One more time credit rating agencies show their incompetence

Yesterday morning S&P released the following information:

“Although we believe these strengths currently outweigh what we consider to be the U.S.’s meaningful economic and fiscal risks and large external debtor position, we now believe that they might not fully offset the credit risks over the next two years at the ‘AAA’ level,” said Standard & Poor’s credit analyst Nikola G. Swann. […] “Our negative outlook on our rating on the U.S. sovereign signals that we believe there is at least a one-in-three likelihood that we could lower our long-term rating on the U.S. within two years,” Mr. Swann said.

So the credit risk of the US is potentially growing according to S&P so it threatens to downgrade the US credit rating. In order to understand how S&P arrived to this conclusion let’s have a look at how credit risk is defined in its 2007 sovereign debt primer:

QUOTE 1: “A sovereign rating is a forward-looking estimate of default probability. […] The key determinants of credit risk [are economic risk and political risk]. Economic risk addresses the government’s ability to repay its obligations on time and is a function of both quantitative and qualitative factors. Political risk addresses the sovereign’s willingness to repay debt. Willingness to pay is a qualitative issue that distinguishes sovereigns from most other types of issuers. Partly because creditors have only limited legal redress, a government can (and sometimes does) default selectively on its obligations, even when it possesses the financial capacity for timely debt service.”

S&P has two ratings, a local currency rating (default risk on domestic-currency-denominated debt) and a foreign currency rating (default risk on foreign-currency-denominated debt). The primer notes:

QUOTE 2: “A sovereign government’s ability and willingness to service local currency debt are supported by its taxation powers and its ability to control the domestic monetary and financial systems, which give it potentially unlimited access to local currency resources. To service foreign currency debt, however, the sovereign must secure foreign exchange, usually by purchasing it in the currency markets. This can be a binding constraint, as reflected in the higher frequency of foreign than local currency debt default. The primary focus of Standard & Poor’s local currency credit analysis is on the government’s economic strategy, particularly its fiscal and monetary policies, as well as on its plans for privatization, other microeconomic reform, and additional factors likely to support or erode incentives for timely debt service. When assessing the default risk on foreign currency debt, Standard & Poor’s places more weight on the impact of these same factors on the balance of payments and external liquidity, and on the magnitude and characteristics of the external debt burden.”

S&P provides a table that shows the number of defaults on foreign-denominated sovereign debt and notes that: “Defaults on sovereign foreign currency bonds occurred repeatedly, and on a substantial scale, throughout the 19th century and as recently as the 1940s.” Interestingly, however, it does not provide such a table for the domestic-denominated sovereign debt. The primer continues by noting:

QUOTE 3: “One might ask why, if sovereigns have such extensive powers within their own borders—including the ability to print money—sovereign local currency ratings are not all ‘AAA’. The reason is that while the ability to print local currency gives the sovereign, and the sovereign alone, tremendous flexibility, heavy reliance upon such an expansionary monetary stance may bring the risk of hyperinflation and of more serious political and economic damage than would a rescheduling of local currency debt. In such instances, sovereigns may opt to reschedule their local currency obligations.”

So overall here is the view of S&P:

Credit risk = economic risk (capacity to pay) + political risk (willingness to pay)
There are two type of sovereign debt: Domestic-currency denominated and foreign-currency denominated
Governments that can tax and issue their own currency have the full capacity to pay domestic-currency-denominated debt
Inflation risk is a form of default risk

Now let’s look if all this makes sense in the case of the US. First, note that the US is a monetarily sovereign country: the federal government issues its own currency, the federal government has no foreign-denominated debts outstanding, and the federal government does not peg its currency in anyways (and so does not promise to convert US dollars into another currency or gold on demand). This means that the US federal government can always meet payments that are due to its creditors anytime (today or in the future) and anywhere (in the US or abroad) by crediting bank accounts. Stated alternatively, there is NO possible risk of default for economic/technical reasons. The federal government cannot run out of money, it has a perfect capacity to pay: economic risk is zero.

The problem with S&P is that it has a shifting definition of economic risk. As quote 3shows, S&P is aware of the absence of economic risk in the case of the US but it proceeds to argue that there is one by changing the definition of default risk to include risk of hyperinflation. OK…hold on… are we talking about default or inflation risk? These are two completely different risks.

In terms of default risk, higher inflation (if revenues of borrowers are indexed to it) increases the capacity of repayment and so LOWERS default risk rather than increases it. If, following S&P’s logic in quote 2, one considers taxes a form of revenues that helps to pay debt, then tax revenues rise with inflation and so lower the risk of default.

In terms of inflation risk, inflation pressures on the demand side are very low and the risk of uncontrollable hyperinflation that S&P refers to in quote 3… give me a break! But there is a deeper misconception at play. S&P (and all politicians in Washington) seems to think that they have some meaningful control over the fiscal balance of the federal government. They do not; as the UK experience, among others, is painfully showing us. The fiscal balance of the government is ultimately driven by the net saving of the private domestic sector and of the rest of the world, as was explained many times on this blog. The following identity holds at the aggregate level:

(G – T) ≡ (S – I) + (J – X)
Government deficit ≡ Net domestic saving + net foreign saving

Stated alternatively, the US government must deficit spend currently and in the near future in order for the private sector to repay its outstanding debt and for foreigners to accumulate dollars. It is only if the private sector as a whole decides to dissave (spends more than what it earns) and/or foreigners decide to dissave (import more than they export to the US) that the government can lower its deficit and potentially run a surplus. Any attempts to go against the desires of the two sectors will lead to a recession. The deficit will go down by itself as the private and foreign sector gain confidence and decrease their net saving.

By conflating inflation risk and default risk in their rating, S&P (and probably other CRAs) creates confusions in financial markets and promotes dangerous ideas. Sounds familiar? Remember all those toxic mortgage products that were rated AAA? One more time, S&P is showing how silly it is to extend a rating methodology that was developed for corporate bonds to other sectors of the economy. Even for corporate bonds, results are far from perfect (remember all those financial institutions that had an AAA rating or so right before they failed?)

So overall there is zero default risk due to economic risk in the US, as an issuer of the currency in which sovereign debt is denominated, the US (like the UK and Japan but contrary to Eurozone countries) can always repay dollar-denominated debts it issued. But what about credit risk due to political risk, i.e. unwillingness to pay?

First and foremost, cases of voluntary default by a monetarily sovereign government on its domestic-currency-denominated date are extremely rare. Following Rogoff and Reinhart, it looks like Japan is a candidate in 1942, a very unusual political and economic situation (and probably a means to limit inflationary pressures by not adding purchasing power to the private sector, so default could be a good policy). It is hard to detect more countries given that the authors do not give us any background about the monetary system at the time of default. The US is not experiencing any massive war exhausting its resources (even that is not enough to increase the risk much), its political system is stable, and Geithner has told us many times that the debt ceiling will be raised that any chance of political problem is at best remote. True, the Congress may tie its own hands and decide not to raise the debt limit, but how realistic is that possibility?

Bottom line, credit risk is remotely remote. You have more chance to be hit by lightening twice during your life than to experience a default of dollar-denominated sovereign US debt. The only really worrisome variable is the stupidity of US congress and its willingness to try to fix something that is not broken. People love Medicare (by far the most popular program of the government), they love social security, and automatic stabilizers are working as predicted. Let them be.