The case for a steadily-increasing federal debt

For every debtor, there’s a creditor; that’s an irrefutable truth of double-entry accounting.  So if the United States government is a multi-trillion dollar debtor, who’s the creditor?  Who’s collecting all those hundreds of billions in interest payments? 

More importantly, if the government bit the bullet and paid down the debt, those creditors would suffer a double-whammy:  they’d lose their assets (the Treasury securities they bought), and their interest income from those assets would stop.  Would that result in a net financial benefit to all parties involved?

The following present-value analysis might surprise some of your friends.  It analyzes three different scenarios: (1) paying off the debt in 30 years; (2) keeping the debt constant; and (3) steadily increasing the debt.  This analysis differs from the present value analyses we usually hear about, because it doesn’t just analyze the government’s side of the ledger, it includes the creditors’ side of the ledger, too.  Double-entry accounting dictates that when the government reduces a liability by buying back one of its Treasury securities from a public owner, it also reduces an asset on that owner’s balance sheet.  Similarly, when the government increases its liability by selling a Treasury security to a public owner, it increases the assets on that owner’s balance sheet.  This present value analysis looks at both sides of the ledger.

Here is a summary of the results.  In a nutshell, when government financial effects are netted against private sector financial effects, the net present value of all three scenarios is zero. 

Not only do the two sides cancel each other out within each scenario, but (surprise, surprise) the net present value of steadily increasing the government’s debt by 5% per year for 30 years (Case 3) is the net-present-value equivalent of completely paying off the debt in 30 years (Case 1).  Before firing off that nasty email to me, click on the thumbnails below, showing the NPV analysis details for each of the three cases.

Click on Case 1, Case 2, and Case 3, in that order:

Note that it doesn’t make any difference to the net present value whether we pay the debt off by running surpluses, versus holding it steady by balancing the dollar budget, versus increasing it continuously by running permanent deficits.  The public is happy to receive those interest payments from the government, and is happy to own Treasury securities—so what sense does it make to eliminate those Treasury securities, thereby decreasing the public’s net financial wealth?  Why not continue running permanent deficits, using the proceeds from the sale of new Treasury securities to enhance our national security (intelligence, diplomacy, military force potential, and homeland security)?  Paying the interest on that debt without ever paying any principal is the net-present-value equivalent of paying off the principal to avoid future interest payments—so what’s the big problem with paying “interest on the debt”?  What’s wrong with running permanent deficits? 

In fact, some will look at Case 3 and say “Hey, does that imply there’s no limit to the amount of debt we could run up, because the NPV is always a wash?”  Well, technically, that’s what the math is saying, but it seems to me there has to be a tipping point somewhere—i.e., the point at which the bond-buying public says, “Wait a minute, that’s a bit too much you’re running up there; I’m going to need a few hundred more basis points before I’ll buy any more Treasury securities.” 

But the good news is, it is also apparent that we are nowhere near that tipping point.  Inflation and interest rates are nowhere near hyper-levels; if we concentrate on growing the economy while simultaneously limiting the debt-to-GDP ratio to the range it’s in today, the tipping point (if there is one) will remain the undetectable supposition it is today.

Besides, one other thing I like about permanent deficits is knowing that foreigners, including some of our OPEC friends, by purchasing Treasury bonds, are helping us pay for better, more effective methods of protecting ourselves against terrorists.  In other words, we’re leveraging “other people’s money” to help fund our nation’s growth rate and national security measures, similar to the way astute private sector corporations use other people’s money to help fund their own growth strategies. 

Bottom line: What’s wrong with permanent deficits and permanently growing debt?  Nothing, as long as we grow the economy enough to hold the debt-to-GDP ratio approximately constant.  The government’s debt is a private sector asset.  Let’s grow it instead of shrinking it. 

Comments

[gunthestops]

I hate when people try to use math, science, and logic to confuse the subject matter and me! Here’s some science for you sparky, ----“Occam's razor”----if it looks like a duck, walks like a duck, and quacks like a duck----it’s probably a DUCK! ---LOL. Now I must go back and attend to the more illogical aspects of my job---the market.

[JimmyJoe]

Yes but when I sell my t-bills and that asset disappears, a new asset shows up to take its place. It seems somewhat disingenuous to write as if that weren't true.

[Counter Revolutionary]

JJ,I don't know what Gunnie was ranting/talking about, but I guess your interpretation is as good as any.

I do believe the concepts Steve is presenting here are going to surprise a lot of politicians in 08. Its all about politics, after all.

[j.k. flamebait]

but, but, only the rich own T-bills and get the interest payments which are funded out of taxes which everybody pays. so this is just taking from the poor and giving to the rich, right? or worse yet, taking from our poor and giving it directly to foreign debt-holders.

[Counter Revolutionary]

J.K.clambake, if you're going to sputter while writing responses we're going to think you are not serious.

[Phil]

Another case for leaning towards a national consumption -- versus a national income -- tax.

Since it takes money to make money, and since "the rich" have the "extra" money to "invest"/"spend," it only makes sense, then, that when government bonds are bought, this is another way that "the rich" are contributing to the "public good" (I hate that phrase!).

Therefore, it becomes all the more obvious that "the rich" not only have the means to provide jobs for other people (one never became employed by "the poor"), but also have the means to further finance government. Only in this case, it's not a tax, and it makes even more fiscal sense, because then "the rich" are getting something directly back for it, too.

Since the Chinese also understand this, only the most absolutely hardened Marxist would intellectually rebel against such a scenario.

[Stephen Reed]

Steve,

What do you think about these recent comments:

"Under Congressional Budget Office projections, the ratio of federal debt held by the public to gross domestic product will rise to about 100 percent in 2030 and ``grow exponentially after that,'' from about 37 percent now, Bernanke said. ``Ultimately, this expansion of debt would spark a fiscal crisis, which could be addressed only by very sharp spending cuts or tax increases, or both,'' Bernanke said in the testimony.

http://www.bloomberg.com/apps/news?pid=20601103&sid=aTgmnxIPdfHo&refer=us

And

"The American budget deficit is clearly a very significant concern for all of us that are trying to evaluate both the American economy's immediate future and that of the rest of the world," [Greenspan] said via satellite at the VeryGC Global Business Insights 2007 Conference.

http://biz.yahoo.com/ap/070226/hong_kong_us_greenspan.html?.v=4

[Steve]

Stephen,

CBO projections are like most of the others; they don't advertise their growth rate assumptions, and when someone digs into the fine print, it usually turns out to be something like 2.5% per year each year into the indefinite future. That, to me, is the scariest number of all -- and also the most humorous.

Ever wonder why the CBO acts as if there is only one possible scenario 25 years from now, when everyone knows there's a wide range of possible growth rates, depending on policy choices we make today and tomorrow? Ever wonder why the CBO buries their growth rate assumptions? I wonder about those things a lot, and am trying to do something about it.

[ilsm]

Steve,

"they’d lose their assets (the Treasury securities they bought), and their interest income from those assets would stop."

No, they get US $ for the T Bills they sold or were retired back to Uncle Sam, now the US $ may be no asset, but if that is the case then the interest earned ain't worth much either.

Taking your model forward, US T Bills are the only thing paying more than inflation, without insufferable risk.

Might happen but I would be very worried if the best any one can do with their money (it is really worthless then) is US T Bills.

(intelligence, diplomacy, military force potential, and homeland security)?

If any of these government investments make sense then private investments should be less risky.

Is there no better investment in the economy because the stuff bought by T Bills (intelligence, diplomacy, military force potential, and homeland security) does not provide for the common defense, nor encourage the general welfare?

Do we invest in the private sector at all?

I suspect and I will have to do some back of the envelop thinking; the case you present moves to the government being the last or only decent investment.

I think Marx, Lenin and later Hitler had that in mind.

I will read some Austrian School tonight.

[Steve]

Check the detailed models; you'll see that the surtax paid by the private sector exactly equals the principal received to redeem the bonds, every year. Net effect is equivalent to just burning the bonds, thereby reducing net financial wealth by that amount.

Also: T-bonds are not necessarily the "best" investment. They are the safest, but your risk tolerance might make you prefer AAA corporate bonds, or some other higher-yielding financial investment. It's our decision in any case.

[Patrick]

ilsm:

"Taking your model forward, US T Bills are the only thing paying more than inflation, without insufferable risk."

this is nonsense. T bills are the only thing with essentially NO risk. of course "insufferable" is a subjective term

"I suspect and I will have to do some back of the envelop thinking; the case you present moves to the government being the last or only decent investment."

again, the case presented illustrates that T bills are the only RISK FREE investment. i can assure you there are other good investments out there, although there is some risk involved

[ilsm]

Huh?

Only half the balance sheet here.

You assume the redeemed T Bills are paid 100% on the prinicple, in some way or another through taxation.

The absence of debt has no value?

There is no opportunity cost to pay interest?

Interest raised by taxes or new debt?

The underlying argument is borrow from the moneyed (creditors) rather than tax them.

[Patrick]

Steve:

Do you know of any sources that break down the composition of foreign debt? The US seems well documented, but I cant find a single other country.

Since every country seems to have debt, shouldn't some foreign debt offset each other?

[Bob]

Steve,

Why does the CBO use a 2.5% growth rate?

I suggest it may have to do with the Fed. While the Fed has never stated that a GDP growth rate is part of their objective, historically they have referenced a "goldilocks" economy when nominal GDP growth is around 2.5%. Somehow they have it in their heads that inflation accelerates to undesirable rates above 2.5% growth.

[Jon Thompson]

"I think Marx, Lenin and later Hitler had that in mind."

Godwin.

[Steve]

ilsm:
Think about it: On Monday, you look in your safe deposit box and see $1000 of Treasury notes. On Tuesday, you look in, and the box is empty, courtesy of Uncle Sam. Do you feel poorer, wealthier, or neutral?

Bob,
I have never been able to understand the logic behind the assertion that too much real growth causes inflation. I do understand that too much money growth for any given level of GDP growth is inflationary; I also understand that too little money growth can hold GDP growth below its potential. The inflation rate, not the money supply, is the signal.

I do not understand why anyone, including a monetarist, thinks he or she is smart enough to know the limits of real growth, and therefore the limit at which the "money supply" should be allowed to grow.

Assuming 2.5% growth, and implementing policy accordingly, risks becoming a self-fulfilling prophecy. Footbinding is the image that comes to my mind.

[Steve]

Patrick,

Sorry, I don't know of any international reconciliation of debt ownership. The Treasury (in its TIC section) has some numbers on how much foreign debt is owned by US citizens, as I recall.

[ilsm]

Steve,

"Think about it: On Monday, you look in your safe deposit box and see $1000 of Treasury notes. On Tuesday, you look in, and the box is empty, courtesy of Uncle Sam. Do you feel poorer, wealthier, or neutral?"

Having taught grad students, my favorite answer is "It depends".

It depends on whether the T Bill is gone because the government defaluted. Then I might feel poorer.

Growing debt eventually grows the risk of the government defaulting. Current account balances would be involved here.

But then we all would be poorer and so relatively I am not worse off than others.

I would feel wealthier if: I had the thousand and could invest it wisely, or if it was taxed for a good cuase, like reducing the US debt load thus improving the net worth, less debt makes flexibility and the government needs to be flexible.

Heck, the government could have avoided the need for the T Bill in the beginning by taxing me.

I was lucky the government borrowed the money from me rather than taxing it.

I've clipped a lot of coupons and am happy so when the piper wants paid that is what happens.

My other teaching comment is: "how would you see me getting poorer?"

Define poorer, in your answer. I really do not want the limited definition of poorer in that I lost a T Bill.

A bit broader.

By the way you had a chart up here which I have copied which shows the debt to GDP of other countries.

Japan's D/GDP was around 160% in 2005.

Italy's was somewhat higher than the US; Germany's about the same Debt/GDP.

Is there a correlation between debt and a 10 year recession in a country with a huge saving rate?

Is the government borrowing the Japanese' excess saving?

[Stephen Reed]

Steve,

Thanks for your response. Good point about the GDP growth assumptions. I guess what worries me is that two of the top macroeconomists (Bernanke and Greenspan) seem to be jumping on the debt doomsday bandwagon. It gives me pause since they don't have a political stake one way or the other, so their economic expertise tends to be a little bit more objective.

I definitely would like the CBO to be more upfront with its assumptions, however. These really long term forecasts tend to be pretty flakely for the most part. What is your experience with these type of forecasts 10 or more years out. Do they tend to have some hint of accuracy in them?

[Jim Glass]

"if the government bit the bullet and paid down the debt, those creditors would suffer a double-whammy: they’d lose their assets (the Treasury securities they bought),"

How so? The creditors would change one asset (T bonds) for another (cash), and there's surely no loss of asset wealth in that.

Then, presumably, since they'd invested their cash previously they'd do it again, like in high-quality corporate bonds, making financing more readily available to productive businesses.

So the question is: Do we want to have bond investors financing through the federal deficit the likes of homeland security grants that in NYS go 3% to NYC and 97% to upstate and cow-country districts as pork distributions, or do we want to have them finance productive businesses?

The size of "the debt" is really a secondary issue. The key issue is the *quality of spending*.

If federal spending *in reality* produces a return to the nation greater than the cost of capital to the federal gov (the interest rate on bonds) then go ahead and run it up! Tax away *and* deficit spend too!

But if federal spending in reality on the whole produces a lesser or negative return, then spending should be slashed down until we can get the rate of return on it to improve -- and the deficit should be cut as a tag along side effect.

[Jim Glass]

"Double-entry accounting dictates that when the government reduces a liability by buying back one of its Treasury securities from a public owner, it also reduces an asset on that owner’s balance sheet."

Well, it reduces "assets: T bonds" but increases "assets: cash" by the exact same amount, so the owner's total assets on the balance sheet remains unchanged.

"Similarly, when the government increases its liability by selling a Treasury security to a public owner, it increases the assets on that owner’s balance sheet. "

Again, the owner's total assets on the balance sheet don't increase but shift, as "assets: T bonds" increase while "assets: cash" -- or very likely, "assets: corporate stocks and bonds" -- are reduced by the exact same amount.

So balance sheet assets in the private sector remain unchanged either way. Since these reflect the discounted value of future cash flows, it confirms what you say, as far as it goes.

But it doesn't mean there is no real economic difference between reduced debt and increased debt. There are two big ones for starters:

1) With reduced debt investors shift investments from financing gov't expenditures to financing private market expenditures. For those of us who believe the latter provide a much higher rate of social return, this is a good thing.

2) More debt means more tax must be collected to finance the carrying of it, that 5% or 6% interest per dollar per year. That interest all comes from current taxes.

And taxes impose a dead weight cost on the economy that increases not with the tax rate but the *square* of the tax rate. So if the tax rate rises by 10% its economic cost rises by 21% ... if the tax rate rises 40% its economic cost rises 2X ... if the tax rate doubles its cost rises 4X, etc.

Thus, *even if* one thinks deficit spending pays a positive return to the nation (in spite of its being used to send Homeland Security funds to cow country, build bridges to nowhere, pay Warren Buffett's Medicare and SS benefits, etc.) still, for it to provide a net benefit to the nation that return must surpass a "hurdle" that rises *exponentially* with the tax rate increases used to finance the deficit's increasing size.

Exponentially rising costs are difficult to carry. That's why the real size of the debt shouldn't rise too high relative to the economy.

[Jim Glass]

"Why does the CBO use a 2.5% growth rate?"

Their middle-case projection is not 2.5% exactly but a rate that declines over time from recent rates to about 2%.

Simple explanation: GDP growth consists of growth in number of workers, times growth in the number of hours they work, times growth in hourly productivity.

Over the last 40 years average real GDP growth was 3.2%, with an annual 1.7% growth rate for the first factor, number of workers.

In coming years, after the baby boomers start retiring, the growth rate in the number of workers drops to about 0.5% -- a decline of 1.2 percentage points.

This reduces GDP growth from 3.2% to 2.0% right there, by arithmetic, other things equal.

Of course, other things might not be equal. In the future people might decide to work a whole lot more hours than in the past, and/or hourly productivity might rise faster than in the past, and/or we might open the borders to invite in a flood of new immigrant workers.

But there's no particular reason for believing so -- and there's just as much reason to believe people will want to work fewer hours than in the past and/or the rate of productivity growth will fall below the historical average, and/or we'll build a wall to keep immigrant workers out.

That is, the 2% projected future GDP growth rate is simply the actual long-term historical average projected into the future. It's as simple as that.

The more complex answer is in the Monte Carlo simulations that the Social Security and Medicare actuaries use to compute the probabilities of all kinds of different scenarios, which one can read about from them, and which give as a middle case what CBO gives as its middle case.