Arnold Kling  

U.S Government Debt Since World War II

Japan and the World Interest R... The Federal Debt/Income Ratio...

In the Washington Post, Joel Achenbach writes,

When I spoke to Peter Orszag, the director of the Office of Management and Budget, he expressed optimism that the administration can balance the primary budget -- not including interest payments -- by 2015.

Below is some analysis of how the ratio of government debt to GDP has evolved in the United States since World War II. The data come from tables B-78 and B-80 of the 2010 Economic Report of the President. I deleted the "transition quarter" in 1977.

The change in the ratio of debt to GDP can be broken into two components. One is the ratio of the primary deficit to to the current debt. The other is what I call the "erosion factor," which is the nominal interest rate minus the GDP growth rate. In a previous post, I suggested an analogy with a owning a rental property. The primary deficit is the operating loss on the property--rental income minus expenses, but not including interest. The "erosion factor" is the interest rate minus the price appreciation of the property.*

As of 1946, the ratio of debt to GDP was 108.67 percent. From 1947 to 1970, it fell to 27.96 percent. A substantial amount of the drop was due to the fact that the government ran a primary surplus in all but four of those years (the exceptions were 1953, 1959, 1962, and 1968), for a cumulative primary surplus of 43 percent of the 1946 debt.

I was surprised by this. I had not remembered these surpluses. One reason is that the primary surplus excludes interest payments. Including interest payments, the government mostly ran deficits, particularly in the 1960's. Another reason may be that the U.S. only began to include Social Security surpluses in the overall Budget late in President Johnson' second term. Had we used the "unified" budget from the beginning, the deficits would have seemed much smaller and we would have counted more surpluses. I assume that the data I am using have gone back and recalculated the "unified" budget for all of history.

The point here is that the sharp drop in the debt/GDP ratio from 1947 to 1970 was only about half due to the fact that the GDP growth rate was higher than the interest rate [correction--more than half was due to GDP growth minus the interest rate. But still, a lot was due to the primary surpluses]. (In fact, GDP growth averaged 3 to 4 percentage points higher than the interest rate from the end of the second World War to 1970. By the way, I calculated the interest rate by simply taking the ratio of interest paid in a given year to the debt outstanding at the end of the previous year.) The other half of the sharp drop was due to all those primary surpluses--the U.S. genuinely ran a responsible fiscal policy, even in the 1960s.

The 1970's were a decade in which we inflated away some of our debt. We ran primary deficits. Real GDP growth was sluggish. But nominal GDP growth exceeded the interest rate by nearly 5 percentage points per year. Bondholders got shafted, and this was not sustainable. In fact, in the 1980's, the "bond market vigilantes" (an expression coined by Edward Yardeni, a Wall Street Fed-watcher of the period) got their revenge, and nominal interest rates actually exceeded the growth rate in nominal GDP. This differential widened in the first half of the 1990's, as bond investors continued to price for more inflation than actually transpired.

The Reagan deficits of the early 1980's, along with the bond market vigilantes, caused the ratio of debt to GDP to rise from a low of about 26 percent of GDP to 40 percent by 1985, and then for the next decade the bond market vigilantes kept the debt ratio climbing, to 49 percent of GDP in 1995. For the rest of the 1990's, the Clinton economy generated primary surpluses in the government budget, and so the debt to GDP ratio fell to 33 percent by the time President Clinton left office.

President Bush's fiscal policies caused the debt ratio to edge up to 36 percent by 2007, and then the recession and the policy reactions of Bush and Obama sent it up to 53 percent currently. The CBO projection is for a 90 percent ratio by 2020.

My reading of this history is that one should not be optimistic that we can simply shrink the ratio of debt/GDP by growing the denominator. A lot of the reduction that took place between 1947 and 2000 was due to running primary surpluses. If we are going to do that again, we will have to do so in spite of the fact that military spending is a much smaller share of GDP (so that arithmetically there is less room to cut) and in spite of the way that Social Security and Medicare are going to be affected by demographics and health care spending trends.

The "erosion factor" of the nominal interest rate minus the GDP growth rate, has been negative at times--particularly in the 1970's. However, you cannot fool bond investors forever, and the best guess is that over long periods the erosion factor will average zero. In other words, we are unlikely to be bailed out by a negative erosion factor. In fact, if investors lose confidence, the erosion factor could become very high, very quickly.

*[For those of you who like algebra, let D = debt, P = primary deficit, i = interest rate, and g = growth rate of nominal GDP, and d = growth rate of debt.

We are interested in d-g, the growth rate of debt minus the growth rate of GDP. If that is positive, the ratio of debt to GDP is rising, and conversely.

d = P/D + i, that is the debt grows at the interest rate plus the ratio of the primary deficit to the initial debt. Subtracting g from both sides of this, we have:

d - g = P/D + (i-g)

On the right-hand side are the two components that I am looking at. P/D is the ratio of the primary deficit to the current debt. (i-g) is the "erosion factor," of the nominal interest rate minus the growth rate of GDP.]

Comments and Sharing

CATEGORIES: Fiscal Policy

COMMENTS (8 to date)
Indy writes:

One should remember the US post-war situation though, with a two-year spike of double-digit inflation rates (topping out nearly at nearly 20%) from July 1946 to July 1948, (and another spike during all of 1951).

Few saw these rates coming. Perhaps some people, cynically savvy of the temptations to debasement of a debt-overburdened government of a war-and-sacrifice-weary nation, saw these things coming, but not “the market” in general, and probably not all those buyers of low-interest long-term war-bonds who watched a hefty portion of the real value of their savings disappear in those 24 months.

Indeed, one of the arguments for not worrying about the present level of debt is that, “It was over 100% of GDP in WWII, and we managed that.” Well, they had the advantage of certain societal conditions which we do not – the population growth rate was high, the US dominated global industrial production in the aftermath of a world war which devastated all our competition, longevity was short, and so entitlement spending was low, there was this thing called the baby boom, the debt was mostly domestically owned, and, oh yes, there were intermittent periods of high inflation, unexpected by the market at the time creditors loaned their money to the government at low interest rates which helped to erase the real value of the debt.

War bonds (the last sold in early 1946, I think), were 2.9% percent (annual) ten year bonds. A old neighbor of mine when I was growing up still had on of the original cardboard cutouts, where you could pop in your 75 quarters ($18.75) in return for a bond promising you $25 in ten years. He was the only one in most of my life to inform me about the post-war inflations. Everyone else always talked about the late 70’s, but he remembered the late 40’s and said they were really bad as well. He thought he lost a lot of money on his bonds, but nobody else but him seemed to remember those details that far back.

Now, the government eventually allowed the terms to be extended for several decades, but let’s say you bought one during the war and you took your $25 in the 1950’s. Using this neato online calculator we can find the decadal inflation rates: Jan 1942-52: 68.8%, 43-53: 57.4%, 44-54: 54.6%, 45-55: 50%.

But the bonds only got 33% after a decade, so no matter when you bought your liberty bonds (and I think over half the country – or 80 million people – did, and at over $2,000 worth per saver and in 1940’s money in the midst of a world war!) you experienced a significant real loss (and the government got a significant escape from war indebtedness) when you collected your proceeds.

Now that's real 'erosion'. Hopefully, we don't replay that strategy in the mid 2010's.

Boonton writes:

It would seem to me there should be a relationship between interest rates and growth. If interest rates were higher than growth then in theory you could earn more by simply stashing your money in savings rather than making true investments. For that to be sustainable, labor and shareholders would have to give up income to bond holders since there's no other place to pay the higher returns.

This implies to me that the long run 'riskless' rate of interest must be lower than economic growth or at least equal to it. Periods where the rate is higher than growth are periods where the bond market is trying to avoid seeing the gov't default on the debt by inflating it away.

steve writes:

Which Reagan and Bush fiscal policies caused debt/GDP to rise? Maybe we should not repeat those. (Reread Romer's classic paper on this general topic.)


kev writes:

Are there any plausible big near-term cuts in spending coming soon? How much could be saved by drawing down our Mideast wars? I heard Obama was considering cutting Soc. Sec. and Medicare in his first term.

Boonton writes:


But the bonds only got 33% after a decade, so no matter when you bought your liberty bonds (and I think over half the country – or 80 million people – did, and at over $2,000 worth per saver and in 1940’s money in the midst of a world war!) you experienced a significant real loss (and the government got a significant escape from war indebtedness) when you collected your proceeds.

Money creation during WWII was very high and wage and price controls were in place to prevent inflation. When Arnold discussed whether the lack of a post-war recession indicated Keynesian economics was wrong (Kling's logic: Less gov't war spending should equal recession), what was missing was a lot of minted money was itching to get spent that couldn't because of rationing. When that was lifted that spending flushed into the economy replacing the gov't spending that was leaving. I suspect that this wasn't really planned but just happened.

But from the perspective above I think we are just looking at two sides of the same coin. If the gov't clamped down and prevented the inflation on the grounds of helping the bond holders it would have had to dramatically raise taxes, slash spending and/or raise interest rates to destroy money. Since the bond purchases were so much a cross section of the population, I suspect the damage would have been much the same. Sure you wouldn't have lost on your bond savings but if you had to spend 6 months unemployed the lost income would more than offset your regained interest earnings.

mulp writes:

Cheney's explanation is simpler:

"Reagan proved deficits don't matter."

Josh writes:

Arnold, I had similar thoughts a while ago. I came up with this graph showing what percentage of the current US debt (this was "current" as of 2008; i.e. it doesn't include any Obama budgets) each year was responsible for:

Basically, the point is that if in year 1, you borrow $100, then in year 2 when you're paying back the $100, you actually get credited with a $100 surplus towards that year's spending.

I recall that I used a bunch of assumptions about how the government pays back its debt - it's a surprisingly hard topic to find information on - but I think the overall picture was robust to a wide range of assumptions (although the individual numbers obviously change quite a bit).

The end result was that only a handful of events - the two WWs, 1982 and 2004 (and to a much lesser extent, the mid-80s "defense buildup" and the 1990 recession) - have any liability at all. Everyone else had a (small) surplus as you noted.

Ritwik writes:

Arnold, the data you present in your article is as compelling a data-based macroeconomic case as I have ever seen - AGAINST your argument.

Take the period in which the US reduced its debt/GDP the most - 1950 to 1969. You present four periods. The average 'growth factor' (taking a quick and dirty arithmetic mean) for those 4 periods is 22.2% Of GDP. Similarly, the average primary surplus is 7.1%. Of net outstanding debt. Which itself went from 79% to 29% in those years (let's take 54%). This translates to 3.9% of GDP.

Of course, 3.9% of GDP and 22.2% of GDP does not translate to debt reduction of 26.1% per period (nearly half of that, due to obvious reasons). However, it gives an idea of the proportion that was contributed by each. 22.2/26.1 is nearly 85%. So cut all the primary surplus that the US ran, and my estimate it would have still reduced its debt/GDP from 79% about 33%. 33% vs. 29% is not a huge deal at all!

Of course, one could argue that if the US was not running primary surpluses, its growth factor would have reduced too (bond vigilantes et al). But then again, if the US was not growing so fast, it may not have been running primary surpluses either.

If I was PK, I'd wear a placard with your data and a 'I told you so' printed on it.

Comments for this entry have been closed
Return to top