The 90 percent Debt-to-GDP Threshold and CBO’s New Debt Estimates

On March 5, the Congressional Budget Office (CBO) gave us a preview of its take on the President’s budget proposals for Fiscal Year 2011 (starting October 1st this year) in a letter to Senate Appropriations Committee Chairman Inouye. CRFB blogged on key features of the preliminary analysis here and here. Buried in the letter is CBO's estimate that debt held by the public would rise to 90 percent of GDP by 2020 under the President’s budget. This is well above the administration’s own estimate (77 percent of GDP), and sets off alarm bells. Recent research by noted economists Carmen Reinhart and Ken Rogoff (R&R) shows that countries grow more slowly when fiscal debt goes over the 90 percent debt-to-GDP threshold.

R&R presented their research in a paper at the latest American Economic Association annual meeting. It has generated a lot of buzz in fiscal wonk circles, as has their recent book. (“Growth in a Time of Debt,” National Bureau of Economic Research Working Paper 15639, January 2010, and This Time is Different, 2009.)

It is worrying enough to think that sometime this decade we will reach a point where our debt is sufficiently high to slow growth in a significant way. However, a closer look at R&R’s work indicates that we probably don’t have to wait until 2020 to arrive at the 90 percent threshold. In fact, we are probably just about there right now.

If you look at OMB's “gross central government debt” (the numbers used by R&R) rather than “debt held by the public” (the numbers more commonly cited by CBO and the administration), our debt/GDP ratio last year was 83 percent of GDP and is projected to be 94 percent of GDP this year. So, the United States may soon be at the point at which our debt level is linked to slower economic growth, according to R&R.

The reasons for the linkage of slower growth to the 90 percent threshold are not well-understood, but R&R and Savastano suggest a plausible explanation in another paper. Individual countries may well have a specific debt threshold above which investors demand an increase in risk premia to hold a country’s debt.  A country’s debt threshold may be based on perceptions of its historical experience [comment: or even technical issues related to its debt and financing structures]. As a country approaches its debt “limit”, interest rates will rise as risk sentiments shift. Growth will be slower as a result  (Reinhart, Rogoff, Savastano, “Debt Intolerance,” NBER Working Paper No. 9908, 2003)

For the United States, what will happen when we cross the 90 percent threshold this year?

There is tremendous uncertainty about the outlook – to say the least. While we’ve seen some signs of nervousness about our rising debt from domestic and international investors in the past year, that nervousness has not led so far to the problems that R&R highlight. The United States has so far retained its appeal as a “safe haven” (perhaps more accurately described as the international lender of last resort).

But, at some point, investor sentiment will shift, at the very least because judgment over the risk:return ratio for relative assets will change as the global economic and financial situation changes. How will investors then regard U.S. government assets relative to other assets in the United States and around the world? With our domestic savings gap likely to remain large, we will be increasingly vulnerable to a shift in investor sentiment as our public debt leverage rises.

So seeing that we are about to cross our high debt threshold now rather than in 2020 should give us pause – at a minimum. While some argue there is no magic number (see Paul Krugman's recent blog), R&R's findings suggest that once debt exceeds a certain share of the economy, there are costs in the form of lower growth. Their research offers a compelling argument about the costs of waiting to make credible, concrete plans to put our fiscal house in order once the economy is on stronger footing.