CEPR - Center for Economic and Policy Research


En Español

Em Português

Other Languages

Home Publications Blogs CEPR Blog The Devastating Interest Burden of the Debt

The Devastating Interest Burden of the Debt

Written by Dean Baker   
Wednesday, 17 August 2011 11:04

It is important to remember that most of the people in Washington debates on economic policy do not know much economics. They tend not to be very good at arithmetic either. That is why they were blindsided by the collapse of the $8 trillion housing bubble that wrecked the economy.

As we get endless pontification about the crushing debt burden it is worth touching base with reality on occasion. In that spirit, CEPR brings you the latest data and projections on the ratio of the federal government's interest payments to GDP, courtesy of the Congressional Budget Office (CBO).

Click for Larger Image


Source: Congressional Budget Office.

As the chart shows the interest to GDP ratio is currently at a crushing 1.3 percent, near the post World War II low. However this figure overstates the burden somewhat. Last year the Federal Reserve Board refunded almost $80 billion to the Treasury. This was interest earned on government bonds and other assets it now holds. That leaves a net interest burden of 0.8 percent of GDP, by far the lowest of the post World War II era.

Of course the burden is projected to rise in future years. The baseline projections shows the burden rising to 3.3 percent of GDP by 2021, the end of the forecast period. This baseline is probably overly optimistic in some respects, since it assumes that the Bush tax cuts are allowed to expire and some other items in current law that will probably not happen.

If we adjust the the baseline for these factors, the debt to GDP ratio is projected to be just over 90 percent by 2021, approximately 20 percent higher than in the baseline. If we raise the interest payments by the same percent, then we get a ratio of interest to GDP of 4.0 percent, still not exactly crushing.

It is also worth noting that if the Fed continued to hold $3 trillion or so in assets, and rebated the interest earned on this money to the Treasury, then it would reduce the net burden of the debt by close to 1.0 percent of GDP. This would mean that even in 2021, if we just left everything to run its course, we would still not face as large an interest burden from the debt as the early 90s.

Okay, this arithmetic interlude is over. You can rejoin the Washington elite and start panicking over the debt again.

Comments (9)Add Comment
lay person
written by Melinda, August 17, 2011 1:27
Could you please compare the US debt ratio to GDP and "crushing interest" to those of the other industrialized nations of the world? It is my understanding that Japan, in particular has HUGE deficits and debt compared to GDP, but as a lay person, i'm wont to compare the numbers with any insight. Thank you.
Comparing stocks to flows
written by Benedict@Large, August 17, 2011 1:43
Not nice to compare stocks and flows. Produces meaningless results. For example, a debt-to-GDP ratio of 90% actually should be expressed as 0.9 YEARS (yes, years ... from grade school arithmetic. Debt (a stock) is $$$, GDP (a flow) is $$$/year. Dividing the two gives an answer in years.) So your 90% actually means it would take 9/10ths of a year to pay off the debt if that's all we did with our GDP.

Now, when a household takes out a mortgage on a new home, what typically would be their debt-to-"GDP" ratio, GDP in this case being annual take-home pay? Well, probably about 400% (!!!), or about 4 years.

So here's the question. How can it be that a banker considers it a safe bet that an individual can handle a 400% debt-to-GDP ratio, but those same bankers go FREAKING NUTS that the United States, the greatest economy in history, might be getting near a mere 100%?

See how ridiculous it gets when you try to compare stocks to flows?
written by coberly, August 17, 2011 3:24

i'm glad you finished your comment, but i was afraid i was going to have to disagree with you.

as it is i think your objection to comparing stocks to flows is a bit pedantic and overdrawn.

if i compare my debt on the house, "a stock," to my income, "a flow," it is easy enough for me to do the internal arithmetic and see that if i am paying 10%, say, on my 400 thousand dollar house, and my income, say a hundred thousand a year, i can... or cannot... afford the payments.
knowing that my debt was "four years" would not help me understand it as well.

i am reasonably sure this was Dean's point. Now that I look at it again, I am not sure it wasn't your point. if so, the missing argument in your letter was "it's the bad guys who compare stocks to flows." But that still leaves me with "needlessly pedantic."
written by coberly, August 17, 2011 3:26
the captcha word obscuration is ridiculous.

Reply to Coberly
written by Benedict@Large, August 18, 2011 5:17
Allow me to restate based upon your criticisms, as I unnecessarily confused two points.

1) My "pedantic" point of expressing the ratio as years as opposed to a percentage is important in terms of effect. 90% is a more startling number (it's almost 100%!) than 0.9 years, which though the same (once years is dropped), appears more significant. Of course if you've written a book meant to raise alarm (which is what was done), you'd want to use the former instead of the latter. It is far better for book sales, even if it is technically inacurate. Which bears to this point. When reading the work of two noted economists, one assumes technical accuracy, and in this case, that has been sacrificed for effect. (Even 90% of a year has less effect (and much greater confusion) than just 90%.)

2) My second point of comparing the national economy to a household could certainly have been made (and perhaps more clearly made) by remaining with the percentage notation. But even this hides the stocks vs flows criticism. In this case, a more proper comparison would be of two flows, one of expected debt service by period, the other of income available by period to service the debt. Debt to GDP claims to be a proxy for this, but for the life of me, I can't figure how. Which brings me back to my first point; that debt to GDP is a number manufactured for its shock value, but containing little if any real content.

I hope that clarifies a bit, and that I have not again been pedantic. :-) Thanks for your interest.

And yeah, those captchas suck.
Interest rates
written by Steve Roberts, August 18, 2011 9:57
Several notes:
Bankers are using sound banking practices by lending to people at 400% income? That's a justification?

As is being shown in Europe and Japan, anyone can afford to pay the interest on a debt when it hits 200% of GDP if the rate of interest is 2%. Heck, I could make payments on a mortgage at the size of 4x my income if you made the interest rate low enough and promised to NEVER raise it. What happens (as is happening in Europe right now) if we can't keep our interest rates low (inflation?) and they zoom up to 6-7%? Historically relevant numbers. How affordable is our debt at those levels.
Yeah, but
written by Brian, August 18, 2011 11:34
We all know interest rates are at historic lows. What happens if, for example, interest rates go back up to more normal numbers such at 5% for mid term loans and 7% for 3 years?

I believe the debt payment would almost double.

This is where is starts to get ugly as you get a compounding effect - doubling of the debt payment increases the additional debt you have to take out, which further increases the debt.

This the the credit crisis you see in most nations that go bankrupt.

And that does not even cover the case where interest rates go above normal. That could triple the payment or more.

The problem is we are getting closer to the edge of the cliff. If the road stays dry (interest rates stay low) we are OK, but if with hit a slippery patch (interest rates increase) we have much less margin to work with.

And I know the US can't go bankrupt per se, but if the FED starts printing to much money there will be all sorts of negative effects that will essentially be a US bankruptcy.
Interest payments as a % of GDP?
written by MLS, August 18, 2011 2:02
This does not make sense to me. The government does not have GDP available to pay interest on debt, they have tax receipts. According to Treasury Direct, the Federal Government's interest expense projects to be about $500 billion for 2011. According to the CBO, the government will collect about $2.6 trillion in total taxes for fiscal 2011. So interest payments make up about 19% of total inflows, before paying a dime for social security, medicare, etc.

Even a slight rise in interest rates such as a return to historical norms implies a very large problem on our hands.
MMT First Generation
written by Warren Mosler, August 19, 2011 11:56
First, when the Fed buys securities, it pays for them by adding dollars to member bank reserve account balances, and it pays it's target interest rate on those balances. So this only saves interest expense to the extent the yield on the securities it owns is higher than the Fed's 'policy rate' for the banking system's cost of funds. Yes, this is currently the case, as you evidenced by the dollars the Fed turned over to the Treasury. But this means that should the Fed hike rates and thereby pay those higher rates on member bank reserves, the interest cost for the US govt overall rises, regardless of the size of the Fed's portfolio.

Second, your article implies that interest payments represent some kind of a burden. Clearly there is no burden in making payment, as interest payments are simply a matter of the Fed crediting accounts at the Fed. There are only real economic consequences if those funds cause spending on real goods and services, in which case that spending would add to GDP, output, and employment. Should aggregate demand become excessive, and unemployment somehow become 'too low' to the point of inflation, those interest payments could then be considered a macro economic 'problem'

For further discussion see:
'the 7 deadly innocent frauds of economic policy' free online at:

Write comment

(Only one link allowed per comment)

This content has been locked. You can no longer post any comments.


Support this blog, donate
Combined Federal Campaign #79613
budget economy education employment Haiti health care housing inequality jobs labor labor market minimum wage paid family leave poverty recession retirement Social Security taxes unemployment unions wages Wall Street women workers working class

+ All tags