Arnold Kling  

Social Security Privatization

PRINT
Disappointing Election Year?... Growth and Economic Literacy...

UPDATE: I thought this was more recent, but it's an older column that Winterspeak pointed to. Still worth reading.

Jeffrey A. Miron and Kevin M. Murphy write,


privatization has no effect on the solvency of the Trust Fund; it reduces inflow to the Trust Fund to the same degree it reduces outflow from the Trust Fund.

They use a different argument from mine on the implausibility of the stock market bailing out Social Security.

for those participants who already owned stocks before privatization, this "new" option has no value. Such participants would simply offset any increased stock holding in privatized accounts with decreased holding in other accounts, leaving the return on their portfolios unchanged

For Discussion. How does the point that privatization is a wash for the Social Security trust fund relate to the argument that the transition cost is zero?


Comments and Sharing


CATEGORIES: Social Security



COMMENTS (62 to date)
Don Lloyd writes:

As briefly demonstrated here ,

I would claim that, if future obligations are not reduced, then no possible current solution or improvement can exist which is not economically inferior to simply making up the SS payout shortfall with newly printed dollars, unless the solution causes the future supply of goods and services to increase.

Privatization, rates of return, etc., are all meaningless and likely economically negative unless they create new future goods and services.

Refutations requested. ( dlloyd@catallarchy.net )

Regards, Don

Brad Hutchings writes:

How does the point that privatization is a wash for the Social Security trust fund relate to the argument that the transition cost is zero?

Both points rely on a dynamic analysis -- over time (he says redundantly).

Arnold... you know what would be really useful? "Arnold Kling's Social Security Spreadsheet". Have it compare current assumptions, band-aid fixes, radical privatization for solvency of Social Security system as a whole and individual retirees over the next 50 years. Have the formulae and input values explicit and some kind of blogo-community process for adjusting them. Post in table form on a dedicated page on this site. If I knew the first thing about this stuff, I'd do it myself. But I do know that it would start to make a visible case for change.

cb writes:

I disagree with both of the comments -

privatization has no effect on the solvency of the Trust Fund; it reduces inflow to the Trust Fund to the same degree it reduces outflow from the Trust Fund

I thought we already agreed that the Trust Fund is a myth. It may not have an effect if you ignore time, but if you immediately privatized SS, you will have the same outlays for the current retirees, with no inflows. The cash would go directly from the Treasury to the financial markets. The reduction in outflows wouldn't occur until today's contributors retire.

for those participants who already owned stocks before privatization, this "new" option has no value. Such participants would simply offset any increased stock holding in privatized accounts with decreased holding in other accounts, leaving the return on their portfolios unchanged

I personally would love to have this 'new' option. 15k/year I could invest in something I can see when I get my statement every month, or 15k I'm loaning to my government in exchange for a yearly statement telling me what my benefits are projected to be, which I happen to have no faith in. I don't understand what would motivate me to decrease my holdings in my current accounts, as you suppose. What am I doing with that money? Spending it?

Tom writes:

I think Miron and Murphy have it bass-ackwards. For starters, they dismiss the notion of returns on private Social Security accounts because the current system doesn't really pay returns. That's true, but so what? The point is that workers who are allowed to drop out of the system would be able to invest their money where it yields more retirement income than they would have received from Social Security.

Yes, those same workers would receive lower Social Security benefits (or the equivalent in a lump-sum payoff for taxes already paid). But, given the declining worker-to-retiree ratio, they have to be better off if they're allowed to put their money into almost any private investment vehicle. It doesn't have to be the stock market; high-grade corporate bonds would do the trick.

The net transition cost is zero because money "borrowed" to cover near-term benefits is "repaid" by lower benefits paid to future retirees. But that has nothing to do with Miron and Murphy's abracadabra with the mythical trust fund.

cb writes:
The net transition cost is zero because money "borrowed" to cover near-term benefits is "repaid" by lower benefits paid to future retirees. But that has nothing to do with Miron and Murphy's abracadabra with the mythical trust fund.

That money that's paying for near-term benefits is theoretically coming from the taxes paid by the people receiving those benefits. Obviously, that's not the case. You can argue that all this is a net zero cost economicaly, but's that's looking at it from a country standpoint. Within generations, today I'm paying money to today's retired people w/ the supposed promise of receiving money when I'm retired. We can go round and round, but anybody w/ a brain my age wouldn't count on my gov't giving me a decent return on my 'retirement tax dollars'. I prefer to take care of my parents myself. I don't need to transfer hard-earned dollars to other people's parents because they weren't responsible enough to plan for their retirement. Have you seen the savings rate in this country, hmmmm... I wonder why the dollar is falling. It really pisses me off that the no saving, heavy spending generation in front of me is coming up with supposed grand ideas to save their kids from themselves.

Let me get this straight, 'money borrowed to cover near-term benefits' is repaid to me in lower benefits to me in the future? So.... the crappy return today's retires got on their retirement funds w/ the gov't is being covered by borrowings, which I have to pay, and... I get lower benefits on the investment I made by paying SS taxes. In other words, today's workers get screwed coming and going. Got to love socialism. Where is that constitution? Somebody must have buried it.

Tom writes:

cb, I'm not talking about the constitutionality of Social Security. I agree that it's unconstitutional. In fact, I recently ran a long post on just that subject. It's here.

I'm talking about the mechanics of privatizing Social Security. It has to go something like this: Younger workers are allowed to drop out of the Social Security system, but at some point the system has to borrow to pay the benefits of persons who are already retired or weren't young enough to drop out of the system. The younger workers who don't have to pay Social Security taxes have money to invest in private accounts, which yield greater returns than the "returns" they would have realized from traditional Social Security. Some of those additional returns are captured by the government, in the form of taxes, to repay the money borrowed earlier. The younger workers keep the rest, and they're still better off than they would have been had they been forced to stay in the system.

As older workers die off, the cost of their benefits is reduced, eventually to zero, which eventually reduces the need for borrowing and taxation to zero. It's the older workers' benefits that go down, in the aggregate, because they're dying off. And as their benefits go down, borrowing to defray those benefits goes down, which means that younger workers have to pay a smaller and smaller fraction of their returns from private accounts. The net effect on a particular cohort of younger workers depends on how the government arranges the borrowing and taxation over time.

There's no free lunch -- but those who are allowed to privatize will definitely enjoy a bigger meal than those who aren't. It's too bad for younger workers that they can't simply walk away from the system and keep all the returns from their private accounts -- but someone has to cover the benefits of older workers who supported those who went before them and now, quite reasonably, expect something in return.

At the end of a transition like the one I've described, no one would be worse off than they expected to be, and a lot of (younger) people woudl be better off than if they had been forced to stay in the system. I don't see a better way out of the Social Security mess.

Lawrance George Lux writes:

The fact exists that the greatest Libertarians today, who got the chance to privatize their SS accounts, would reemerge as 'Welfare Babies' if and when they ever needed some benefits or aid. The privatization of SS accounts, therefore, is not a wash.

The nonexistence of the SS Fund is a reality, but a reality which could be changed, by remodeling it as a National Banking system. SS taxes would be devoted to paying off the Treasuries currently held by the Fund, which would serve as its core Reserve. The SS Fund would be committed to paying all liabilities, and investing in viable risk ventures--as do other banks. This Fund would start as a aggregating Bank initially as We are still running SS tax surpluses. This will eventually disappear, but already invested Capital would provide additional funds.

This avoids the pitfalls of individual investing, as well as the operating losses of Investment fees. It does not call for immediate increase in National Debt, unless Congress and President insist on continued deficit spending. Subject to normal rules of banking and Government banking law, it would probably be the safest means to alter the funding of Social Security.

There remains the added need to Standardize Benefits, limit Medicare benefits, and establish an unlimited-Pay SS taxation. lgl

Jim Glass writes:

Privatization is a remedy for the fundamental problem facing SS -- not insolvency, but the fact that it is about to start paying negative returns to millions of today's young workers, and no more than trivial positive returns to anybody, as the SS actuaries show.

SS funding gap remedies that are simple and sensible in principle -- such as delaying retirement age as the population gains longer and healthier life expectancy -- fail the political test because they are benefit cutbacks that drive these already negative-to-miniscule rates of return further down.

A fundamental promise of FDR and SS's founders was that SS would always pay a "fair" positive return to everybody. Of course its great popularity has resulted from paying *much more* than a fair return to the first generations of participants.

That closing the funding gap per se is not the difficult or serious problem facing SS can be seen from the fact that big funding gaps in SS have been closed *repeatedly* since the 1940s with no serious problem by raising taxes and (later) cutting benefits.

Today's problem is that they've run out of room to do so while maintaining the promised positive returns for participants -- and starkly negative returns stand to make SS as unpopular in the future as big positive ones made it popular in the past.

Real returns figure to be 25% lower than even the formula ones the SSA actuaries give above because of the funding shortfall, as either a tax hike or benefit cut to close the shortfall reduces return on contributions accordingly -- and *that's* the problem of closing the funding gap.

That said, privatization certainly can play some role in reducing the funding gap.

E.g: In the SSA figures referred to above, note that a low-income single male gets a little less than zero return on SS contributions -- let's round up so he gets $1 forty years from now for $1 contributed today, both current value. (Ease of computation!)

Now say that instead of paying $1 of SS tax today he invests in real savings. The gov't now must borrow $1 to cover the lost tax revenue to pay somebody else's benefit. The average real rate on long-term gov't bonds since going off the gold standard has been 2% -- say it pays that on the borrowing. Meanwhile the individual invests the $1 in SS savings to earn 4% (no reason to go nuts with growth stocks, investment grade corporate bonds and the like will do).

Forty years in the future the individual will have $4.80 in savings instead of a $1 benefit. The gov't will have to pay off $2.20 in bonds, assuming it compounded its interest charges, but owe $0 in benefits to the individual. That creates a net gain of $1.60 betwen them that can be divided between them by some formula with both coming out ahead.

Being that that amount is 160% more than the entire benefit owed to the individual on his contributions under the current benefit formula, it is nothing to dismiss lightly.

An additional fiscal benefit is that the government's borrowing is shifted forward to today when it can be afforded, and is reduced 40 years from now when it will be borrowing up to its neck trying to finance Medicare.

So while privatization won't close the SS financing gap by itself, its potential contribution towards doing so shouldn't be sneezed at.

Bernard Yomtov writes:

I think you misrepresent their statement when you quote them as saying

privatization has no effect on the solvency of the Trust Fund; it reduces inflow to the Trust Fund to the same degree it reduces outflow from the Trust Fund.

What they say is (my bolding)

In fact, privatization worsens Trust Fund solvency, since the Trust Fund receives less revenue but still pays promised benefits to existing and future retirees. This is a matter of arithmetic: if the Trust Fund takes in fewer taxes and pays the same benefits, its balance must decline.


What if instead of just establishing private accounts, privatization also reduces the stream of promised benefits by the amount of reduced taxes? Then privatization has no effect on the solvency of the Trust Fund; it reduces inflow to the Trust Fund to the same degree it reduces outflow from the Trust Fund.

In other words, your quote is correct only in the context of reduced benefits. Sure. Eliminate benefits entirely and you'll save a ton of money.

Tom writes:

As someone said in an earlier comment, the solvency or insolvency of the trust fund is irrelvant. The real issue is how to provide a given level of benefits at the lowest cost to participants. There's only one way to do that: privatization. Social Security doesn't yield a real return; it's merely a transfer-payment Ponzi scheme that's going to begin claiming victims in about 14 years. Private accounts, on the other hand, do yield real returns. Therefore, privatization must provide a given level of benefits at the lowest cost to participants. The only question, then, is how to get from the present transfer-payment scheme to a system of private accunts.

Don Lloyd writes:

Tom,

...Private accounts, on the other hand, do yield real returns....

The returns may be real, but they provide little or no nutrition if you try to eat them.

For a given supply of goods and services in the future, an increased level of returns simply transfers goods and services from employees to SS recipients. These transfers will be lower than those that occur from an increase in the future payroll tax rate, but the same as those transferred by making up the SS shortfall by printing up an equivalent number of new dollars and including them in the SS payments.

Increased returns that are not accompanied by an increased future supply of goods and services are useless unless they are predicted and used to reduce current payroll taxes.

Regards, Don

Boonton writes:
SS funding gap remedies that are simple and sensible in principle -- such as delaying retirement age as the population gains longer and healthier life expectancy -- fail the political test because they are benefit cutbacks that drive these already negative-to-miniscule rates of return further down.

A fundamental promise of FDR and SS's founders was that SS would always pay a "fair" positive return to everybody. Of course its great popularity has resulted from paying *much more* than a fair return to the first generations of participants.

1. SSI is an Insurance system, not a savings system. 'Returns' were never promised, only benefits if someone became a certain age (and later benefits if someone became disabled).

Now say that instead of paying $1 of SS tax today he invests in real savings. The gov't now must borrow $1 to cover the lost tax revenue to pay somebody else's benefit. The average real rate on long-term gov't bonds since going off the gold standard has been 2% -- say it pays that on the borrowing. Meanwhile the individual invests the $1 in SS savings to earn 4% (no reason to go nuts with growth stocks, investment grade corporate bonds and the like will do).

Forty years in the future the individual will have $4.80 in savings instead of a $1 benefit. The gov't will have to pay off $2.20 in bonds, assuming it compounded its interest charges, but owe $0 in benefits to the individual. That creates a net gain of $1.60 betwen them that can be divided between them by some formula with both coming out ahead.

A static analysis if I ever saw one.

1. Increasing US borrowing today will likely increase interest rates. Increasing interest rates will boost the cost of the bond that your hypothetical investor will have to pay off (or his children will pay off in the form of taxes...rather ironic they will pay taxes for debt incurred today just like we pay taxes for our 'debt' incurred to our parents thru their SSI payments).

2. Increasing interest rates, of course, hurts stock returns attacking the benefit side of your equation.

The SSI trust fund surplus today is artifically raising stock returns and dampening gov't bond returns. You can't assume that will all stay the same if you just flip a switch and reverse the structure of the trust fund.

Boonton writes:
As someone said in an earlier comment, the solvency or insolvency of the trust fund is irrelvant. The real issue is how to provide a given level of benefits at the lowest cost to participants. There's only one way to do that: privatization. Social Security doesn't yield a real return; it's merely a transfer-payment Ponzi scheme that's going to begin claiming victims in about 14 years. Private accounts, on the other hand, do yield real returns. Therefore, privatization must provide a given level of benefits at the lowest cost to participants. The only question, then, is how to get from the present transfer-payment scheme to a system of private accunts.

Of course SS yields a real return. By reducing borrowing today costs are cut in the future. If $100B less of 30 yr bonds are issued in 2000 then coupon payments that would have been made from 2001-2031 are eliminated.

Private accounts are not some type of magical money machine. If the economy grows by 4% a year then by definition the economy's ability to provide retirement benefits is only growing by 4% per year.

Imagine a simple thought experiment; imagine '401K world' where everyone has 401K's only. Like our world, the ratio of workers to retirees is falling. What must happen? As 401K accounts are cashed out, in order for people to have their expected retirements workers must allocate a larger portion of their paycheck to their 401K (compare to increasing payroll taxes). Otherwise there will not be buyers for the assets being sold out of the 401Ks and prices will fall leaving retirees with less in their accounts than expected (compare to benefit cuts).

I would preserve social security but reduce its benefits thru indexing the retirement age to lifespan and perhaps introducing some type of 'SSI-lite' that would let people opt for lower benefits coupled with lower taxes (or perhaps a tax refund for simplification). I would also account for our improving health (for some but not all elderly) by reconfiguring the payroll tax above 65 yrs old. I suspect by offering modest benefit increases for staying in the workforce we could exploit a Laffer Curve scenero where some retirees can find it in their interest to return to the work force coupled with a positive effect on the gov't's cash flow.

This would indeed cut SSI's expected 'return' but insurance doesn't have an expected return! I expect that my house WILL NOT burn down, I expect that my car WILL NOT be stolen or destroyed. If my expectations are accurate my insurance expenses are wasted with a -100% return. However if my expectations are wrong I'm treated to a very nice return.

SSI has a nice return if you happen to live a very long time with limited resources ($1000/mo to someone with very little assets left at 80 yrs old probably feels like $1700/mo to a welll off 68 yr old), if you are a homemaker who became widowed or if you are seriously disabled. The way to get back to a system that is more insurance and less 'all in one retirement plan' is to reduce the extras that have accidently built up in the system.

Jim Glass writes:
A fundamental promise of FDR and SS's founders was that SS would always pay a "fair" positive return to everybody. Of course its great popularity has resulted from paying *much more* than a fair return to the first generations of participants.

"SSI is an Insurance system, not a savings system. 'Returns' were never promised..."
~~~~

That is false, they were.

The founders of SS -- FDR personally -- explicitly promised "fair returns" on contributions to SS that were positive and comparable to market rates. They in fact stated they expected the average yield to participants to be a real 3%, that on government bonds, varying somewhat with the individual's situation, somewhat more for low contribution levels and less for higher ones, but positive for *everybody*.

As to the "insurance system" claim that keeps being made by defenders of the status quo, this is rather ironic:

FDR personally insisted on the "insurance system" terminology to make sure that SS *was* funded with savings, on an actuarially sound basis, with returns on benefits computed *as a commercial insurer* would compute them.

That is why he called it an "insurance system" -- while others who wanted to create an unfunded paygo system as you describe *opposed* using the "insurance" terminology.

When FDR was presented with plans that weren't funded on an actuarially sound basis as an *insurance program* would be, he literally threw the first one in the trash in the office to make his point, and later vetoed an alternative.

As a result, FDR's SS Act of 1935 was a funded plan in which benefits were explicitly tied to contributions earning a positive rate of return -- a principle publicly promised by FDR and SS's other founders.

(All the changes that turned SS into the paygo system it is now came *later*, starting in 1939 when after first being over-ridden on his veto of the first ones, FDR gave up fighting them because he had to fight WWII. Then SS was *really* turned into what it is today in the 1950s and 1960s, when it was amended to increase benefits and cut funding almost every two years. And of course it was bankrupt by 1980.)

So it's really rather bizarre to hear defenders of the SS status quo say SS doesn't have to be funded with real investments *because* it's an insurance program! FDR must be popping wheelies about this, wherever he is.

His "insurance" terminology was talking about the *funding*. What real-world insurance program isn't funded and actuarially sound?

"... only benefits if someone became a certain age..."

It's also a torturing of the English language to call that "insurance" -- that's what's known as a "pension".

E.g. if you go to the SSA web site you will find its actuaries describing the SS retirement benefit as a "defined benefit pension plan", and if you go into the historical section you will find discussions of how the pension benefit's funding moved from insurance basis -- as commercial insurers fund pensions -- in the early days, to paygo.

I mean, c'mon: Real "insurance" only pays off upon the occurrence of an insurable risk that is uncertain to happen. Health insurance doesn't pay off if you don't need medical treatment, auto insurance doesn't pay off if you don't have a car accident, and life insurance doesn't pay off if you don't die during the policy period.

If SS is "insurance against poverty in old age" then what's the insurable risk that occurred that Warren Buffett is collecting on? (And Bill Gates will too?)

Nothing. None. Warren's collecting on his credits in a defined benefit pension plan, pure and simple -- although the plan happens today to be unfunded and massively actuarially unsound on a paygo basis.

To call that pension an "insurance benefit" is a bit of word play worthy of the Ministry of Truth.

And Warren's receiving a far higher-than-market rate of return on his credited contributions - which will be paid for by today's young when they receive a negative-to-zilch return on theirs, in flat contradiction to SS's explicit founding promises.

Now, one doesn't have to be a political genius to realize that they won't like that, and that as the future majority of voters they may take it upon themselves to change the program in ways today's defenders of SS wouldn't at all approve, if their objections aren't anticipated.

Tom writes:

Don Lloyd says The returns may be real, but they provide little or no nutrition if you try to eat them. If the returns are real (and they are if they're invested in private equity and debt), then they do provide nutrition. That is, they underwrite an increased future supply of goods and services.

Boonton says SSI is an Insurance system, not a savings system. 'Returns' were never promised, only benefits if someone became a certain age (and later benefits if someone became disabled). Social Security isn't insurance, it's a transfer-payment scheme. So, do payments into SS actually yield "returns" or not? Boonton later says Of course SS yields a real return. By reducing borrowing today costs are cut in the future. If $100B less of 30 yr bonds are issued in 2000 then coupon payments that would have been made from 2001-2031 are eliminated.That's not a return, that's a re-timing of the flow of transfer payments. Reducing borrowing today requires raising taxes today, which means that today's workers are simply paying now for what they might or might not get later. They could put the same money to work in private accounts and be better off.

But, Boonton says, Private accounts are not some type of magical money machine. If the economy grows by 4% a year then by definition the economy's ability to provide retirement benefits is only growing by 4% per year.Actually, private accounts foster, and participate in, real economic growth. Social Security taxes do no such thing. There's the rub. You don't get growth without capital investment, and private accounts, unlike Social Security, underwrite capital investment.

Boonton also says 1. Increasing US borrowing today will likely increase interest rates....2. Increasing interest rates, of course, hurts stock returns attacking the benefit side of your equation. But he forgets that the diversion of money from SS taxes to private accounts will tend to drive down interest rates, thus boosting the value of stocks (for the benefit of those who choose to invest in stocks). And the rate of return on private debt will always be higher than the fictional return on government bonds held by the SS trust fund.

That's all for now. I'm enjoying this dialogue, and I hate to cut and run, but I am going to be out of touch for a couple of weeks. I'll check in when I return.

Jim Glass writes:
Of course SS yields a real return. By reducing borrowing today costs are cut in the future.

I really like this idea that not funding a major liability produces a positive real rate of return. Not saving is saving! I've got to use this in my life.

If $100B less of 30 yr bonds are issued in 2000 then coupon payments that would have been made from 2001-2031 are eliminated.

Yup. There's savings every year for 29 years, all the way until that really big coupon comes due in the end.

Jim Glass writes:
"...returns may be real, but they provide little or no nutrition if you try to eat them..."

"If the economy grows by 4% a year then by definition the economy's ability to provide retirement benefits is only growing by 4% per year."

The real economy is no kind of constraint on Social Security. Irrelevant. It's not like a plague of retirees is going to be consuming all of the nation's goods, with nobody left to produce then.

"The annual cost of Social Security benefits represents 4.3 percent of Gross Domestic Product (GDP) today and is projected to rise to 6.6 percent of GDP in 2078." SSA

Besides which, it's not the US's real economy but the world's that matters. Dollars can buy things from abroad.

SS is just a government program approaching a time of political trouble because it has made promises it can't keep. Nothing more.

Boonton writes:
Reducing borrowing today requires raising taxes today, which means that today's workers are simply paying now for what they might or might not get later. They could put the same money to work in private accounts and be better off.

Reducing Fed. gov't borrowing today reduces taxes tomorrow (to repay the principle of the bond that would have been issued today) as well as eliminates the interest payments that would have been made over those 30 years.

Assume for the sake of argument that the SSI trust fund surplus does not cause increased Federal spending. In that case a real return is generated simply by the fact that the Federal Gov't is paying down debt today (or not increasing it as fast as it would otherwise).

I really like this idea that not funding a major liability produces a positive real rate of return. Not saving is saving! I've got to use this in my life.

If $100B less of 30 yr bonds are issued in 2000 then coupon payments that would have been made from 2001-2031 are eliminated.

Yup. There's savings every year for 29 years, all the way until that really big coupon comes due in the end.

Actually you missed it.

1. It's the year 2000, SSI 'surplus' is $100B. Gov't borrowing was going to be $500B (6% interest rate for 30 years). Now it is only $400B. In 30 years it is estimated that SSI will have to pay $100B for retirees from the year 2000.

2. Gov't makes coupon payments of $24B per year for 30 years (6% * $400B to make things simple) or $720B. To make it simplier, let's assume gov't waits until 2031 to make both the coupon payments and principle payment in one big lump sum without compound interest accruing. At the end of 30 years it must also pay off the bonds of $400B which brings you up to $1,120B. Finally, SSI was amazingly accurate in its prediction of expenses this year but let's have a little fun and say its $99B (say there was a little variance in SSI's estimates of lifespans back in 2000). So by 2031 gov't must tax or borrow $1,219B.

3. Imagine in 2000 SSI was really a 'pay as you go' system and taxes were cut by $100B so as to generate no surplus in the trust fund. Gov't borrowing this year is $500B. Over 30 years the coupon payments will be $900B ($500B*6%*30yrs). At the end of 30 years gov't will have to also pay back $500B bringing you up to $1400B and finally the $99B in SSI payments leaves you at $1,499B

In this simple model you see a savings of $280B. That return, though, is shared by all of society. Taxpayers (or the loanable funds market) will find their burden in 2031 to be $280B less than what it otherwise would have been.

Of course, a more dyanamic model would account for the fact that by not borrowing $100B extra in 2000 the economy would be able to invest that much more in private capital. So by 2031 the economy should be a bit larger because of the investment made in 2000.

Boonton writes:
The real economy is no kind of constraint on Social Security. Irrelevant. It's not like a plague of retirees is going to be consuming all of the nation's goods, with nobody left to produce then. "The annual cost of Social Security benefits represents 4.3 percent of Gross Domestic Product (GDP) today and is projected to rise to 6.6 percent of GDP in 2078." SSA Besides which, it's not the US's real economy but the world's that matters. Dollars can buy things from abroad.

Imagine a 401K world that is just like our own except everyone uses 401K’s as their retirement plan. Like our world, the population is slowly getting older and the number of workers to retirees is deceasing. What has to happen in such a world?

First, as the workers retire they will fund their living by selling off assets in their 401K. In order for them to sell off their assets someone else must buy them. In other words, the young people will be buying the shares, bonds and other assets that the old people are selling. So far so good.

Compare this to SSI world. As people retire they collect benefits. The benefits come from the taxes collected from the young.

Second, as the number of retired people relative to young people increases the amount of selling exceeds buying. There are two possible responses. One response is that the price of the assets will fall and retirees will receive less for them. This would mean a lower living standard for them. Another response is that the workers will contribute more of their income towards their 401K’s thereby allowing the prices to remain the same.

Compare to SSI world. Fewer workers mean either benefits have to be cut or each worker must pay higher payroll taxes to support the retired.

In either case the economy of 2031 can be thought of as a pie. The pie represents what the economy will produce in 2031. The retired in 2031 will want to consume as much of that pie as possible as will the non-retired. While an individual’s wealth may grow faster than the economy there is no way everyone’s ability to consume (which is what wealth is) can grow faster than the entire economy at once. If the economy is growing at 4% per year then there’s no way everyone inside of it can be growing at 5%, even if they are allowed to put their social security money into the stock market!

Jim’s ‘rest of the world’ is just a distraction that adds complication without adding to understanding. Germany or Japan are not going to give the US money for its retired population in 2031. What could happen is that residents of the US could invest abroad in economies that are growing faster and thereby escape from the constraints of just the US economy. In theory, if the rest of the world looked like the US economy the SSI Trust Fund could, instead of buying US bonds at 5%, buy bonds from younger emerging countries at 10%.

But we are still left in the same boat. The consumption that takes place in 2031 is limited to the goods and services produced in 2031. The machine doesn’t change just because we zoom out to look at the world economy rather than just the US economy. Having money in your account is not the same as having goods and services. The money and its ‘return’ is like chips in a poker game. They just entitle you to a larger share of the pot but they don’t make the pot any bigger in themselves. A person whose retirement account grew at 5% while the economy grew at 0% simply is entitled to a larger share of the goods and services produced by the economy while some other people(s) are getting a lower share.
This is very sensible if you think about it without money and simply in terms of workers to retired people. If you want to go from 10 workers per retired person to 1 worker per retired person then either that one worker will be 10 times as productive or he will have a poorer living standard because of his burden (or the 10 retirees will have a less enjoyable retirement.

cb writes:

Tom,

Didn't mean to imply I was directing my mini-tirade at you, your posts make the most sense me. I'm just irritated because, and I think you said it, the entire SS system is a Ponzi scheme - new 'investors' monies are being used to pay off the first 'investors' return, that is a classic Ponzi scheme. I agree with you whole heartedly that the issue is how to change it.

I'm no economist, but I know that if we stopped SS today, the gov't would have a boatload of 'liabilities', and no assets to pay for them. One could make the argument that my future stream of taxes are an asset, but aren't those assets also increased by the liabilities of payouts to me when I retire. The bottom line is that the monies 'owed' to today's retirees suppossed to be returns on their invested tax dollars. Somebody want to show me where that is. Somebody mentioned earlier something about the SS Fund holding treasuries. Huh? Explain please.

Aside from the fact that this whole thing is a socialist Ponzi scheme, most of the posts advocating privatization seem to be basing their arguments on the belief that monies lent to the private sector have better returns than monies lent to the government. Yep. It's called a belief that individuals know better what to do with their money, acting in self-interest, than some arrogant do-gooder in Washington (hope I'm not giving away my political persuasions). However, I would like to see any change to SS to have some safeguards, I wouldn't want to see a bunch of old people w/ no income because they invested their retirement taxes in Enron-like investments (I guess mutual funds will be the only choices).

Aside, I find it interesting that this belief in the power of government to do good has seemed to correspond w/ the rise of economic illiteracy. I don't know for sure, I'm relatively young, but economic conversations in my world, on the internet, and on TV generally leave me depressed.



Bernard,

It goes without saying that establishing private accounts reduces the stream of promised benefits. That doesn't reduce outflows because people working aren't retiress, it's yesterday's workers that are now retired and receiving benefits, unless you mean to cut them off.


Boonton,

You're entire argument seems to be that by keeping the system the way it is now, we have lower debt, and therefore lower taxes, therefore young people like myself are better off than if we raise the debt. Uh, no. This is a Ponzi scheme, cut and dried, and therefore the returns generated to late investors are determined by the players making the rules, which in this case is the baby-boomers. Nothing personal, but can I have my taxes back? I prefer the capital markets. You did, though, include a paragraph at the end about money being invested in the economy would theoretically cause the economy to be larger. Yes, sir, it's called supply-side economics. You also, somewhere, made an assumption that surplus' were not spent. My entire beef with our government is that the surplus generated from baby-boomers SS taxes were indeed spent. All those fund that are suppossed to pay for their benefits are not there, gone, up in smoke in the Great Society and pork. Guess what, I'm paying for it and I'm pissed.


Whew. I feel better. Interesting nuggets to chew on -

Currently, half of all SS taxes are paid by employers (which, coincidentally is also the employee, if you're self-employed). It will be interesting to see how the proposals handle that. Are those funds going to be continued to be remitted to the gov't, I mean, it's not the employees money (arguably it's the consumers, or owners of the companies money)? Maybe those funds can be used to pay the existing benefit liabilities? I doubt it is enough, but maybe. Somebody before mentioned taxing the returns of privatized accounts. Maybe the combination could make a significant contribution.

I don't see anyway that anything of significance can be accomplished without raising the retirement age - (I agreed with you, Boonton!!)

Will significant adaptation of Health Savings Accounts lower projected health costs? I believe so, people use 'free' products more than they do when they're not.

I will not attempt a robust dynamic analysis of privatising SS, but some thoughts that I think people are not considering. When you remove, say $100B, from the Treasury coffers and put them in the hands of investment managers (who are those people going to be anyway - will they be subject to Social Investing laws, how will the participate in corporate governance issues, etc. ), it is true that treasury rates will go up statically because the demand for funds by the treasury will go up. Arguably, rates of return in corporate securities will go up because they are tied to treasuries. Modern financial theory says returns are based off the risk free rate, which is treasuries. Corporate bonds all trade to a spread over the comparable treasury. However, maybe other variables in the financial theory will change - perhaps the expected return. Then again, a $100 increase in demand for corporate securities will lower the corporate spreads. I'm tired of typing (and thinking), but my point is that there are many variables that have as yet to be covered in this thread. Somebody said something about a static analysis, all of the analysis' have been relatively static.

The bottom line is that socialist Ponzi schemes are un-American. Returns generated by the private sector are necessarily higher than gov't 'returns'. SS was a fine idea during the depression, but it's time to change. You can either tinker with the numbers and pretend it'll be OK, or you can fundamentally change the system.

I still miss the 'there's no net economic change' point, I just know that my money is being transferred to someone else, with the promise of getting a 'return', all the while the demographics of our country say otherwise, according to the experts.

SS is just a government program approaching a time of political trouble because it has made promises it can't keep. Nothing more.

So sayeth Jim Glass.

Cheers!!!

Boonton writes:

cb,

There's a lot of issues there that you touch upon. I'll just try to hit up a few of them:

1. I disagree with your name calling 'socialist', 'ponzi scheme' and so on. SSI works on a simple premise, workers pay taxes and retired people collect benefits. Benefits are indexed to how much taxes the retired paid when they were workers but unlike a Ponzi scheme SSI could run forever. As long as the economy has workers and retired people (I'll set aside Arnold's more dramatic speculations about nanotech inventing youth pills and so on).

2. The SSI money goes to purchase gov't bonds. The bonds would have otherwise been sold to private investors. This is how the gov't usually borrows money. If someone wants to loan money, they can loan it to the gov't (by buying bonds) or to a company (by buying a corporate bond) or elsewhere in the private sector (say by loaning it to their son-in-law or any other possibility).

The SSI Trust fund has only been spent if you believe the surplus in the fund caused spending to increase by an equal amount. While this may seem obvious knowing human nature and how a person would probably behave if they had access to a nice stash of money they wouldn't have to pay back for decades...I have reason to believe this is probably not the case.

3. You're going too fast over rates of return. Gov't doesn't 'invest' and earn a return. Gov't is charged a rate for borrowing so by borrowing less (or by paying off debt when it is running a surplus) gov't is reducing that cost. The rate is set by the private sector. If the private sector is charging gov't 6% then gov't can earn 6% by paying down its debt.

Things only get a little tricky if gov't pays off all of its debt and still has a surplus. This was briefly looked at near the end of the Clinton years when there were some serious projections that showed the gov't retiring nearly all of its debt in the next few decades but I wouldn't expect this to ever become a major issue.

cb writes:

Can we verify that the fund is buying treasuries w/ surplus funds? Anybody know for sure? My understanding is that the fund is basically an accounting paper fund, money going to the gov't goes into the big pool.

Boonton writes:

Yes you can verify that. The Treasury creates special non-marketable bonds that the fund purchases. Switching the bonds to marketable is basically nothing more than changing some digits in a computer database.

When you deal with this amount of money there is rarely anything 'real' there like a safe filled with $100 bills. It is all entries in computer databases and accounting records. The same goes for huge companies & other large institutions.

The magic question is did gov't spending go up because the trust fund was running a surplus? I suspect it hasn't (with some exceptions, for example Nixon enacted a bump in benefits before his re-election in order to win the senior vote).

cb writes:

Where can I find information on this subject? I would like to verify that the amount of taxes paid by workers for the last 50 years is still there. Thanks.

cb writes:

BTW, spending has been going up. After watergate, there was a specific change in the legislature that made it easier for Congress to spend, I can't remember exactly what it was. Line item veto, maybe.

Whether that spending is using SS funds, I don't know.

cb writes:

Also, are these bonds in the SS fund counted in the federal debt. If not, then this is just a fake accounting transaction.

Jim Glass writes:

The trust fund bonds have no economic meaning whatsoever, as is easy to see, and as the Treasury itself admits.

1) Balance sheet wise: They are a liability of the US government payable to the US government (*not* to social security recipients). Thus, each $1,000 bond is to the US government both an asset of $1,000 and a liability of $1,000, which equals net value of $0. (+ $1,000 - $1,000 = $0).

2) Practical effect: They have exactly $0 value in funding future SS obligations. This is easy to see by comparing our world in which they exist to one in which they didn't.

In every year after SS goes cash flow negative, with benefits paid exceeding payroll taxes collected, to pay promised benefits the government obviously will have to make up the difference between the two. Call this amount for any year $X.

So in the year, say, 2030, benefits exceed payoll tax collections by $X and the goverment has to make up the difference.

[] Without the bonds in the SS trust fund the government would have to make up the shortfall by using general revenue of $X -- by raising income taxes, cutting other spending, borrowing more, increasing admission fees to national parks, etc., in the amount $X.

[] With the bonds in the SS trust fund the government will make up the shortfall by redeeming bonds in the trust fund of value $X ... and to get the money to redeem the bonds it will have to use general revenue of $X -- by raising income taxes, cutting other spending, borrowing more, increasing admission fees to national parks, etc., in the amount $X.

As $X equals $X precisely, we can see that economically and fiscally the current situation is identical to the penny to a situation where the government never bothered to have any SS trust fund or bonds in it at all.

The trust fund is a big $0.

Now, to get back to answer the various parts of your question precisely...

"Also, are these bonds in the SS fund counted in the federal debt...

The SS trust fund bonds ...

* Are not counted on the national balance sheet because they have no net value, as asset +$1,000 plus liaiblity -$1,000 = $0.

* Are not counted in what is generally called the "national debt" -- the amount of debt the Treasury owes to the public, because the are not an obligation to the public, the Treasury owes them to itself.

* Are counted in the government's tallies of "intra-govenment debt" that it owes to itself, which are included in other measures of government debt.

You can read the numbers for both debts and their total here.

Note that during the "surplus years" the national debt (owed to the public) went down, but the intra-government debt kept rising, so the total of the two kinds kept rising.

And remember that all the bonds representing intra-government debt -- those in the SS trust fund and other similar funds for other benefit programs -- have absolutely no value, net $0, because they are owed by the government to itself.

The intra-government debt is simply an accounting tally of promised benefits the government says it will pay in the future, that it will have to collect taxes in the future to finance, in addition to the taxes it will have to collect to finance the "national debt" owed to the public.

"... If not, then this is just a fake accounting transaction."

Well, yeah. If General Motors issued General Motors bonds to General Motors, held them in an account payable to General Motors, that it legally could use for any purpose it desired, called that account a "trust fund", and said that trust fund secured its employees' pensions, everyone involved would go to jail.

And at the trial it would be like the penultimate scene in "The Producers" ... "We find the defendants incredibly guilty!"

Boonton writes:
Well, yeah. If General Motors issued General Motors bonds to General Motors, held them in an account payable to General Motors, that it legally could use for any purpose it desired, called that account a "trust fund", and said that trust fund secured its employees' pensions, everyone involved would go to jail.

Of course the Saturn Division of General Motors could borrow money from its parent company. GM would have 'loan to Saturn' as an asset & Saturn would have 'loan from GM' as a liability. When you look at GM as a whole, though, the accounting nets to zero. But this isn't a meaningless fiction, by treating Saturn as a distinct 'company within a company' GM is better able to manage the unit for profitablity and is able to track how many GM resources Saturn is using.

Accounting wise the Trust Fund is forcing the gov't to tax more than it is spending on SSI today. That is real savings which generates a net return as I demonstrated.

Where can I find information on this subject? I would like to verify that the amount of taxes paid by workers for the last 50 years is still there. Thanks

I imagine you may want to start looking at the SSI web site as well as the Treasury Dept's web site. You will probably find it very dry and technical. More importantly, it will probably be more accounting centered than economics centered.

BTW, spending has been going up. After watergate, there was a specific change in the legislature that made it easier for Congress to spend, I can't remember exactly what it was. Line item veto, maybe.

Whether that spending is using SS funds, I don't know.

Line item veto has next to nothing to do with spending. In the early 70's Nixon, in a bid to win the senior vote, raised SSI benefits dramatically. I would accept that as one instance where the fund surplus lead to higher gov't spending. Cases like that aside, I don't think there is an argument that the trust fund has caused gov't spending to increase as much or more than the funds surplus.

Don Lloyd writes:

Jim,

The trust fund bonds have no economic meaning whatsoever, as is easy to see, and as the Treasury itself admits.

It is still shocking to see someone discuss SS as if the Trust Fund actually contained any economic assets or had any economic meaning.

However, that's only the beginning of the story. Even if the Trust Fund did contain actual dollars from payroll taxes that by some miracle were not spent by the government, the Trust Fund would still be entirely economically useless in terms of meeting future SS payout obligations.

When it comes time for the Trust Fund to make a payout in excess of payroll taxes, whether it pays out dollars that it has stored for decades or whether it just asks the Treasury for newly printed ones makes no economic difference at all to the economy. It is the SS payouts that provide the purchasing power that allows retirees to bid up the prices of and acquire goods and services independent of the source of the dollars that exceed the future payroll taxes paid in.

This is why increasing and saving current payroll taxes cannot help meet future shortfalls, but are just equivalent to a current contraction of the money supply and a future printing of new dollars.

Regards, Don

Boonton writes:
This is why increasing and saving current payroll taxes cannot help meet future shortfalls, but are just equivalent to a current contraction of the money supply and a future printing of new dollars.

I disagree. Imagine that the gov't expects to have to spend $500B in 2011. Each year the gov't, to prepare for this, runs a surplus of $100B. (to keep it simple, assume the national debt is $1000B and interest rates are 0%). Each year $100B of dollar bills are collected. The gov't then buys back $100B of bonds with those dollar bills. The money in the economy doesn't contract, it flows from the taxpayer to the gov't to the owner of the bonds.

The national debt falls from $1,000B to $500B. In 2011 the gov't borrows $500B and spends it on whatever it has to. Like someone paying off his credit cards because he knows he'll have to soon put a new roof on his house, the gov't in this simplistic example successfully saved for a future expense.

Bernard Yomtov writes:

A question for Jim Glass and others who ridicule the idea of a trust fund:

Suppose the excess of payroll taxes over benefits paid - the trust fund - were invested in private securities instead of treasuries. Let's say, to be conservative, the money was invested in high-grade corporate bonds.

Would you then think there was a trust fund?

cb writes:

Good point, Bernard. The point is whether or not those debt securities held by the fund are listed as debt by the feds. If not , then it's a paper transaction. If not, then our debt is substantially larger than being reported. Conceptually, it is easy to understand that the gov't owing money to itself is a wash, the question is whether the fund is considered a separate entity. If there are assets, separate assets (treasuries or not) that exist to fund the baby boomers benefits, then it's a completely different story than if it's netted out and those funds have actually been spent.

I am also curious about the projections. I saw a program on TV the other day talking about the generation behind me, that apparently is much larger than mine. I have no idea if that's true, but if it is, wouldn't those workers taxes be ample enough to pay benefits to tomorrow's retirees. I know enough about numbers to know that slight changes in assumptions can significantly change projections, especially 50 years out.

boonton,

I've been to both sites, as well as the Cato Institutes SS site, none of which I could find anything of signifance, probably didn't look real hard. I would like to see detailed #'s and assumptions, and again, whether those securities held in the fund are counted in the debt is important.

I was wrong, it wasn't the line item veto, it was the committee and subcommittee bill of rights, effectively transferring enourmous power (along w/ other bills) away from committee chairman. I don't remember the #'s I saw, it was years ago, but there was a significant increase in spending that started in the 70's. There's a good article in the Autumn '04 Wilson Quarterly about the increased partisonship in Congress that mentions it (not in reference to increased spending, but it was a consequence nonetheless). This is not to make an argument that Congress has spent the surplus, but merely that spending has increased. I still say that if those SS bonds are in the debt #'s, then they represent assets of the fund, and therefore have not been spent, but I am not convinced of that, as of yet. Bernard is right, whether or not they're gov't bonds, corp bonds, or equity, the point is whether or not the gov't is netting those bonds when they report gov't debt. The fed, for instance, does not net their holdings of treasuries, although they do remit the interest and principal payments to the treasury, net of costs to run the fed.

Don Lloyd writes:

Boonton,

Imagine that the gov't expects to have to spend $500B in 2011. Each year the gov't, to prepare for this, runs a surplus of $100B. (to keep it simple, assume the national debt is $1000B and interest rates are 0%). Each year $100B of dollar bills are collected. The gov't then buys back $100B of bonds with those dollar bills. The money in the economy doesn't contract, it flows from the taxpayer to the gov't to the owner of the bonds.

Breaking the debt into separate payments doesn't change anything, but simply means that the analysis needs to be done 5 separate times in different time periods. You could run a surplus of $500B this year and buy back the $500B in bonds immediately. This doesn't represent a money supply contraction any more than increasing the payroll taxes in a future year does to fund the SS payments for that year. Your buying back of the bond in advance is the same as levying payroll taxes this year and paying a future SS payment now, 30 years in advance.

Regards, Don

Don Lloyd writes:

A question for Jim Glass and others who ridicule the idea of a trust fund:

Suppose the excess of payroll taxes over benefits paid - the trust fund - were invested in private securities instead of treasuries. Let's say, to be conservative, the money was invested in high-grade corporate bonds.

Would you then think there was a trust fund?

You may not remember, but Alan Greenspan once got in hot water by suggesting that one way to make sure the Trust Fund contained real economic assets would be to buy German bonds (and not bullets and bureaucrats, for example). At maturity, the proceeds would be converted to dollars and could be paid out in SS payments.

But this simply gives the SS recipients more purchasing power to bid and compete for a given supply of goods and services in the future, the same as if they were simply given newly printed dollars.

Private securities are only different to the extent that their purchase may result in a larger supply of goods and services in the future if effective use is made of the current security sale proceeds by the private sector.

Regards, Don

Jim Glass writes:
A question for Jim Glass and others who ridicule the idea of a trust fund:

Suppose the excess of payroll taxes over benefits paid - the trust fund - were invested in private securities instead of treasuries. Let's say, to be conservative, the money was invested in high-grade corporate bonds.

Would you then think there was a trust fund?

I'll be glad to answer that -- but first a question for those who enthuse over the SS trust fund as a real "trust fund".

What tangible value do you believe the SS trust fund has?

Do the bonds in the trust fund have a net positive value on the Treasury's consolidated balance sheet of the US?

Do the bonds in the trust fund reduce at all the amount of general revenue the government will have to use to meet the gap between SS payroll tax collections and SS benefits promised in out years, compared to if there was no trust fund?

Does the trust fund legally assure that SS benefits promised today will be paid, with trust fund bonds being payable to SS beneficiaries the way US Treasury bonds are to the public -- so that, say, if a future Congress decides it doesn't want to raise the general revenue to pay the bonds it can't reduce benefits by a corresponding amount?

If the answers to these questions are No, No, No, then what exactly is the substance of the trust fund that deserves respect?

Bernard Yomtov writes:

Jim,

I notice you dodged my question, despite being "glad to answer it." Ever consider a career in politics?


Still, I'll answer yours.


What tangible value do you believe the SS trust fund has?

The trust fund holds treasury securities. Those have a tangible value, damn it. Just like corporate bonds do. Anyone who thinks they don't is not worth talking to about this issue.


Do the bonds in the trust fund have a net positive value on the Treasury's consolidated balance sheet of the US?

Depends on whether you consider the trust fund a separate entity or not.

Do the bonds in the trust fund reduce at all the amount of general revenue the government will have to use to meet the gap between SS payroll tax collections and SS benefits promised in out years, compared to if there was no trust fund?

Complicated question. That's why I asked mine. The simple answer is no, but that conceals more than it reveals. The government has to pay back the bonds sold to SS, true, and use general revenue to do so. But this has nothing to do with Social Security.

The government borrows money from the trust fund to pay for other activities. If the trust fund were privately invested the government would still have to borrow that money, only it would do so in the capital markets. Instead of owing that money to the trust fund it would owe it to other bond holders. The debt owed to SS is not a SS problem. It is a problem of fiscal irresponsibility in the whole government. Notice that this problem would exist even if there were no such thing as Social Security.


Does the trust fund legally assure that SS benefits promised today will be paid, with trust fund bonds being payable to SS beneficiaries the way US Treasury bonds are to the public -- so that, say, if a future Congress decides it doesn't want to raise the general revenue to pay the bonds it can't reduce benefits by a corresponding amount?

Actually, yes. Because a future Congress may also decide it doesn't want to pay off bonds held by other investors.

And let me comment on this:

Well, yeah. If General Motors issued General Motors bonds to General Motors, held them in an account payable to General Motors, that it legally could use for any purpose it desired, called that account a "trust fund", and said that trust fund secured its employees' pensions, everyone involved would go to jail.

Complete nonsense. Companies routinely owe huge sums to their pension plans. This is exactly equivalent to what you describe. Unfunded pension liabilities in this country are about $350 billion. Until recently GM had a $10 billion unfunded pension liability. It erased this by selling bonds and using the proceeds to cover it, but there was nothing illegal about the underfunding.

When private companies go bankrupt the pensioners often lose out. Not illegal. The claim that private pension plans are wonderfully run in accordance with the highest standards of prudence and responsibility is just wrong. It doesn't help your cause.

Don Lloyd writes:

Bernard,

The trust fund holds treasury securities. Those have a tangible value, damn it. Just like corporate bonds do. Anyone who thinks they don't is not worth talking to about this issue.

See this

The government admits as much in the FY 1996 Budget document entitled Analytical Perspectives, p. 258:

"These balances are available to finance future benefit payments and other trust fund expenditures -- but only in a bookkeeping sense. Unlike the assets of private pension plans, they do not consist of real economic assets that can be drawn down in the future to fund benefits."

Regards, Don

Boonton writes:
If the answers to these questions are No, No, No, then what exactly is the substance of the trust fund that deserves respect?

As a mechanism that permits the gov't to build up savings to pay for higher expected expenses in the future.

Jim Glass writes:
If the answers to these questions are No, No, No, then what exactly is the substance of the trust fund that deserves respect?

As a mechanism that permits the gov't to build up savings to pay for higher expected expenses in the future.

Yes, your whole house of cards is built on this Joker. But talk about static analysis -- increases in tax revenue leave spending by Congress unchanged!

Let's see, how realistic is that?

When SS tax revenue started pouring from the SS Act of 1935, Congress immediately increased spending to consume it. See the SS Act of 1939 and the related political debates and legislation.

On Brad DeLong's web site you can find his conversation with Congressional Budget Office director Rudy Penner on why the spending cap restraints collapsed in 1998 -- Penner: "It was the surplus" which then was entirely from SS taxes. Congress's spending is restrained by deficits, there is nothing restraining spending of taxes -- increases in tax revenue result in quickly increased spending until the deficit restraint hits again. (see recent history!)

Or google up Daniel Patrick Moynihan -- who was both a member of the Greenspan Commission that reformed SS in 1983 and later the Chairman of the Senate Finance Committee, so he well knew about both the SS Trust Fund and government spending -- and his explanation of how the spending of the trust fund "surplus" by the government was not theft but, in his word, "embezzlement" ... and a net negative because it allowed Congress to spend on low-value projects that it didn't have the nerve to ask to increase income taxes for. Which is why his own reform proposal among other things made SS pure paygo, eliminating the extra tax collections credited to the trust fund, as an improvement in government finance.

So if folks like Moynihan and Penner were right, and spending is in fact highly elastic with tax revenue, then the answers for my questions about the effects of the current trust fund arrangement are: "No, No, No, No Savings Either, and Wasteful Spending is Deceptively Slipped Past Taxpayer Scrutiny Too."

OTOH, if you want to enter any expert testimony in support of your claim that Congressional spending is static and inelastic in response to tax revenue, so that when Congress "forces itself" to increase taxes savings in the same amount occur, I'd be glad to read it.

As far as I see, most people are of the opinion that Congress has not yet found any way to "force itself" to not spend revenue it forces itself to collect.

Jim Glass writes:
I notice you dodged my question, despite being "glad to answer it." Ever consider a career in politics?

I didn't dodge it -- just trying to define terms, to try and understand what you think a "trust fund" is and what its effects are.

What tangible value do you believe the SS trust fund has?

The trust fund holds treasury securities. Those have a tangible value, damn it. Just like corporate bonds do. Anyone who thinks they don't is not worth talking to about this issue.

Ahem. You're disagreeing with the US Treasury here. Do you think really think its opinion is so worthless?

Do the bonds in the trust fund have a net positive value on the Treasury's consolidated balance sheet of the US?

Depends on whether you consider the trust fund a separate entity or not. ...

Bah. The consolidated balance sheet of the US is a real thing, published by the Treasury.

The bonds are an asset on it or they are not. What "you consider" has nothing to do with it. It is a matter of fact.

Should I politely suggest that anyone who can't answer this question of simple fact is not worth talking to about this issue?

No, I'll be more polite and note (again) that the bonds are not on the balance sheet, because the Treasury deems they have no tangible value to it -- unlike other assets, such as cash and foreign bonds it may hold -- and explain why they don't.

Imagine that you owe $10,000 to, say, you're mother, due on 1/1/2010. To assure you will be able to pay it you write an IOU to yourself, "I owe me $10,000 due 1/1/2010". You then put in an envelope and label it "trust fund".

Have you created an asset of tangible value to you that will secure your debt to your mother?

Of course not. Why, exactly? Because it leaves your position totally unchanged. Before, you had to pay the $10,000 to your mother from your income (or other assets) and after writing the IOU to yourself you still have to pay the $10,000 to your mother from your income (or other assets). So, $0 difference to you, $0 value to you.

Now, if you have a very good credit rating you might be able to sell your $10,000 IOU to a bank or a German or somebody (assuming you change it from "I owe me" to "I owe bearer") for the full $10,000. It will then be an asset worth $10,000 to them.

But you still wind up net $0. Because if you use the $10,000 you get from the German to pay your mother, then you then have to pay $10,000 to the German at the same time from your income (or other assets), just as before. So the IOU you write to yourself is never worth anything to you. It's a big $0.

It is exactly the same situation with the SS trust fund bonds. They are IOUs the Treasury issues to itself. So they have $0 value to it. So it does not report them on its balance sheet. When it redeems them to fund future SS benefits it will have to raise general revenue exactly as if they did not exist. Zero change to it, zero value.

It could sell the bonds on the market -- but this produces zero ($0) net value to it because then when the bonds are redeemed it will have to pay the bond buyers the exact same amount it would have used to pay SS benefits. Again, $0 change to it, zero value.

OK? That's why the SS bonds are not on the Treasury's balance sheet -- and there's nothing subjective about it.

As Don noted, and I did before, this is described every year by the government itself in the Perspectives, so it's not a secret.

Does the trust fund legally assure that SS benefits promised today will be paid, with trust fund bonds being payable to SS beneficiaries the way US Treasury bonds are to the public -- so that, say, if a future Congress decides it doesn't want to raise the general revenue to pay the bonds it can't reduce benefits by a corresponding amount?

Actually, yes. Because a future Congress may also decide it doesn't want to pay off bonds held by other investors.

Wrong again. The Constitution requires full payment on all bonds payable to the public. Congress can't decide not to pay. But the Treasury can cancel the SS bonds owed to itself at will, or use them for any non-SS purpose it wishes, as the Perspectives notes. And the Supreme Court has said Congress can cut or eliminate any and all SS benefits at will -- even those funded by the "trust fund"!-- just like any other government program.

When you make mistakes like these, you know, maybe you should be less eager to declare others "not worth talking to about this issue", eh?

Jim Glass writes:

Now to answer Bernard's question.

But first: What is a trust? Is the SS "trust" a trust? The government in the Perspectives says it is not in both the legal and common senses of the word, and I don't know why anybody would want to argue with the government on this, but anyhow...

From the law dictionary on my desk:

"Trust: a legal entity created to hold assets for the benefit of certain persons or entities who have legal right to such benefit, with a trustee managing the trust..."

Three requirements there: (1) Legal entity, (2) holds assets, (3) for the benefit of third parties who have defined legal rights in the assets.

Let's say I owe Bernard an unsecured debt of $100,000 due 1/1/2020. He observes that I'm not setting aside any money to pay it, so he gets nervous and insists I set up and fund a "trust" to do so.

OK, I write myself an IOU "I owe me $100,000 due 1/1/2020", put it in an envelope, and write "trust fund" on the outside. Should Bernard be happy?

I hope not! This arrangement is 0 for 3. There is no separate legal entity. The note to myself is worth $0 on my balance sheet -- it's not an asset. And if I win the lottery or something and actually pay $100,000 to myself in 2020 Bernard has no claim to that money via the IOU -- I can use it to pay off my gambling debts and Bernard gets stiffed. Bernard still objects!

So, to make him happy this time I go to my lawyer and set up a legal trust -- a separate legal entity, managed by a bank trust department -- contribute to the trust a AAA grade corporate zero-coupon bond maturing at $100,000 on 1/1/2020, and set the trust terms so the $100,000 will be paid by it to Bernard on 1/1/2020 by law -- I can't stop payment or get that money myself.

Now all three requirements are met, we have a trust funding my debt to him, and Bernard is happy.

Now, say, Bernard expects to receive SS payments in 2020 of which $X will be financed by the SS trust fund. Is the SS trust fund any more of a legal trust fund than my envelope labeled "trust fund"? Let's see..

* The SS bonds are just a government IOU to itself which the Treasury gives $0 asset value on its balance sheet. No asset. 0 for 1.

* There is no separate legal entity holding the bonds apart from the Treasury. (This is not a subjective, "it depends if you consider.." thing. There's not.) 0 for 2.

* The bonds, such as they are, are payable to the Treasury, not to SS beneficiaries, and Congress can stop payment on all SS benefits at any time, including all those from the trust fund. It can cancel the trust fund bonds, use them to pay Medicare instead, terminate SS entirely, whatever. Payment to Bernard is not legally enforceable (unlike payment to Bernard on the T-bonds he personally ownes). 0 for 3.

No trust by the legal definition and common useage of the term. Nope.

Bernard asks:

Suppose the excess of payroll taxes over benefits paid - the trust fund - were invested in private securities instead of treasuries. Let's say, to be conservative, the money was invested in high-grade corporate bonds. Would you then think there was a trust fund?

Let's see again:

Such bonds are assets on the Treasury's balance sheet (they are not an IOU to oneself, but from someone else - big difference!) so we have assets now. 1 of 3.

But if all else stays the same -- the bonds are owned by the Treasury not a separate legal entity, are payable to the Treasury, and Congress can use their proceeds as it wishes, other than for SS -- then we are stuck at 1 of 3. No trust.

Try this: Congress separates the SS trust fund as a separate legal entity. It converts the SS bonds to public Treasury bonds subject to the Constitutional guarantee of payment, like other T-bonds, and contributes them to the trust. The trust fund is legally committed -- just like private-sector retirement plan trusts -- to pay to SS beneficiaries the bond proceeds, even if Congress kills all the rest of Social Security.

Now we have all three requirements met -- even using government bonds, because the T-bonds are assets to the trust, since it is a separate legal entity from the government, so the bonds owned by the trust are not an IOU to itself. (While on the Treasury's balance sheet they now are a liability, not a $0.)

Yes, now we have a trust! Even funded with government bonds!

Of course, a trust fund funded with such bonds would reduce the government's future tax cost of funding SS by $0, compared to if the trust fund did not exist, as explained at length previously.

But that's a different issue from whether there's a trust that meets the legal definition and common useage of the word or not.

With a real, legal trust, Congress would be legally (Constitutionally as per the T-bonds) on the hook for the portion of SS benefits funded by the trust fund. It could play only with the other 90%.

Jim Glass writes:
And let me comment on this:

Well, yeah. If General Motors issued General Motors bonds to General Motors, held them in an account payable to General Motors, that it legally could use for any purpose it desired, called that account a "trust fund", and said that trust fund secured its employees' pensions, everyone involved would go to jail.

Complete nonsense. Companies routinely owe huge sums to their pension plans. This is exactly equivalent to what you describe.

"Equivalent" -- nonsense. That's what SS actually does claiming it is funding benefits through the trust fund.

No business can claim it funds a retirement plan like that without go-to-jail time -- because, as you point out, it is the same as leaving it unfunded. Glad you agree.

To claim you have funded a plan while leaving it effectively unfunded is dishonest!

No business ever funds a defined benefit pension plan by contributing its own bonds to the plan. It is absolutely illegal because it is a transaction without economic substance, giving an IOU to itself with $0 asset value to it, to the plan.

Hey, wouldn't it be just great if we could all pay our debts like that, by writing out 30-year IOUs to ourselves and distributing them to our creditors!

Unfunded pension liabilities in this country are about $350 billion.

Um, we are talking about funding plans, not unfunded ones.

No company ever funds a plan with its own bonds.

Until recently GM had a $10 billion unfunded pension liability. It erased this by selling bonds and using the proceeds to cover it...

Yes, it sold its bonds to third parties and contributed the cash it received in return -- an asset in anyone's book -- to its pension plan.

It did not contribute its own bonds to its pension plan. Nobody does that.

Except the government with Social Security, of course.

... but there was nothing illegal about the underfunding.

As a matter of fact, GM had to ask the government for permission to underfund, by requesting a Section 412 waiver of the funding requirement, justified by business hardship and a promise to make good soon, with evidence that it would be able to do so. As do all other employers that underfund pension plans must do.

It is most definitely illegal to underfund without getting a 412 waiver that grants government permission to do so.

Many say the government is too generous in issuing such waivers, but that is another story.

Of course, the government is very generous with itself on that score.

Boonton writes:
Congress's spending is restrained by deficits, there is nothing restraining spending of taxes -- increases in tax revenue result in quickly increased spending until the deficit restraint hits again. (see recent history!)

Really? Explain why spending was rather well restrained during the Clinton years yet has gone thru the roof one the surplus disappeared? You nicely express the ATM theory of gov't.

The ATM theory is this, you go to the ATM to take out $20 for lunch and you notice on the receipt you have $100 more in the bank than you thought. You shrug and say what the hell and take out another $40 for movies and dinner. That's not how gov't works, no one says 'you have $1.1T to spend, figure out what you want to spend it on'. It's more of 'we want to send $200B here, $500B there and spending on interest is automatically $350B and so on'.

Like the supposed collapse of American education, you'll have convetional wisdom on your side but not the facts.

Boonton writes:
Bah. The consolidated balance sheet of the US is a real thing, published by the Treasury.

The bonds are an asset on it or they are not. What "you consider" has nothing to do with it. It is a matter of fact.

Should I politely suggest that anyone who can't answer this question of simple fact is not worth talking to about this issue?...No, I'll be more polite and note (again) that the bonds are not on the balance sheet, because the Treasury deems they have no tangible value to it -- unlike other assets, such as cash and foreign bonds it may hold -- and explain why they don't.

The balance sheet is not a 'real thing'. Balance sheet's are accounting conventions that describe an underlying business but they are just that conventions. Business existed before double entry bookkeeping evolved into our modern day financial statements and accounting principles.

Imagine that you owe $10,000 to, say, you're mother, due on 1/1/2010. To assure you will be able to pay it you write an IOU to yourself, "I owe me $10,000 due 1/1/2010". You then put in an envelope and label it "trust fund".

Have you created an asset of tangible value to you that will secure your debt to your mother?

Of course not. Why, exactly? Because it leaves your position totally unchanged. Before, you had to pay the $10,000 to your mother from your income (or other assets) and after writing the IOU to yourself you still have to pay the $10,000 to your mother from your income (or other assets). So, $0 difference to you, $0 value to you.

Cute and I agree you have created value in your little trust account there. So what? You aren't describing SSI because SSI actually runs a surplus taking in more in taxes than it pays out in a benefit. In your little example, that would be like saying "I owe mom $10,000, therefore I'll put my overtime income of $250 a month into paying down my credit card balance".

Whether or not you save $10,000 by this method, every $250 payment you make strengthens your financial position for the day of reckoning when you'll have to pay your mom.

Bernard Yomtov writes:

Jim,

Thanks for the seminar. I'll try to respond on the trust fund issue. Meanwhile, suffice it to say that since the government is not a private organization, and has the power to change laws and set taxes, the notions are somewhat different, but not nearly to the degree you claim. Note that the bonds held by Social Security are in fact backed by the full faith and credit of the US government.

(BTW, where can I find this government balance sheet you refer to?)

In any case, nothing you wrote refutes my basic point:

The crisis facing Social Security, if there is one, stems from a presumed inability of the government to pay back the money it has borrowed from Social Security.

If there were no Social Security the government would have borrowed the money elsewhere, and be presumptively equally unable to repay it.

Hence the "Social Security crisis" is not a "Social Security crisis" at all, but a general fiscal crisis.

You want to address it before it gets too bad? Fine. So do I. You think there's a better way to do Social Security than what we have? Fine. Let's hear it. But let's recognize that these are two different issues.

Jim Glass writes:
Congress's spending is restrained by deficits, there is nothing restraining spending of taxes -- increases in tax revenue result in quickly increased spending until the deficit restraint hits again. (see recent history!)

Really? Explain why spending was rather well restrained during the Clinton years yet has gone thru the roof one the surplus disappeared?

====
Um, spending (including tax cuts) went through the roof before the surplus disappeared -- as soon as it arrived, during the Clinton years, with the big bipartisan spending hikes before the 1998 election.

They are what made the surplus disappear, right? If you don't believe so, tell us what you think did make the surplus disappear.

And once again, Rudy Penner, the head of the CBO, did explain that.

The post-Reagan deficits resulted in the bipartisan spending caps being imposed. Look at the legislative record. Then the spending caps blew apart when the surplus hit and the restraint imposed by the deficits disappeared.

At that point -- as *you just said* -- in response the to the unexpected surplus that was projected to last a decade, Congress and White House (and both parties, starting in 1998) went hog wild with spending and tax cuts wiping out the surplus until the unexpectedly large deficit arose after the unexpected recession, and the spending brakes went on, stabilizing the deficit where it is now.

And now you actually are citing this experience of the politicians immediately wiping out an unexpected a surplus that could have lasted a decade in support of your claim that government spending (including tax cuts) is inelastic to revenue increase! And as an example of how government saves!! Gosh golly.

Look, you're arguing with Penner, the former CBO head, not me, when you deny that the deficit restrained spending and the elimination of the deficit resulted in the spending caps being blown off. If you want to do that be my guest.

But unless you can cite some academic evidence or similar persuasive authority to support this strange and heroic claim that gov't spending (including tax cuts) is inelastic to revenue changes -- as I invited you to do before -- then you are pretty lonely on this. And the experience of 1997-2003 isn't exactly the best empirical support for that claim!

Jim Glass writes:
Bah. The consolidated balance sheet of the US is a real thing, published by the Treasury.

The bonds are an asset on it or they are not. What "you consider" has nothing to do with it. It is a matter of fact...

The balance sheet is not a 'real thing'. Balance sheet's are accounting conventions that describe an underlying business but they are just that conventions...

The consolidated balance sheet of the US govermment is a real thing published by the Treasury. The asset value to it of all securities owned by it are listed on it.

If you want to hand-wave away the accounting rules involved as mere "conventions" with no real meaning that can be adjusted to fit the needs of your argument, fine.

Enron adjusted those mere conventions to fit the needs of its business, and there were hardly any
practical consequences at all, for the business or its employees' retirement accounts or anybody!

Jim Glass writes:
"Note that the bonds held by Social Security are in fact backed by the full faith and credit of the US government."

Really? To whom?? For that to be so, somebody has to be able to *invoke* that backing. Who can? Can you? Nope. Can any citizen? Nope.

Only the party that owns the bonds can. That's the government. Dang, invoking the full faith and credit of itself against itself to force itself to pay itself is going to have a big impact on its financial situation. So that "full faith and credit" business is very meaningful!

"(BTW, where can I find this government balance sheet you refer to?)"

2003 Financial Report of the United States Government

From the balance sheet...
Assets/other assets/securities: $37.7 billion.

That's somewhat less than the $1.5 trillion of bonds in the SS trust fund.

Ah, there's a note:

"intragovernmental debt holdings [the SS trust fund bonds, etc] are eliminated in the consolidation of these financial statements"

... for reasons previously explained. And also explained in the Perspectives.

"In any case, nothing you wrote refutes my basic point:

The crisis facing Social Security, if there is one, stems from a presumed inability of the government to pay back the money it has borrowed from Social Security."

False, wrong. That's only about $1.5 trillion. The gov't could pay that over future decades easy.

The problem is the $10 trillion current value unfunded SS liabilities on top of that -- which the gov't might be able to fund over future decades, if there wasn't another $40 trillion to $60 trillion current value of other unfunded retirement liabilities on top of that.

With all these liabilities pegged to grow with growth in the economy, so there is no way economic growth is going to make them go away.

There is no way to close a funding gap of $50 trillion to $70 trillion current value without some combination of major benefit cuts and major tax increases. And the only way to mitigate those to some extent is with an injection of private investment into the system to finance some of the long-term future costs with real savings instead of taxes and cuts.

The SS trust fund is a red herring -- near irrelevant to all this, because even nominally its balance is trivial compared to >$50 trillion, while in economic reality its value at funding future expenditures is exactly $0.00.

The only relevance it has is political in that millions of people beleive it really secures their future SS benefits by financing them in advance with real assets that have been saved to pay them in spite of whatever fiscal problems might hit the government -- as a real trust would -- in spite of the fact that of course it does no such thing.

If there were no Social Security the government would have borrowed the money elsewhere, and be presumptively equally unable to repay it.

Hence the "Social Security crisis" is not a "Social Security crisis" at all, but a general fiscal crisis.

??? If there were no Social Security and Medicare, etc., there'd be no $50 trillion to $70 trillion current value of unfunded liabilities and there'd be no fiscal problem at all. Sure.

C'mon, you don't think that the "coming insolvency of the US" that DeLong is always going on about is due to a piddling $1.5 trillion owed to the SS trust fund, do you?

Or that the 20% of GDP that will be going to interest on the debt alone as projected by GAO in 2045 if entitlements aren't brought into line is due just to the interest on a mere $1.5T borroweed from SS?

Boonton writes:
And once again, Rudy Penner, the head of the CBO, did explain that.

The post-Reagan deficits resulted in the bipartisan spending caps being imposed. Look at the legislative record. Then the spending caps blew apart when the surplus hit and the restraint imposed by the deficits disappeared.

At that point -- as *you just said* -- in response the to the unexpected surplus that was projected to last a decade, Congress and White House (and both parties, starting in 1998) went hog wild with spending and tax cuts wiping out the surplus until the unexpectedly large deficit arose after the unexpected recession, and the spending brakes went on, stabilizing the deficit where it is now.

What is missing here is the fact that most of the battle over the budget comes down to discretionary spending, the portion of the budget that Congress actually decides line by line each year. This piece of the budget, because Congress feels it has direct control of it, is what is usually debated and argued about. This is also the part of the budget to which that spending caps and other devices are usually applied. Unfortunately (or fortunately depending on your POV) this is only a fraction of the total budget. Many items such as interest, unemployment insurance payments, SSI, Medicare/Medicaid and others are on 'automatic pilot'.

Nobody who turned 65 or who was laid off in the Reagan years was told to wait for his check because 'spending' caps were being imposed due to the deficit.

Boonton writes:
But unless you can cite some academic evidence or similar persuasive authority to support this strange and heroic claim that gov't spending (including tax cuts) is inelastic to revenue changes -- as I invited you to do before -- then you are pretty lonely on this. And the experience of 1997-2003 isn't exactly the best empirical support for that claim!

Let's start with:
http://www.cbo.gov/showdoc.cfm?index=4985&sequence=11 the table that shows revenues and outlays and public debt as a % of GDP.

Going from 1962 to 2003 revenue as a % of GDP has been between 16-20% with an average of 18.2% (note that 2003 is unusually low at 16.5%). I added a column to the 'debt held by the public' to indicate change from the previous year. So, for example, in 1962 debt was 43.6% and in 1962 it was 42.3% and the change was -1.3.

So I build a simple regression model:

Change in debt = B + B2 * Revenue

If your theory is correct, B2 will either be 0 or positive. Why? Because as revenue goes up total spending will increase to either match the revenue increase or will be even greater resulting in additional debt. The SSI trust fund is netted into these figures so it doesn't matter if the spending increase is discretionary or caused by SSI paying out larger benefits.

The results:

Regression Statistics
Multiple R 0.556718244
R Square 0.309935203
Adjusted R Square 0.292241234
Standard Error 1.81363567
Observations 41

Coeff.____________T Stat
B = 22.7059_______4.146
B1 = -1.2551______-4.185

What this means is that gov't borrowing from the public goes down as revenue goes up as a % of GDP. At least when we look at over 40 years of observations. Spending may indeed rise but paying off debt increases *more* which means that the SSI trust fund is a vehicle for gov't saving.

Boonton writes:
The consolidated balance sheet of the US govermment is a real thing published by the Treasury. The asset value to it of all securities owned by it are listed on it.

If you want to hand-wave away the accounting rules involved as mere "conventions" with no real meaning that can be adjusted to fit the needs of your argument, fine.

First of all accounting rules developed as business needs demanded more and more objective statements. At one time, for example, many businesses used simple double entry bookkeeping to produce an income statement and that was all they needed. As business became larger and securities laws required publically traded businesses to audit and make public their financial statements the conventions became more rigerous. Gov't accounting, though, is not as advanced hence the Treasury Department's 'balance sheet' should not be viewed in the same light as Microsoft's.

Second even with businesses and GAAP there are plenty of 'mere conventions' in balance sheets. Depreciation of property is a big one. If you own several homes each year your business's balance sheet is going down because that property is being 'depreciated'. In reality your assets are probably increasing since real estate has been appreciating in value. Likewise there are times when GAAP requires one to report stocks & bonds at their purchase price even though 'objectively' it is easy to find their current market value. Inventory as well is often reported at its purchase price even though its true value may be far less.

There is sometimes a real distinction between accounting substance and economic substance. Trying to analyze a national system like SSI using GM's pension plan as a template is not always the best way.

BTW, did you respond to Banard's question? Where does the Treasury publish this 'balance sheet'? To be honest, this is the first time I even heard about it and I question whether it has any economic significance.

Boonton writes:

My apologies Jim, I see where you linked to the T-Dept's 'Balance Sheet'. I still question whether this is a useful document or more of an accounting convention without economic substance.

Out of curiosity, I did another regression analysis with our table. I compared the change in revenue as a % of GDP to the change in outlays. In order to capture SSI I added the SSI surplus or deficit to outlays. The model is:

Change in net outlays = B + B1 * change in revenue

If Jim's assertion is correct, B1 needs to be positive. If revenue goes up then outlays goes up as well. The results?

Regression Statistics
Multiple R 0.195100212
R Square 0.038064093
Adjusted R Square 0.01339907
Standard Error 0.840952055
Observations 41

____________df________SS________MS
Regression __1_____1.091381_____1.091381096
F_____________Significance F
1.543241715_____0.221554244

The Coefficients are:
Coefficients_____________________t stat
Intercept_______0.060844566______0.463
X Variable 1____-0.205372797_____-1.24

The regression tells us that the change in spending actually goes down when revenue goes up! This isn't as shocking as it initially sounds. Many of the gov't's revenue sources are tied to the overall economy as is its spending. When times are good revenue goes up automatically and many spending elements like unemployment and welfare go down. Even SSI might experience spending cuts in a good economy, some people will choose to keep working if they are doing very well thereby cutting SSI payments and increasing both SSI revenue and general tax revenue.

Unfortunately, though, while Jim's theory can be rejected the statistics are not significant enough to accept an alternate theory that revenue increases actually causes spending cuts (look at the t-scores, less than 80% confidence).

Boonton writes:
??? If there were no Social Security and Medicare, etc., there'd be no $50 trillion to $70 trillion current value of unfunded liabilities and there'd be no fiscal problem at all. Sure.

C'mon, you don't think that the "coming insolvency of the US" that DeLong is always going on about is due to a piddling $1.5 trillion owed to the SS trust fund, do you?

Or that the 20% of GDP that will be going to interest on the debt alone as projected by GAO in 2045 if entitlements aren't brought into line is due just to the interest on a mere $1.5T borroweed from SS?

Hang on a second Jim! You're lumping SSI in with Medicare. They are actually two very different programs. SSI is either totally sound or maybe 1% off depending on how economic growth goes over the next 40 years or so. Medicare is in for big problems because the cost of medical care is growing faster than the economy. Unlike SSI, though, I'm very doubtful whether we can really project the cost of providing retirees medical care 40 years out into the future.

Bernard Yomtov writes:

Jim,

Calm down. As Boonton points out, you've suddenluy switched gears from SS to "SS and Medicare." And if you want to start taking PV's out to 75 years and beyond be my guest. Just remember not to change your discount rates or growth assumptions by a fraction of a percent, or you'll get pretty dramatically different results. Given the confidence intervals of these estimates they have pretty limited usefulness. What is the PV of the defense budget out to infinity? Pretty big, I'd guess.

The money held in the Social Security Trust Fund is held in treasury bonds backed by the full faith and credit of the government. By law these funds can only be used to pay Social Security benefits. In other words, the treasury is legally obligated to provide the money to pay the benefits. Can Congress change the law, change the benefits? Sure. But that doesn't mean that the system as it operates today, under the laws governing it, is in serious trouble, or that the whole thing is a fiction. Is your 401K a fiction? What if Congress puts a tax on the assets in it? After all,if we have fiscal crisis who knows what will happen.

If you don't like Social Security, fine. But stop the Chicken Little act.

cb writes:

I don't understand what is wrong with you guys. Everybody gets that the fund holds treasury securities, or 'certificates' as they're properly called. However, the way the gov't does their accounting, they offset those securities against the debt because 'we owe it to ourselves'. The problem is the certificates are essentially an INCREASE in debt, because the gov't is spending the cash in the fund and replacing it w/ a piece of paper. That debt issuance is not reflected in the debt figures, which is duplicitious. Even the Treasury says so, so I don't get why you guys keep saying it isn't. If the fund has 'assets', shouldn't the other party in the transaction have 'debt'? Yes, but they don't count it that way.

Boonton writes:
The problem is the certificates are essentially an INCREASE in debt, because the gov't is spending the cash in the fund and replacing it w/ a piece of paper. That debt issuance is not reflected in the debt figures, which is duplicitious. Even the Treasury says so, so I don't get why you guys keep saying it isn't. If the fund has 'assets', shouldn't the other party in the transaction have 'debt'? Yes, but they don't count it that way.

What you are missing is the fact that what is essentially happening is that gov't is taking debt out of the public's hands today and putting it in the trust fund to get rid of cash that is not needed to pay benefits. Tomorrow it will take that debt out of the fund and put it back in the public's hands in exchange for cash to pay benefits.

If you need an accounting analogy consider a cash rich corporation that buys its own stock back today but plans to sell its stock again on the market years later.

Jim Glass writes:
"The money held in the Social Security Trust Fund is held in treasury bonds backed by the full faith and credit of the government."

You didn't answer the question -- who can invoke the backing? The government against *itself*? ;-)

Did you ever promise your creditors: "Don't worrry, I'll collect that guarantee against myself if I don't want to pay later. So all my debts to you are secure!"

"By law these funds can only be used to pay Social Security benefits. "
What law??

Bernard, it has already been pointed out to you, with a link to the Perspectives that explain it, that this not so. It is false.

There is not a thing in the world preventing Congress from giving a haircut to SS and applying any or all of the trust fund bond money to anythinbg else if it wants. Say, Medicare, should Medicare costs shoot up and Congress want to fund it without raising taxes. Or whatever.

Being that this has been pointed out to you repeatedly, you are engaging in simple repetitive denial here.

Is your 401K a fiction?

Of course not. It is a real trust that meets all three requirements -- including owning real assets under anybody's accounting rules. That's different from the SS trust, you will agree, right?

What if Congress puts a tax on the assets in it

Congress does tax it. But it can't confiscate it, the Constitution says so. Thus we know for a fact that the assets in it will be delivered to me upon retirement. But we do not know, for a fact, how much anyone will get out of the SS trust fund. This will be entirely a function of the politics of decades from now.

We also know that SS including the trust fund is 25% underfunded overall, 40% underfunded for the young. So we know as mathematical certainty that somebody is going to get screwed relative to what's been promised them in current law -- because it's mathematically impossible to close that gap without somebody getting that much less in benefits or paying that much more in taxes or both, relative to what they've been told in their SS Personal Earnings and Benefits Statements that they get today, which lay out their promised benefits.

Now, if you really want to insist there's no difference between that and a funded private sector defined benefit pension plan holding real assets in a real trust that can't be stopped from paying out full promised benefits ... well, I don't know what to say.

"...stop the Chicken Little act"

Who, me? Little ol' me?

When DeLong and Krugman go on and on about the looming "insolvency of the United States"?

And when the GAO report I linked to before projects the 'end of government' in 2045 when just interest on the debt by istself is a greater portion of GDP than the entire federal govt today?

And when the Comptroller General of the U.S. describes the future finances of the US thusly:

"'We start with a haircut,' he said, thumbing through the pages that show federal spending starting to explode as Baby Boomers reach retirement. 'The next page is a scalping, and then we go to decapitation and finally disembodiment.'"

And you call me 'Chicken Little', sheesh.

What do you call all them?

Boonton writes:
You didn't answer the question -- who can invoke the backing? The government against *itself*? ;-)

Did you ever promise your creditors: "Don't worrry, I'll collect that guarantee against myself if I don't want to pay later. So all my debts to you are secure!"

This is starting to sound like an argument with someone who demands to know what backs US dollars. What isn't reflected on the balance sheet, Jim, and what doesn't apply to the case of an individual who owes a creditor is that the US gov't enjoys taxing authority. I do not hence Visa will just have to trust me that I will have a good job for the next 30 years and will be able to pay my credit cards.

cb writes:
What you are missing is the fact that what is essentially happening is that gov't is taking debt out of the public's hands today and putting it in the trust fund to get rid of cash that is not needed to pay benefits. Tomorrow it will take that debt out of the fund and put it back in the public's hands in exchange for cash to pay benefits.

If you need an accounting analogy consider a cash rich corporation that buys its own stock back today but plans to sell its stock again on the market years later.

What you are missing is that the gov't is spending the cash, and putting certificates in the fund w/o increasing their reported debt. Your analogy is wrong, because the gov't is not spending the cash to buy assets. It may be legally possible to at a future date sell the certificates, but the point is that those certificates are not counted as debt. Again, as the treasury admits, there are no assets in the fund.

Boonton writes:

The Treasury does report the debt owed to the Trust Fund, it simply isn't listed under 'debt held by the public' because it isn't held by the public. Likewise a company that buys back some of its own shares will 'keep them' in a special account that shows up on the stockholders equity portion of the balance sheet.

Don Lloyd writes:

Boonton,

Likewise a company that buys back some of its own shares will 'keep them' in a special account that shows up on the stockholders equity portion of the balance sheet.

Your analogy is better than you know. A company that holds its own stock is NOT holding an economic asset. The stock that it holds can be either destroyed or increased without effectice limit without impacting either the company or its shareholders.

Regards, Don

Boonton writes:

So let's imagine a company that is cash rich right now but suspects it will have a major expense in 10 years. Let's suppose this expense is so large that it knows there is no way it can pay for it without borrowing.

If the company doesn't feel confident in its ability to invest the large lump of cash today in other companies and it doesn't feel there are worthwhile internal investments to make at this time, then what would be wrong with the company buying back its own stock? Granted the future expense is still there but by reducing equity held by the public today the company has the option of reissuing the shares in the future.

This leads me to another issue I have with Jim's 'Treasury Balance Sheet'. A company's balance sheet should list all the assets a company has. A company can only make use of those assets to pay its liabilities and have equity left over. Government can own assets (like military bases, roads, post offices etc.) but gov't's primary source of revenue is taxation not earnings on its assets (in fact, most of its assets will have no earnings at all).

Since taxation is the primary source of gov't revenue the true underlying asset of a gov't is the economy that it has jurisdiction to tax but to my knowledge no one has developed a method of depicting that in a meaningful way on a balance sheet.

Comments for this entry have been closed
Return to top