« God has been located | Main | Denmark's humorous socialist »


Well, I'll note that the "intra-governmental debt" is economically irrelevant, as it is owed by the govt to itself, and thus has an asset/liability value of exactly $0.00 on the Consolidated Balance Sheet of the United States.


I mean, it's not owed to anybody!

That "ig-debt" (I think of the debt held by the public as "ugh-debt") merely represents political promises of future spending that may (and surely will) be changed amid the politics of the future -- no different than promised spending for agricultural subsidies, school aid, whatever is politically supported at the moment.

E.g. the SSTF bonds just represent promised future SS spending. But Congress has the absolute right to reduce SS spending to a level that would make the bonds unneeded, so they'd *never* be paid. (And it is entirely possible that Congress will in fact do so post 2019 or so, by means testing "the rich" out of benefits, to avoid the large income tax increases that will otherwise be needed then to redeem the bonds). In that case the bonds could roll over unto enternity, never paid off.

To make such "flexibility" legally possible, Congress has given the ig-debt bonds some very unique traits (contrary to the popular myth that they are the exact same as ugh-debt bonds).

E..g.: the ig-debt bonds are payable on demand regardless of maturity date, and as long as demand is not made they roll over forever. (Let's see anybody either here or in China try to buy T-bonds with *those* terms!)

Also, since the ig-debt is owed by the govt to itself, it is unenforceable against the govt by anybody else -- it is payable only if the govt chooses to pay it ... to itself!

So ig-debt bonds are not owed to anybody ... are unenforceable against the govt ... have no maturity date (upon which default could occur if not redeemed) and thus can legally roll over until the year 10,000 ... and are payable by the govt only voluntarily, just as is the case with any "regular" expediture ... and then are payable only to the govt itself.

Which leaves them being not much in the way of a debt -- in fact, no kind of debt at all!

Which is why the Balance Sheet of the US reports them at $0.

OK, so drop the ig-debt off the chart, and we see that the ugh-debt -- I mean the *real* debt -- of the US is owned 49% by foreigners (instead of only 29%).

Although the real debt also is $4.1 trillion (41%) smaller than the total given on the chart including the ig-debt.


Where did you put the debt held by the Federal Reserve this time?

If you didn't already include it in the debt "Owned by US Gov't" then it should be taking a big bite out of the publicly held debt these days.

Right, it's not in the intragovernmental debt, so it's part of the top slice. I should probably start breaking that out, for more clarity. Similar to all of the publicly-held debt, it will be perpetually rolled over, so the question of the so-called debt burden comes down to the affordability of the interest payments to the (non-governemtal, non-Fed) public.

I'm sure you are all TECHNICALLY correct that debt doesn't matter as it will be rolled over ad infinitum, and that OASDI debt doesn't count since the government owes it to itself and can choose to default with no consequences. However, I wonder if that is really realistic in view of the very real political commitments that all (Democrat and Republican) politicians have made to future recipitents of Social Security and Medicare and in view of the likely decrease in value and security of the currency given the vast increase in spending that the current administration is pursuing.

I am concerned that your economic theory is based on assumptions that will not be politically acceptable and that therefore the debt will matter and will be required to be funded.

As a retired CFO of large corporations, I am quite aware of the accounting techniques that can be used to obscure reality. However, the senior citizen lobby will certainly drive from office any politician who threatens their SS and Medicare funds. I think it is correct that the Obama administration will attempt to "means test" SS as Medicare is already "means tested". However, this will be a drop in the bucket, and I suspect the idea that the debt incurred by spending those OASDI funds in the past will have to be funded and paid to the citizenry. That would imply either more private debt (assuming the market will bear it) or increases in taxation that will suppress economic growth. For that reason, I think that ignoring this major and growing amount of debt may be theoretically accurate, but in practice will be disastrous. As a side note, the current recession has already diminished the OASDI tax revenue to the point that it is deficit to current obligations.

Finally, consider how you would view the credit worthiness of a government that defaults on its promises to its citizens. I suspect that would color your judgment as to how secure your investment in U.S. Debt would be.

Jerry McInvale


You're right, being "technically" correct isn't always the same as being pragmatically correct, or politically correct. However, I for one will take two out of three. If I can't have all three, I choose to forgo the political correctness in favor of attempting to change it.

My economic theory is based on the near-universal agreement among economists of all ideologies that a ratio of debt to some measure of debt-carrying capacity is far more meaningful than the absolute debt level (in dollars). Most have settled on the ratio of debt to GDP. I use that a lot because of its widespread use, but I personally prefer the ratio of interest to tax receipts or interest to outlays.

Those ratios satisfy the tests of technical correctness and pragmatic correctness -- although they fail miserably when it comes to political correctness. According to most politicians (who are highly skilled at reading opinion polls) I am supposed to fear deficits and debt, and elect them so that our grandkids might be saved. But I don't fear deficits and debt IF THE ECONOMY IS GROWING AT A PACE SUFFICIENT TO RENDER THEM HARMLESS. Today we have $10T debt in a $14T economy; I'd be *happy* to know that our grandkids would be "saddled" with $100T debt (ten times as much) in a $140T economy (ten times as much), or a $210T economy (15x, putting us that much farther from "default").

Therefore, shifting the debate to the important questions of if, when, and how the economy might grow at a sufficient pace is what I spend my time on; consequently, I have no time left for the more-popular debate about if, when, and how we might start running surpluses again in order to pay down the debt level.

Getting back to a healthy, growing economy is what it's all about in the short run; getting the most growth we can out of that economy is what concerns me most about the long run. Sustainability of social security and Medicare (and everything else about the federal budget) depends on it. [The so-called unfunded liabilities we always hear about would be completely offset by the present value of future tax receipts at a real growth rate of between 4 and 5 percent, last time I ran the numbers. Too bad the FRUSG doesn't even attempt to quantify the PV of future tax receipts (see Robert Rubin's comments about "assets" on page 8 of this document - http://tinyurl.com/crmod7 ); if it did, that would be a basis for getting into the important debate about growth rate assumptions, how to enhance them, which programs help it, which don't, etc. In its absence, however, the politicians will continue having a fear-mongering field day with the unfunded liabilities number.]

In any case, a mix of "debt and equity" (borrowing and taxation) enhances the economy's growth rate, in my judgment -- just as it does for most successful private sector firms. SS and Medicare sustainability depend on sufficient growth, but political correctness doesn't yet recognize growth as something policy can affect; growth is a mysterious thing that "just happens." I'll continue trying to change that naive mindset, however.

"I'm sure you are all TECHNICALLY correct that debt doesn't matter as it will be rolled over ad infinitum, and that OASDI debt doesn't count since the government owes it to itself and can choose to default with no consequences. However, I wonder if that is really realistic in view of the very real political commitments that all (Democrat and Republican) politicians have made to future recipitents of Social Security and Medicare..."

It's not merely technically true, it is REALLY true, which is why the ig-debt is counted at $0 by the Treasury on the Balance Sheet of the US.

As to how real the political commitments to future program recipients may be ...

1) Remember what happened when SS hit a funding gap in 1983 -- benefits were cut to close 50% of the gap (typical political 50-50 split-the-baby compromise). So we have precedent as to how the political system reacts, and the coming funding gap circa 2020s is HUGELY larger.

2) Consider the political desires of seniors *in the 2020s* -- don't make the mistake of projecting today's political conditions upon them then.

At that point, to stay even with Medicare and Social Security spending, income taxes will to be increase by over >50% from today's levels, all projections say, e.g....
... with about a third of that needed to service the SS Trust Fund bonds to pay SS benefits.

In the 2020s, in that situation, as a voting retiree do *you want* to pay a 15% income tax increase to pay your own SS benefits? If, say, the alternative "means test the rich out of benefits" could save you from that tax?

[] If you are in the middle class, means-testing the rich out of benefits saves you from a major tax hike on your fixed income -- including on your own SS benefits! -- and you come out ahead. Vote to means test!

[] If you are *the rich*, do you really want to pay a 15% tax increase on all your great pension, investment, business income, to save your relatively piddling SS benefits? No, means-testing yourself out of benefits to avoid that tax puts you ahead on net. Vote to means test!

[] If ALL seniors themselves vote to means-test to avoid the tax hike, then who else is going to vote *for* a big tax hike on themselves? Corporations? Young families raising children? Special interest groups that will have more federal money available to them if SS spending is reduced via means testing? Probably no, no, no ....

Can you find *anyone* who as a matter of self-interest will vote FOR redeeming the SFTF bonds via the tax increase?

If not, why wouldn't Congress obey the will of the voters' and let the bonds roll over until the year 3,000 of beyond?

That's how *real* the SS ig-debt is, as a matter of political pragmatism.

(Note that means-testing was already introduced into SS to cut benefits for the rich in 1983, and is already on the table as a fix for the future problem.)


Have no desire to be argumentative. My concern is simply the fallout from "breaking the promise" to repay admittedly unsecured debt.

To your points. Were benefits really reduced in 1983, or was the projected rate of increase reduced. I don't really know, but I am unaware of any actual reduction in the history of SS.

As I understand the administration's definition of "rich", I am pleased to be a member of that group and accordingly since my retirement plans required me to enroll in Medicare, I can assure you that I already pay 250% of the normal monthly cost, and my social security will be taxed fully at normal federal rates. Also, there are not enough of us "rich" to make a substantial difference in the deficit versus the cost of future benefits, even if all of us are stiffed completely. Therefore the choice the government will face is either to be a deadbeat who refuses to pay his/her debts or to increase taxes tremendously. Either way, the impact on the country's financial condition, its reputation and debt ratings may be significant.

The issue isn't whether the bonds will roll over until eternity, it is whether the government will meet it's financial commitments. If not, I suspect the government's reputation will be the same as any other deadbeat.

Personally, I can get by anyway and expect to have to - but as I have paid in well over $500K in my working career, it is a bit annoying to be told never mind, we don't really have to live up to our commitments that we may to justify the payments. You may be correct that it is not legally required - but what about ethically?

Jerry McInvale

"Jim, Have no desire to be argumentative. My concern is simply the fallout from 'breaking the promise' to repay admittedly unsecured debt.

"To your points. Were benefits really reduced in 1983, or was the projected rate of increase reduced. I don't really know, but I am unaware of any actual reduction in the history of SS."

Don't worry about being argumentative. Steve will tell you I'm a friendly guy (when not on usenet).

Benefits were reduced in 1973 IIRC, and on a much larger scale in 1983. In 1983 in two major ways:

1) Delaying retirement age for younger workers (plus a number of lesser provisions such as cutting the cost of COLAs). That reduced benefits for future retirees, while protecting current ones (the most motivated voters!)

2) Introducing disguised means testing by making benefits subject to income tax -- the give-away key difference being that the tax collected on SS benefits was not sent along with all other income tax into general revenue (to fund the Cold War, etc.) but instead sent back to the SSA.

Thus, the net result was that current SS benefits -- the *net* transfers from the SSA to SS recipients -- were reduced by a means test that happened to exactly coincide with income tax rates.

The Greenspan Commission members and politicians who made this change knew full well they were effectively imposing a disguised means test -- but an overt means test was never going to fly politically, and what works is what works ... that's politics.

It's worth noting that this "hidden means test/tax" has been increased steadily since 1983 -- the Clinton tax law increased the portion of benefits subject to tax from 50% to 85%, and the dollar threshhold at which it starts applying is not adjusted for inflation, so it gets smaller each year, making more benefits subject to the "tax" that goes back to the SSA, reducing net benefits per person.

The result was that the politicians in 1983 projected their funding gap would be closed 50% by benefit cuts and 50% by the payroll tax increase they imposed.

However, of course ...

3) The major payroll tax increase for workers starting then further slashed NET benefits of then-young future retirees.

Workers retiring before around the year 2000 received from SS $16 trillion more than they contributed (todays money at present value). No wonder they loved Social Security!

But future retirees will receive *less* than they contribute. Social Security makes them poorer on a lifetime basis.

So in a real sense, the "promises to future workers" attributed to Social Security have already been broken. The 1983 changes, structured to alleviate the *short term* political problems of Congress, saw to that.

Now, as to the 2020s, it's easy to say "all promised benefit payments should be paid" -- but one also then has to say *who should pay for them*.

The projected SS funding gap for then is hugely larger than in 1983 -- two points of GDP, about a 15% across-the-board income tax increase from today's level on all income: corporate, investment, wages ... and on retirees' pensions, IRAs and of course their Social Security benefits too(!)

Should middle-class retirees take such a big income tax hit on all their fixed retirement income just to fund their own SS benefits (which will be subject to the tax increase too)?

Will they vote for it, if they have a clear option to avoid it -- just means tax the rich out of benefits?

And again, consider "the rich" -- if a 15% tax increase on *all* their income would cost them more than their (for them) piddling small SS benefits, what are they going to vote and lobby for?

In sum, it's easy to say "we have an obligation to pay all promised SS benefits" ... but you then have to close the cash flow circle: *who* should pony up the 2 points of GDP annually to pay the benefits?

Note well, if you drop that whole cost on "the young" of then -- so that from today's level where SS was made a money-loser for the post-1983 young, SS then becomes a *much worse* money-loser for post-2020 workers -- SS will become a **black hole of losses** for that period's younger workers.

Is that fair? How will they react to that politically?

OTOH, being that people over 40 have most of the nation's income, if you increase taxes on older people and seniors to pay the money right back to them for their own benefits ... what's the point of that?? (And the deadweight economic cost of a 15% tax increase would be a very *real* economic cost of society).

Means-testing the rich seems to me the most economically efficient and progressively fair method of closing the cash-flow circle -- and being that politics is already driving things in that direction, also the most likely to actually take place.

But whatever one's opinion of that, Remember this key fact:

SS is a paygo, cash flow program. This means that going forward *benefits must equal taxes*, to the penny!

There is no way around that. So whatever the future promise of SS is can't be changed, on net. There is no way to either "keep it" or "break it" to future participants as a group, as a whole.

Because SS has until now paid out $16 trillion more than it collected, in the future participants MUST get back $16 trillion LESS than they put in, lose that amount to it!

There is NO WAY way to change this $16 trillion. The only question is *who* is going to pay it, and *how* -- through reduced benefits or higher taxes, or a combination.

That's what "the promise of Social Security to future generations" today comes to.


Let me try it like this. What if we said there will be no increase in taxation for the future, and therefore benefits must be paid from current cash flows plus the "debt" that was incurred because the government chose to use past surplus payments into OASDI for other reasons. If we accumulated the sum of those debts at conservative interest rates - treasuries? - then I would be OK with whatever benefits were available from that sum plus "paygo" cash flow. If the government wants to means test it, I may think that is unfair, but so be it. The Administration and Congress would have to "fess up" that the greatly reduced benefits happened beacause they had made spurious commitments with no way to fulfill them. Perhaps that would convince the voters to more seriously examine the "promises made" in view of the credibility or lack thereof of the Congress.

I emphatically do not believe that future generations should pay for my "retirement". Funds for SS and Medicare have already been confiscated by the government for their own purposes, and that is not the fault of nor should it be the liability of future generations. However, pretending that it didn't happen by refusing to recognize the "debt" strikes me as a bit deceptive.

Bottom line, I am well aware that the commitments can not and will not be met. But, I would like to see some recognition of the swindle.

Jerry McInvale


OASDI "trust fund" accounting is a legacy of FDR's only way of selling it to Congress: keeping it "separate" from the general fund, and making it seem like a safe of some kind that holds the money people "contributed" as it piles up, waiting for those people to retire. It worked, and it passed Congress.

In fact, total cash in (to the govt) must equal total cash out, every year -- otherwise, the Treasury would be either creating or absorbing money from the economy, which is the Fed's job, not the Treasury's. Combined with trust fund accounting, that means cash flow surpluses and deficits at the department level must net to zero. Bottom line: for years, we've payroll-taxed working people a little too much to support the cash outflow to their contemporary retirees, and the difference has supplemented general fund outlays; soon, that will reverse, and the outflow to retirees will be composed of more than just inflow (of payroll tax receipts) from workers.

That reversal won't hurt, if tax rates don't increase (i.e., neither income nor payroll tax rates). The only two ways to keep tax rates from rising, but also get increased cash flows from taxes, is (1) to have more income-earners paying taxes, and (2) to tax higher incomes-per-worker at those unchanged rates. Those two requirements are the long way of saying we need "economic growth": more people working, and higher income-per-worker.

Regarding the trust-fund accounting situation, I wrote a tongue-in-cheek article a while back, to illustrate how the "trust fund raid" could be completely paid back in six years, with absolutely no effect on anybody's cash flow. Here's the magic wand: http://tinyurl.com/cecprv [...and here's an earlier article on the same subject, the trust fund "raid": http://tinyurl.com/bkd574 ]

Tongue-in-cheek articles aside, the point is this: cash in equals cash out, regardless of what we label it. Economic growth is the imperative.


I'm not sure I understand your point other than the fact(with which I heartily agree) that growth is important. As of the end of Fiscal 2007, the SS & Medicare "Trust Funds" had accounting assets of $2539.2 Billion and the accounting debt owed the trust funds by the U.S. Government General Revenues was the matching $2539.2 Billion with interest accrued annually based on Treasury Bond rates in excess of 4 year maturity. At some point to get beyond the theoretical, this is either a real debt or it is not. I'm not sure from your comments whether you believe it is a debt that will have to be repaid or not. Which is it?

I certainly hope for robust growth, but whether you believe the Obama Administration or the Congressional Budget Office, there appears to be no prospect of growth overcoming spending in the next decade. And, I would note that even in periods of robust growth in the past 50 years, deficit spending has been more the rule than the exception - even including the infamous trust funds. Therefore, I'm not sure that your assumption of increased revenue from growth and/or taxation will be adequate to fund ongoing expenditures and the deficits coming from the "trust funds". If you are confident of that, what is the source of said confidence? If not, then how does this remain a theoretical problem without actual consequence?

Either the government will have to "default" on the "bonds" held by the trust fund with very public consequences, or it will have to increase spending to fund the annual trust fund expenditure deficits. In either case, it will not be a pretty picture and will affect the credit rating of other government debt instruments. So, to me, this is not at all a theoretical issue. What am I missing?

Jerry McInvale

"Jim, Let me try it like this. What if we said there will be no increase in taxation for the future, and therefore benefits must be paid from current cash flows..."


"plus the 'debt' that was incurred because the government chose to use past surplus payments into OASDI for other reasons."

But this requires an increase in taxation, so it contradicts your prior statement.

Example: SS projects to owe promised benefits exceeding payroll tax by $400 billion in 2030. By that date...

[] Without the SS bonds, Congress would have to increase general revenue (income taxes) by $400 billion to pay those benefits.

[] With the SS bonds, Congress will have to increase general revenue (income taxes) by $400 billion to pay those benefits (while redeeming $400 billion of trust fund bonds).

As $400b minus $400b = $0, the existence of the trust fund and its bonds ("the debt") is meaningless, worth $0 -- so there is no way to pay benefits but from current taxes.

Congress would have to make the exact same decision of how to increase taxes/reduce benefits by 2030 whether the trust fund and "the debt" existed or not.

Steve is exactly right that the gov't runs on current cash flow in real time, that is all.

Social Security benefits provided under the 1983 changes ...

[] In the past, post-1983, cost less than the cash payroll tax revenue the gov't collected, and Congress spent the surplus cash flow on other things.

[] Going forward, post-2018, will cost more than the cash payroll tax the gov't will collect, so Congress will have to increase taxes/cut benefits/both to close the cash flow gap.

That is all. The entire story in substance. Simple as that.

"The SS debt" is economically and financially meaningless, which is why it is reported at $0 by the Treasury on the Balance Sheet of the US. As it is meaningless, there is no meaningful way in which it can be "repaid".

The only real functions of "the debt" have been in politics, in giving Congress political cover to spend taxes labeled "FICA" on other things -- and as a device that wonderfully distracts from the real issues in arguments over "reform".

"If the government wants to means test it, I may think that is unfair, but so be it."

It won't be the government that decides that. It will be future voters deciding how much they want to tax themselves to pay themselves.

"The Administration and Congress would have to 'fess up' that the greatly reduced benefits happened because they had made spurious commitments with no way to fulfill them ... I would like to see some recognition of the swindle"

I'm with you on that, and that day is coming, as surely as arithmetic is true. In the 2020s. If we live until then we'll see it with our own eyes.


Regarding the SSTF & Medicare TF bonds you write: "At some point to get beyond the theoretical, this is either a real debt or it is not."

They are NOT a real debt, period. That is why they are carried as a debt liability of $0 on the Treasury's balance sheet of the US (link provided in earlier in this thread).

I don't know why this is so difficult for so many people to understand.

If there is a debt, please specify to whom it is owed.

Do not say "The SSTF bonds are a debt to Social Security participants". That is false, totally untrue. You, as a participant in Social Security have no legal right or interest in the SSTF bonds. Zero. None.

This completely separates and distinguishes the SSTF bonds from "debt held by the public" you may own. If you personally own a T-bond, payment to you on it is guaranteed by the Constitution, and the US's failure to pay you on its maturity date indeed would be a default.

But Congress has a total and absolute right to cut Social Security program spending, just as with any other spending program (Flemming v Nestor, US Supreme Court). Say that in the 2020s -- following the will of the voters -- it cuts benefits to you and others, by say delaying the retirement age combined with means-testing, so the bonds will never be "needed" at all.

Can you then object to that benefit cut? No! No more than you could if your agricultural subsidy was cut. Can you object that the bonds weren't redeemed? No! You have no legal interest in the bonds at all! Is there a "default" on the bonds? No!

Remember, the SSTF bonds are owed by the govt to *itself* -- and they have no maturity date! So how can default on them possibly occur? When would it occur? The bonds have already been rolling over since 1984 with no default. They can roll over until the year 3,000 with no default. By their terms.

Moreover, the govt has the explicit right to change the use of the bonds from SS to any other purpose whatsoever: to put into the Medicare trust, give to the Defense Dept., whatever. Default on them is not legally possible.

"Either the government will have to 'default' on the 'bonds' held by the trust fund with very public consequences..."

No! No default, no consequences! Let me give an example that may make this clear.

A generous Uncle like you promises to pay his nephew $200,000 when he retires at age 65.

The nephew asks: "Great, but can I really count on this?"

Uncle: "Absolutely. But it is only a non-binding verbal promise I can revoke at any time."

Nephew: "Huh? Why won't you make your promise binding?"

Uncle: "Because things may change. I may need the money myself. You never know."

Nephew: "Huh? But how can I possibly rely on it then?"

Uncle: "Look, if it makes you feel better, I will write a legally binding note for $200,000, secure it by all my assets, and make it payable on your retirement date. AND I will create a trust, name it the "Nephew Retirement Trust" and put the note in that trust. OK?"

Nephew: "Yeah! That's better! So the note will be payable to me, and be held safely in the trust to which I am the beneficiary. Sounds good!"

Uncle: "No! No! My note will be payable *to me*, and the beneficiary of the trust will be *me*. Not you. I told you, I might need that money myself someday. My promise to you is non-binding and I can revoke it at any time."

Nephew: "But that gives me absolutely NOTHING!"

Uncle: "What more do you want, you ingrate nephew? I've promised you $200,000, backed it by a legal and fully secured note, and put it in the 'Nephew Retirement Trust'. WHAT MORE DO YOU WANT? "

The nephew's answer here depends on who the uncle is. If he is...

Uncle Jerry, then ...

Nephew: "You're a madman! I see why everyone in the family called you my 'funny Uncle'. Get away from me and my money!"

Uncle Sam, then ...

Nephew: "You're right! What a great deal! Absolute security! Who could ask for anything more?"

Exact same situation in both cases, only the Uncle's name differs. Go figure.

Question: in either case, if Uncle doesn't pay Nephew, does Uncle "default" on the note?

More questions: If you, Jerry McInvale, write out a fully legal note for $200,000, secure it by your home, and then make it payable to *yourself*...

(1) Have you created a "real debt" that will affect your credit rating as compiled by Equifax, etc.?

(2) If you then don't pay yourself on the note, will you legally "default"? ... Will your credit rating be ruined as a result? ... Will a judge hear your case if you file a suit to enforce payment on the note or alternatively collect damages, against yourself?

Whew! You guys wear me out.

1. Suggesting that the government's unfunded liabilities would be funded if only we could count the present value of future tax revenue... No good. At the rate we've been going the last 75 years, the present value of future tax revenue is already more than consumed by the present value of future new benefits and promises, leaving the current lot of unfunded liabilities... unfunded.

2. Regarding this $4-trillion-plus of "government debt owned by the government" - I agree the book value of the asset and liability is zero. This represents the "money printed out of thin air" that will cause big inflation? And isn't this the most rapidly growing section of Steve's pie chart?

The comments to this entry are closed.