2011b. United States of America long-term rating lowered to 'AA+' on political risks and rising debt burden; outlook negative. August 5.

Standard & Poor's. 2011c. Principles of credit ratings. February 16.

Standard & Poor's. 2011d. Sovereign government rating methodology and assumptions. June 30.