How Phony Is The Social Security Trust Fund?

Pretty phony.  The NY Times Magazine has a long article on Social Security by Roger Lowenstein which both Kevin Drum and Paul Krugman endorse, so we know there will be trouble.  Ducking the argument about the "reality" of the Social Security Trust Fund is just one of the article's distortions.  Here we go:

IS THE TRUST FUND TRUSTWORTHY?

The second debate concerning solvency is over whether the securities in the trust fund will be honored or whether, in Moore's pointed imagery, the fund will resemble a bank ''after it's been robbed by Bonnie and Clyde.'' This seems an odd preoccupation. Social Security does not own junk bonds or third-world debt; it invests in U.S. Treasuries, considered the safest investment on the planet. Since 1970 there have been 11 years in which Social Security has operated at a deficit; each time, it redeemed bonds from the trust fund without a fuss. Goss, the agency's actuary, says he has no doubt it will be able to do so again. ''Absolutely,'' he said when asked if the trust-fund bonds are sound.

This isn't what some conservatives have said. Paul O'Neill, the former treasury secretary, went so far as to say that Social Security has no assets. In anti-Social Security literature, the ''no assets'' contention isn't even debated; it's treated as gospel. According to Michael Tanner, head of the Cato Institute Project on Social Security Choice, the agency's pauperism has turned America's seniors into ''supplicants'': after working and paying taxes their entire lives, ''they earn the privilege of going hat in hand to the government and hoping that politicians decide to give them some money for retirement.'' The implication is that the money isn't there: graybeards will have to beg for it.

"Treated as gospel".  Well, I suppose so, in the same way that "2+2=4" is treated as gospel.  And the "conservative" view, properly characterized, is not even in dispute amongst serious economists.  But to be clear upfront - responsible conservatives are not saying that the Trust Funds assets will not be honored by the US Treasury.

Let's roll through this:  the "conservative" complaint is not that the trust fund assets are "worthless"; it is that they have virtually no legal or economic significance.

Imagine that we come to 2018, and Social Security payroll tax receipts are less than legally mandated benefits.  Because we have a trust fund, what happens?  The Soc Sec Administration collects interest from the Treasury on its "assets", turns around, and says, we need the cash to cover the benefit checks we are writing.  The Treasury then makes a payment to the Soc Sec Administration from the general fund.  To cover this, the Treasury must, working with Congress, either (a) run a surplus in other funding; (b) increase outstanding debt, (c) reduce other spending, or (d) raise new taxes.

Now, imagine that we did not have a trust fund.  Congress has still mandated a certain level of benefits for 2018, and the payroll tax will not be sufficient to cover them (under current projections).  Consequently, the Soc Sec Administration will go to the Treasury to cover the shortfall.  And to cover the shortfall, the Treasury, working with Congress, will either (a) run a surplus in other funding; (b) increase outstanding debt, (c) reduce other spending, or (d) raise new taxes.

Please note that the choices are the same whether we have a trust fund or not.  In that economic sense, the trust fund assets are meaningless.  And this point is not in dispute among serious observers.  In 1999 (Pre-BushCo, for you libs out there), the US Treasury said this about the trust funds:

The Federal budget meaning of the term ``trust'' differs significantly from the private sector usage. The beneficiary of a private trust owns the trust's income and often its assets. A custodian manages the assets on behalf of the beneficiary according to the stipulations of the trust, which he cannot change unilaterally. In contrast, the Federal Government owns the assets and earnings of most Federal trust funds, and it can unilaterally raise or lower future trust fund collections and payments, or change the purpose for which the collections are used, by changing existing law.

...These [trust fund] 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.

Instead, they are claims on the Treasury that, when redeemed, will have to be financed by raising taxes, borrowing from the public, or reducing benefits or other expenditures. The existence of large trust fund balances, therefore, does not, by itself, make it easier for the
Government to pay benefits.

Paul Krugman tackles these issues in a recent column and a longer article, and does not attempt to pretend that the assets have economic significance.  Instead, he focuses on the legal and political heft of the trust fund.  However, he is toe-dancing down the line between advocacy and deception, and critics will say that he steps out of bounds into falsehood.  Let's look at his column:

Privatizers say the trust fund doesn't count because it's invested in U.S. government bonds, which are "meaningless i.o.u.'s." Readers who want a long-form debunking of this sophistry can read my recent article in the online journal The Economists' Voice (www.bepress.com/ev).

The short version is that the bonds in the Social Security trust fund are obligations of the federal government's general fund, the budget outside Social Security. They have the same status as U.S. bonds owned by Japanese pension funds and the government of China. The general fund is legally obliged to pay the interest and principal on those bonds, and Social Security is legally obliged to pay full benefits as long as there is money in the trust fund.

There are only two things that could endanger Social Security's ability to pay benefits before the trust fund runs out. One would be a fiscal crisis that led the U.S. to default on all its debts. The other would be legislation specifically repudiating the general fund's debts to retirees.

That is, we can't have a Social Security crisis without a general fiscal crisis - unless Congress declares that debts to foreign bondholders must be honored, but that promises to older Americans, who have spent most of their working lives paying extra payroll taxes to build up the trust fund, don't count.

Politically, that seems far-fetched.

Emphasis added.  Now, I agree with his sense of the politics.  However his defenders will need to insist that he is speaking metaphorically when he writes of "legislation specifically repudiating the general fund's debts to retirees", a point that is buttressed by his subsequent reference to "promises to older Americans".

In fact, the government does not need to specifically repudiate the bonds held by the trust fund; all it needs to do is reduce, by new law, the benefits to which retirees are entitled (i.e., the "promises").  If the new benefits are reduced below the level covered by the payroll tax, the Trust Fund will be back in surplus and have no statutory right to demand redemption of its bonds.  Despite the presence of ample trust fund "assets", and even though a general default on treasury debt will not have occurred, retirees will not be paid full benefits.  Hmm, tell me again how meaningful those assets are.

Is this legal ability of the government to re-write the rules controversial or in dispute?  No.  The 1999 Treasury excerpt above is quite clear when it says that "the Federal Government owns the assets and earnings of most Federal trust funds, and it can unilaterally raise or lower future trust fund collections and payments, or change the purpose for which the collections are used, by changing existing law". 

Legally and metaphorically, we owe a "debt" from the general fund to Social Security beneficiaries, which is backed by Trust Fund "assets".  However, a future change in laws can reduce or eliminate the legal debt, without triggering a default on publicly held Treasury debt.  My suspicion is that Paul Krugman understands this point, but hopes that his readers will not.

Now, does this mean that the Trust Fund has no legal significance at all?  No - the reduction in benefits after 2018 would have to occur as a result of an affirmative action, namely new legislation reducing benefits.  Absent new law, the Soc Sec Administration is authorized to redeem its Treasury binds in order to pay benefits, the Treasury is obliged to honor its debt, and the Soc Sec debt is included in the current debt ceiling. 

After (if) the trust fund is exhausted, the ground rules change - the Soc Sec Administration is still legally obliged to make a certain level of benefit payments, but it lacks the authority to spend more than its payroll tax receipts.  At this point, the government really would be defaulting on a promise - without Congressional action, people who made years of high payroll taxes will get an overnight reduction in their benefits.

As to the economic significance of the trust fund, the only case to be made is that its presence has large political and small legal impact, and hence affects people's expectations about future policy.  In this view, cutting benefits in 2018 will be politically difficult, because of the accounting mechanism of the Trust Fund.

So, to summarize - the consensus view, as expressed by the Clinton Treasury and Social Security critics, is that the Trust Fund assets have almost no economic or legal significance outside of their impact on the political process.  Roger Lowenstein did not do a good job of presenting this.

MORE:  Our search for the confused begins with the NY Times editors, who told us this on Jan 10:

In suggesting that 2018 is doomsyear, the president is reinforcing a false impression that the trust fund is a worthless pile of I.O.U.'s - as detractors of Social Security so often claim. The facts are different: since 1983, payroll taxes have exceeded benefits, with the excess tax revenue invested in interest-bearing Treasury securities. (An alternative would be to, say, put the money in a mattress.) That accumulating interest and the securities themselves make up the Social Security trust fund. If the trust fund's Treasury securities are worthless, someone better tell investors throughout the world, who currently hold $4.3 trillion in Treasury debt that carries the exact same government obligation to pay as the trust fund securities. The president is irresponsible to even imply that the United States might not honor its debt obligations.

Well, legislating new, lower benefits will achieve the effect of making those bonds "worthless" without upsetting foreign investors.

Here is a quote in the Jan 2 WaPo from someone who surely knows better:

These 'IOUs' are Treasury bonds, one of the world's safest investments," said Robert Greenstein, executive director of the liberal Center on Budget and Policy Priorities. "The Treasury, the White House and Congress cannot choose not to pay interest on the bonds or not to redeem them -- unless they're willing to have the U.S. government default for the first time in history."

As I said, I am sure he understands the issues; presumably, he believes that framing it as a default on Treasuries may be helpful in advancing his agenda.

Comments

[martin]

So they're going to argue politics and you're going to argue legalities. Good plan. Not.

[TM]

I prefer to think of it as "fact" versus "fiction".

As it dawns on people that defenders of the status quo, such as the NY Times editors, are either (a) lying, or (b) confused, the debate may swing my way.

[Benjamin]

TM >>> I prefer to think of it as "fact" versus "fiction".

Well put.

[Appalled Moderate]

TM:

Social Security gives me a headache -- but I've just taken two asprin, and am prepared to display my ignorance to the world....

What you are truly arguing is that the US cannot continue to promise to pay pensions at the current level without increasing income, or somehow changing investments. D-day for that line of argument is not 2018, because benefits do not go down then.

So, basically, whether the trust fund exists or not isn't an issue to make the heart race. It's when Social Security has to find other sources of income to fund promised benefits. And, frankly, this discussion is peripheral to that day of reckoning. Guess that puts me in martin's camp.

[TM]

So, basically, whether the trust fund exists or not isn't an issue to make the heart race.

Understatement of the week, and its only Tuesday.

It's when Social Security has to find other sources of income to fund promised benefits.

On its own,the payroll tax is insufficient as of (per projections) 2018.

[Appalled Moderate]

TM:

Social Security has a claim on general fund revenues (via the promises to pay it has accumulated) well past 2018. When that claim becomes insufficient to fund benefits -- and government action is required to keep things as they are -- is the reckoning.

Since I think most of the people retiring in the years after 2018 "know" that the SS Trust Fund does not have an existence unto itself, the day that fund starts running a deficit is going to be somewhat meaningless. (With the government, outgo always seems to be outrunning income, so why would a boomer's conscience be shockled when this starts to happen in Social Secuity?)

It's hard to credit Bush with much wisdom when he is working to solve a year 2030 budget problem, and is busily running the 2005-6 budgets into the red.

[Dave Schuler]

And to cover the shortfall, the Treasury, working with Congress, will either (a) run a surplus in other funding; (b) increase outstanding debt, (c) reduce other spending, or (d) raise new taxes.

Good summary. Or, as I've said before, there's going to be a re-adjustment of spending priorities. Frankly, I don't know why this is so shocking. We re-adjust spending priorities to adapt to new circumstances all the time.

I actually think that the anti-Social Security folks have done a pretty good job at making the case for bringing actuarial soundness to the program. Bit of a sitting duck, really. Those arguments have been made and well known for almost three quarters of a century.

The anti-Social Security folks have several problems on their hands. First, the private accounts proposed don't bring actuarial soundness to the program. Consequently if their argument for implementing them is based on the actuarial problems of the Social Security system their argument fails. Second, if the same growth assumptions are applied to the current system as are being posited by the private accounts supporters, there's no crisis in Social Security. So much for a comparative advantages case. Third, the idea that we're going to abandon the current program is politically impossible. We may means test it; we may extend the SSRA somewhat. We won't abandon the plan.

It also bears mentioning that the national savings arguments are largely sophistry. Is it savings if you purchase a U. S. bond? Of course it is. Is it savings if the U. S. government takes the same amount of money from you and purchases a bond? If you answer yes to that question, then it makes no difference to the national savings rate whether private accounts exist or not. If you answer no to the question, as I say, it's sophistry. If you answer “it depends” I would submit that the same “it depends” applies to private accounts.

[Cecil Turner]

"So, basically, whether the trust fund exists or not isn't an issue to make the heart race. It's when Social Security has to find other sources of income to fund promised benefits."

The problem with that logic is that long before that point, a lot of people are going to be paying taxes into the system to provide "promised benefits," knowing they certainly won't receive those benefits. And, of course, they'll be paying higher income taxes as well to cover the gap caused by taking Social Security "trust fund" monies from the rest of the budget. And finally, they'll get to bear the sizeable transition costs to whatever system follows.

I despair of finding useful cost data covering the various options. If tweaking the current system will make it work, then we can probably kick the can down the road a bit. But my sense of the graphs is that we can't . . . and if we have to transition, D-day is well before SS starts to draw monies from the general fund (in 2018?).

[TM]

I agree with Cecil, famed blog-commenter. The crisis arrives when enough folks start to wonder if they will get the benefits they have been promised, asnd also wonder why the folks ahead of them in line (i.e., older) are not sharing the pain.

The idea that we can blithely pay full benefits to folks until 2042, and then reduce payments to 75% of promised benefits, will strike youngsters as daft long before 2042.

Whether that means it is a crisis today depends on how many folks feel affected, how vocal they are, and how seriously they take the projections.

[Cecil Turner]

"I agree with Cecil, famed blog-commenter."

I'm not too sure what that means exactly, except I still need $1.50 to get a [bad] cup of coffee. But it was chuckle-worthy.

"Whether that means it is a crisis today depends on how many folks feel affected, how vocal they are, and how seriously they take the projections."

Concur. Also, how confident they are that the powers-that-be are actually committed to fixing the system.

[SteveL]

Do people seriously think that holding one's own debt is an asset? I mean, get serious. Sure, I could retire on interest if only I sold myself $100,000,000 worth of bonds. Presto, I'm rich!

[TM]

Mr. Turner, you've been away too long. The "famed blog-commenter" bit referred to your auto-Insta-link of the other day.

You should get a blog. Imagine the traffic!

[Jim Glass]

Isn't Krugman the fellow who said we should be more willing to call people who willfully deceive "liar"?

Well, I'll take his advice: He's a lying liar here, lying to deceive -- only *worse*: he's a weaseling lying liar who *tries* to throw in a weasel qualification at the end so he can say "I didn't lie".

Only being a economist, rather than a lawyer who knows how to weaseling disclaimers (like me!), he even screws up at that.

Let's parse a bit of what he said -- even just this little bit leaves us impressed at his mendacity per word...

"The short version is that the bonds in the Social Security trust fund are obligations of the federal government's general fund, the budget outside Social Security. They have the same status as U.S. bonds owned by Japanese pension funds and the government of China."

Bull. Let's put aside the fact that US bonds held by China and Japanese pension plans are *assets* to them on their balance sheets -- and are *not assets* on the US balance sheet -- which will be a mere $5 trillion dollar balance sheet difference in 2018. Forget that. ;-)

Just consider the fact that the bonds held by the trust fund are *legally unique* in that they are *demand bonds*. They can be cashed in at the Treasury at will *before* they mature -- and if they are not they simply roll over.

This is a nontrivial distinction -- you and the Chinese government and Japanese pension plans *cannot buy* bonds like this!

Here's one way this nontrivial distinction matters: if the government chooses not to cash these bonds in by not demanding redemption of them by itself, it DOES NOT DEFAULT ON THEM. They just roll over. (How else would you issue a debt to yourself?)

If the government could default on them it already *would* be -- because they are nominally 15-year bonds (the term used to set the interest rate on them), the government's been issuing them for 22 years, and it hasn't cashed in any of them yet. Default?? No, automatic rollover.

So Krugman says these bonds have "the same status as U.S. bonds owned by Japanese pension funds and the government of China."??

Yes, sure, the exact same status EXCEPT for the minor facts that:

(1) the Chinese and Japanese *can't buy* bonds like these;
(2) The US government is free to choose to not redeem them at maturity *without* defaulting, just as it is doing *now*, whereas bonds owned by the Chinese and Japanese must be redeemed on maturity or default occurs; and, oh yeah
(3) in 2018 the trust fund bonds will represent a net $5 trillion *liability* of the government to SS participants that will have to be financed *with* tax hikes, while the same amount of bonds would be $5 trillion of *assets* to the Chinese or Japanese, which could be used to finance expenses *in lieu of* collecting taxes.

Other than those three ... well, no, there are other differences too but I won't go into them now.

Those are quite enough to make Krugman a "Liar! Liar! Liar!" -- unless he wants to protect his reputation for honesty by claiming ignorance leading to analytical incompetence. Continuing....

"The general fund is legally obliged to pay the interest and principal on those bonds, and Social Security is legally obliged to pay full benefits as long as there is money in the trust fund."

Unless it votes to reduce benefits -- just like it did in 1983! Krugman doesn't remember back that far??

"There are only two things that could endanger Social Security's ability to pay benefits before the trust fund runs out. One would be a fiscal crisis that led the U.S. to default on all its debts."

I love this straw man, it's very popular on the left these days!

"The other would be legislation specifically repudiating the general fund's debts to retirees.
That is, we can't have a Social Security crisis without a general fiscal crisis - unless Congress declares that debts to foreign bondholders must be honored, but that promises to older Americans, who have spent most of their working lives paying extra payroll taxes to build up the trust fund, don't count."

Exactly like Congress did in 1983!!

This is Krugman's weasel -- he's saying here: 'Everything I said above is false because Congress could simply cut benefits, and I know that, but it's true anyhow because Congress would never, ever be so mean as to cut benefits to our old, dependent, retired senior citizens.'

And this is a complete crock. In 1983 benefits ran ahead of the money to pay them and Congress closed the gap 50% by *cutting benefits*.

Of course it didn't cut the benefits of the poor dependent elderly. It cut the benefits of the *young*, which is why they now receive *negative returns*, and it introduced means testing for the wealthy old (in carefully disguised form).

Krugman's forgotten this already?? Somebody remind him now, and ask him:

What will keep Congress from doing the same thing again after 2018 -- cutting future benefits, so it won't have to incur the income tax cost of paying down some or all of those bonds, choosing to roll them over with NO DEFAULT instead -- when the situation will be exactly the same as it was in 1983?

Except for the fact that the tax increase needed to pay off the trust fund bonds will be HUGELY LARGER than in 1983 -- a >20% increase in income taxes for that *on top of* another 40% increase to pay for Medicare.

As it happens, I also recently posted the numbers for these necessary tax increase, at http://www.scrivener.net/2005/01/on-stuff-and-nonsense-about-social.html

So Krugman is a flat liar about the bonds, unless he wants to plead arrogant ignorance. As to his picture of the minimal politcal risk of Congress reducing benefits to boot helpless poor old retirees after 2018, the question is ... is he a political naif who can't even remember 1983, or a willful deceiver?

[Appalled Moderate]

Is the problem with maintaining social security as is that we can't run those kind of deficits in the future? (That seems to be the general thrust of the comments)

If it is, why does Bush feel comfortable running those kind of deficits now? Those deficits are structurally far worse than they appear, because the payroll tax beings in more than is being paid out in social security benefits.

I guess I suspect the administration's motives here, because fiscal prudence has never been a priority with Bush. My guess is that Bush wants to take Social Security away from the Demos as an issue once and for all, and he figures that an account balance approach (what you see is what you get -- invest wisely) will do that. This could be politically brilliant. Whether it's good policy....well, we might know in 2020 -- a year when our hindsight ought to be be particularly acute.

[martin]

Ah yes, TM-but don't forget in the leadup to the Medicare Bill-fiction won!

[Harry Arthur]

If I was a "twenty-something" I'd sign up to dump the tee into Boston harbor again. My children and grandchildren will continue to have their hard-earned money confiscated by the federal government which will provide them with a negative rate of return on their forced "investment."

This whole SS program is an accounting disaster that no private company would ever be allowed to implement. Paint it anyway you want, but to my simple mind issuing IOUs to yourself is not an honest means by which to fund a retirement program. I'm still struggling to get my mind around how the bonds in the SS "trust fund" can be both government assets and liabilities simultaneously. Perhaps I was asleep during that lecture.

I'm betting that the first step will have to be (and should be) that we immediately reindex the cost of living adjustment to inflation rather than wages. I say that as a boomer who will be eligible to retire in 2011. Of course this will be spun politically as a "reduction" in benefits in the same way that any reduction in the rate of increase is always a cut in Washington.

Now where is that tee?

[Harry Arthur]

Perhaps I was thinking of golf instead of dumping "tea" in the harbor. Must have been asleep during the spelling lecture also. :-)

[Bill Arnold]

Re the
"To cover this, the Treasury must, working with Congress, either (a) run a surplus in other funding; (b) increase outstanding debt, (c) reduce other spending, or (d) raise new taxes." times 2
argument, the difference between trust fund and not trust fund is that the trust fund was built with a surplus resulting from a too-high regressive tax. If we weren't building a trust fund (which seemed fraudulent even at the time in the 80s, i'll admit, since it was clear at the time that a debate like this would happen), then we could have paid lower social security taxes on the first NN k of our income and used the money for something else or saved it in private savings, or bought gold coins, the rate being adjusted every year to match as closely as possible expected disbursements. But we didn't. (Else I'd have those gold coins.) We paid those extra taxes, which were used to disquise the magnitude of the general fund deficits and indirectly were used to justify tax rates lower than they needed to be (e.g. cuts on progressive taxes, which pretty much were high-end tax cuts).

So I'm confused, perhaps. Long day. Point out the flaw please? Why doesn't the "class warfare" argument hold, that the trust fund was a loan to the affluent, and we need to raise rates on them (which by now includes me, a class traitor) to force them to pay back the loan? The history of top end tax rates (e.g. http://www.truthandpolitics.org/top-rates.php) might be a way out of this argument, not sure.

Or is the argument simply that it is legitimate to have the argument, that there are many ways of fixing the problem?

[Neo]

This is one of those old good new/bad news stories.

First the good news: there is a trust fund with funds currently invested in U.S. Treasuries, considered the safest investment on the planet.
Then the bad news: there is a trust fund with funds currently invested in U.S. Treasuries, considered the safest investment on the planet.

U.S. Treasuries are considered the safest investment on the planet because they are backed up the the U.S. government and it's taxing authority over the U.S. taxpayer. For this U.S. taxpayer, this is a real case of self-insured trying to feel content that he/she is insured, when no matter what happens, you pay. Sort of, heads I pay, tails I pay.

In 2018, the U.S. Treasury either has to find someone to buy a growing number of the notes being "redeemed" by the "trust fund", or find the money (with full expense to taxpayers) to "redeem" and retire those notes, the crisis will be all too real and growing.
When the 2052 (?) date arrives, the stock of "self-IOUs" will run out, but the ever increasing burden to the taxpayer will not. Taxpayers won't even notice that anything new as occurred, as they pay not matter.

I won't bore anyone with talk of the strategic effects of transferring over a trillion dollars of U.S. Treasuries to investors, mostly European and Chinese. Not even anything about the leverage they will have over U.S. foreign and domestic policies.

[Aaron]

At what point will young workers begin to demand salary paid in cash udner the table to avoid paying the payroll taxes that they know will be coming back to them at 75%?

For Apallled Moderate: means testing, increased taxation, whatever...I think you are missing the point.

This is not a a CLASS issue, but a GENERATIONAL one. To be equitable, means testing should apply to EVERYONE right now, not wait until 2018 or whenever.

And...we are ran deficits lately to fight a couple wars and to stimulate the economy after the bubble bust. This is normal. We can cut some spending in Bush's 2nd term I hope.

[Ravi]

I basically agree with your position - the trust fund has political and moral, not economic significance. But I think the political and moral difference cannot be dismissed so easily. The "rest of the government" borrowed money from Social Security to support lower tax rates and higher spending than would otherwise be possible. When the tables are turned, how is the "rest of the government" anything but a deadbeat if it refuses to pay? And I include in that schemes to reduce benefits to avoid the need to redeem trust fund bonds (though politically, I think passing such a scheme is unlikely - what percentage of the population will be receiving or close to receiving SS benefits in 2018 again?). I agree were such a benefit reduction enated it is a close legal case, but if the benefit projections show that the government *never* has any intention of redeeming those bonds, I think the debt has been de-facto repudiated and that is at least a plausible argument.

And, as many estimates have shown on a number of occasions, the tax increase required to support the redemption of the trust fund bonds isn't that large - repealing the 2001 tax cut (for example) would do it. And, honestly, the thought that we'd return to the tax system that produced such the horrible economic results of the 1990s really keeps me up nights. Actually, I'm being sarcastic: I'm persuaded (by Robert Frank's argument among others) that progressive taxation (ideally progressive consumption taxation, though I wouldn't trust *this* crowd with fundamental tax reform) increases overall economic efficiency by redirecting resources from pointless competition for effectively limited status goods (like penthouse apartments in Manhattan and Patek Phillipe watches, that derive a substantial portion of their value from their exclusivity) towards broad-based benefits (consumption and saving) for the middle class and the poor.

Most importantly, I agree that there are budget problems and something needs to be done now, but I am dumbfounded that people honestly believe Social Security is the highest-priority problem (when it is going to be running a surplus for the next 13 years, even under pessimistic assumptions). The real issue is that the general fund is spending too much and taxing too little (and note that, given past precedents and public statements of insiders like Grover Norquist you should expect 4 more tax cuts over the next 4 years). And the crisis arrives, not in 2018, but whenever foreign (primarily Asian) central banks decide that they have enough dollar reserves. At that point, the dollar will crash, interest rates will spike (because foreign lenders will demand compensation for the currency risk they are bearing), there will be a recession (spikes in long-term interest rates tend to do that) and today's deficit will look like chicken feed. I think its plausible that, for example, long-term interest rates will double, say 4-8% on the 10 year Treasury bond. And this would double the federal governments interest cost at a time when tax revenues are dropping because of the recession. That may not even be enough... Europeans have (in nominal terms) lost something like 40% of the value of their dollar assets over the last 2-3 years. Do you think an extra 6%/year (since I think intereset on Euro-denominated bonds is something like 2-3%) would really bring them back to the dollar?

To me, Social Security reform is deck-chair rearrangement gone amok. Particularly since the additional immediate borrowing required to finance the program is, to me, likely to trigger a currency crisis in the first place. Instead of worrying about the storm that is years away, we need to be focused on the leak in the hull now or only those of us who are heading for the lifeboats (in other words, acquiring overseas assets) are going to survive... But at least there is one benefit of having Republicans in charge: they won't be willing to (punitively) tax overseas assets because the party financers wouldn't like it.

[TM]

But I think the political and moral difference cannot be dismissed so easily.

I don't mean to give the impression that I am dismissing it. However (one might argue), when the Times feels obliged to substantially misrepresent its own side by pretending that a reduction in benefits equals a legal default on Treasuries, one might conclude that it is the Times that feels its moral argument lacks impact.

I'm persuaded (by Robert Frank's argument among others) that progressive taxation... increases overall economic efficiency by redirecting resources from pointless competition for effectively limited status goods (like penthouse apartments in Manhattan and Patek Phillipe watches...

I am not persuaded. First, when I look atEurope, with its progressive taxes and high unemployment, I do not see a model of economic efficiency.

Secondly, the examples you cite for luxury goods are not convincing (And I'll grant that you are just typing out a comment, and may have better examples elsewhere). But really, what limits the world availability of Vatek watches, other than the designer's desire to limit suppply, and his ability to create buzz? I'll tell you this - even more limited are the "Minuteman Autograph Edition" Timex watches, but who cares?

My elaborate (and entirely obvious) theory about garish status symbols in this country is that people pursue them because of (a) physical mobility); (b) social mobility), and (c) a desire to score with the hot chicks.

In Old Europe, life is a bit like "Cheers" - everybody knows your name, and your family, and your background, and so on. Consequently, a person can still get into the fancy nightclubs without wearing a gold chain or a driving a cool sportscar to impress the doorman and bouncer.

Here in the land of opportunity, anyone can be anything. Consequently, in order to show people you really are a heavy hitter, you need to wear it, drive it, live in it, and vacation at it (only the last cannot be directly observed, but folks can admire your tan).

Until folks surrender their desire for status, we will have status symbols. And folks like Vatek will meet the demand. (I still have my Autograph Edition available, too...)

[TM]

OK, from Bill:

the trust fund was a loan to the affluent, and we need to raise rates on them (which by now includes me, a class traitor) to force them to pay back the loan?

I would probably frame the debate that way if I were a Dem. And as a bonus, it should be easy to find plenty of folks who were thirty (or yopunger) in 1983, paid the higher payroll taxes for years, and are planning to retire around 2018.

But the other side of the story doesn't match - "the affluent" is not an eternal group (as you have found out, membership changes). So Dems will argue that we need to raise taxes on high earners in 2005 in order to recoup taxes not paid by high earners in 1985. Will it eventually dawn on them (or voters) that they are talking about different people?

And as bonus, the Bush tax cut was mentioned in one of these CBO studies (IIRC). The component for "the rich" was about 0.7% of GDP; the bit for the restof us was about 1.4%of GDP.

And as you recall, Howard Dean favored repealing the whole thing; Krugman and others advised against that, as being political suicide; and Kerry finally argued for undoing only the cut for the rich.

So politically, 2/3 of the first Bush tax cut is off the table. (And I can't find a link, so I can't say whether the CBO, or whoever, was looking at both tax cuts, or just the first.)

Pretty well researched, huh?

[Jim Glass]

"And as bonus, the Bush tax cut was mentioned in one of these CBO studies (IIRC). The component for "the rich" was about 0.7% of GDP; the bit for the restof us was about 1.4%of GDP."

IRS Statistics of Income data is in for the Bush income tax cuts as shown on 2002 tax returns.

Result: The lower your income the *more* your tax rate was cut, while tax rates actually rose for the very rich. And the rich suffered a substantial drop in income, so they wound up paying a higher tax rate on reduced income.

This entirely *progressive* result -- as unexpected as one might think it is after all the talk of "Bush tax cuts for the rich" -- apparently is not newsworthy enough to be reported by the NY Times, as it hasn't been mentioned in it yet by Krugman, David Cay Johnston, Uchitelle, or anyone else as far as I can see.

In fact, the only person I know who's published the actual numbers is, well, me....

http://www.scrivener.net/2005/01/2002-year-rich-got-poorer-but-paid.html

[Buckaroo]

Jim, the only reason tax rates for "the rich" went up in 2002 was that capital gains income decreased. Capital gains is taxed preferentially. Thus, the effective tax rate for all income for "the rich" increased, even though the marginal rate for ordinary income was lower.

As you know, the tax cut for dividend and capital gains income in 2003 is a cut that disproportionately benefits the rich. So, lets look again at those tables for 2004.

Incidentally, income tax revenues for 2004 remained about 15% below the levels of 2000, and that's before adjusting for inflation, much less for GDP growth (if you adjust for inflation, you have to go all the way back to 1996 to match revenue). So, we've had a series of tax cuts that have now demonstrably resulted in a decline of govt revenue, even with record corporate profits that were the result mainly of accomodative federal reserve policy.

Maybe it's time to reconsider whether tax cuts can grow us out of fiscal troubles, and get back to good old fashinoned fiscal conservatism, where we set the tax rate more or less to match spending and projected future obligations.

We could avoid most of the problem of repaying the SS trust fund simply by avoiding running up the publicly held debt for the next 15 years. We'd also be much less at the whims of foreign central banks, who will be rolling over most of the publicly held debt and for which we'll be forced to find new buyers about five times between now and 2018.

The current annual deficit of $550 or so $Billion is a much bigger problem than planning now to begin paying the interest on the bonds held by the Social Security trust.

Of course, it wouldn't suit conservative interests well if we are actually in a position to make good on the bonds held by the trust fund. So, we probably won't, at least not for the next four years. But it's good to know that conservatives are actively planning how to default on short to medium term Social Security benefits.

[Ravi]

TM, you're missing my point - I'm not saying that people should not pursue status symbols, I'm just saying that to the extent an increasing share of consumption is directed towards status symbols that decreases overall economic efficiency. Suppose you want to buy an exclusive watch: Does it really make a difference if you can get level of exclusivity you want with a $5000 watch that has 60 diamonds versus a $10000 watch that has 286 (the prices are made up but the number of diamonds is real)?

I think it doesn't make a difference. And I think the increasing share of consumption directed towards luxury and status goods means that we have notably underestimated the diminishing marginal utility of income. Another way of saying that is that I think the country would be better off (and the status-good purchaser no worse off) if there were incentives that redirected that $5000 from the status goods arms race towards more productive uses. (whether that is private savings, funding the government or whatever).

I should also note that decreasing the consumption of status goods without making status good purchasers notably worse off is something that cannot be achieved privately. You could decide to buy less status goods, but then you would also end up with less (relative) status. But if status goods (through taxes or savings incentives or whatever) just cost more for everyone, people could keep more or less the same relative status even when the share of consumption devoted to status goods falls.

[Ravi]

As for the political and moral argument about the trust fund, it depends on what sort of benefit reduction you are talking about. If you are talking about a benefit reduction to close the post-2042 actuarial gap while still honoring the trust fund bonds, I agree that is not the moral equivalent of a default. But if you are talking about a benefit reduction specifically designed to avoid redeeming the trust fund bonds (to avoid the impact of that redemption on the rest of the government), I do think that is the moral (and as argued above possibly legal) equivalent of a default. Particularly after the early-80s reform, people paid higher taxes that were supposed to be reserved for Social Security. If we decide that reservation is no longer a good idea, I think the least that we owe the public is an honest, straightforward debate on that specific issue.

If you want to repeal the portion of the law that says that taxes collected for Social Security must be spent on Social Security, be my guest. I think that is bad policy (and worse politics), but at least it would be honest. Hiding the default in a footnote of a benefit formula, on the other hand, is deceptive and dishonest.

[Buckaroo]

One way the creditor nation central banks can stop accumulating dollar assets is simply to spend them on raw goods.

That lets them keep exporting to the US, while passing off the currency risk, and forcing inflation (or decline of the dollar) on terms favorable to the creditor nations.

[Buckaroo]

It's when Social Security has to find other sources of income to fund promised benefits.

On its own,the payroll tax is insufficient as of (per projections) 2018.

On its own, if we go to "Plan 2", which is by most assumptions the model Bush intends to propose, the payroll tax is insufficient to pay current benefits immediately, and would remain worse than simply doing nothing for about fifty years.

We'd have to run much, much more red ink (I'd say we'd have to make a massive tax increase if I was using the same language some conservatives use to describe the problem of paying back the trust fund treasury bonds) to allow diverting 2% of payroll rather than simply letting the trust fund run out of assets sometime between about 2042 and 2052.

Bush administration projected in 2004 that the incremental debt to privatize (at 2%) would add 23% of GDP to the federal debt by 2036, and each year between 2004 and 2050 would have a bigger general budget deficit than leaving SS untouched.

What we've got is an albatross of a plan masquerading as a solution to a general budget crisis, but which would in fact make things a whole lot worse for practically everyone who's in the working population today. It might make things better 75 years from now, if everything goes to plan.

I think a plan that takes 75 years to make things better qualifies for the old saying, 'In the long run, we're all dead'. We would never get to the long run, because we'd go through an Argentina in the mean time.

[TM]

Don't cry for me, Argentina...

Thatis more of Krugman's deliberate deception. Much of Argentina's debt was denominted in currencies other than its own (mainly dollars). Consequently, they are limited in their repayment options.

Although I do not recommend it as a method of sound public finance, the US can simply invite foreign central banks to a Printing Press Party, and ask them whether they want their bonds repaid in tens, twenties, or hundereds. We will naturally stamp each bit of currency with the magic words "Legal Tender for all debts public and private", and tell them to beat it.

Krugman understands this, of course, but rarely takes the trouble to mention it to his readers. However, we have Brad DeLong on our side for this (sort of - he is speaking of the trade deficit, and fretting about Social Security/Medicare reform)

[Buckaroo]

OK. So we'd be another England.

Is it worth it to you to push the USA off the top of the hill just so we can phase out Social Security?

[Ravi]

Yes, running the printing presses is (currently) an option. But given the unsustainability of our budget and current account deficits (not to mention the pace of reserve accumulation by foreign central banks), I would not be so certain about our future ability to issue dollar-denominated debt. Given the lack of (local) political will to do something about current imbalances, I expect that when the currency crisis hits that, at a minimum, private lenders will start demanding inflation-hedged bonds (i.e. TIPS) and foreign ones might well demand debt denominated in their currency. And after that, then what?

I will also note that, even today, net interest costs are (1.4% of GDP) are far less than the structural general fund deficit (5% of GDP). So even if you inflate away all of today's debt you will still need to borrow, and the interest costs on that would (I imagine) be punitive. After all, at that point, the US government will already have been proven to be a deadbeat once... And none of this really does anything about Social Security (though I suppose inflating away the trust fund balances is a sneakier way to repudiate them...).

[Ravi]

I should add that (from Social Security's perspective) there's any easy way to solve the problem of hyperinflation. Just pass a law requiring the bonds in the SS Trust Fund to be treated like TIPS (instead of like ordinary bonds). I'd love to see a debate on that (since opposition to the measure would be the same thing as saying you're considering a hyperinflation to get around our current fiscal problems). Of course, such a blatant contemplation of an inflation would probably provoke an external financing crisis, but (absent more budget discipline than we've seen in the past 4 years), I think that's inevitable anyway...

[doug]

Let's see how can we use general fund money to pay for social security commitments. How about some of the billions we are using for the "missle defense system" or for Trident submarines or the F22 fighter or....... All these systems which won't work or be necessary against our potential adversaries. See there is plenty of money you just have to identify your priorities. How about an increase in taxes, hey that would work too! As has been noted all over the blogsphere the amount of money we are talking about is relatively small with respect to overall funding through those years.

[Jim Glass]

Jim, the only reason tax rates for "the rich" went up in 2002 was that capital gains income decreased...

~~~
What, you mean the rich got poorer too (as I said)?

Why yes they did, with income down 70% at the top level due to capital losses replacing capital gains.

Now the strange thing is how the Times' Krugman/ Johnston/ Uchitelle triad, after writing alarming story-after-story about the "Top 1%" and the "New Plutocracy" based on IRS year 2000 boom-year capital gain data, has not found this evisceration of the new plutocrats a story worth mentioning even once over the last two years.

Well, except for one very amusing spin attempt by Johnston who wrote a story Income Drops For All Americans accompanied by bar charts which showed income up for all Americans except the Top 1% who had theirs down 60%.

Being that the term "Top 1%" had apparently dropped out of Johnston's vocabulary since the prior year, he had no way do describe the resulting decline in aggregate income except as a decline in income for "all Americans". I guess he didn't know the art department was going to run those bar charts with his story. ;-)

The capper was that on the web version, down where it lists related stories, were his stories from the prior year: "Top 1% See Income Surge", "Top 1% Widen Income Gap Above Other Americans"...

Then "Income Falls for All Americans". It was so funny I almost spit my beer.

But enough of me. Now, you were explaining why, down through all the wage earning levels, the Bush tax cuts were *progressive* in reducing tax rates the most at the lower levels all the way through?

[Buckaroo]

Capital gains income is volatile. Positive some years. Negative others.

Much of the change in distribution of aggregate tax rates for the top few percent would have occured with no tax law change at all, because the rich have much more capital gains income than others, and capital gains are taxed at a lower rate than regular income, and lots of people registered capital losses in 2002.

Aside from that, the 2001 tax bill phased in cuts for the top bracket, including the bizarre approach on the Estate Tax in which exemptions are increased every year for nine years, then the Estate Tax vanishes for one year (the "Push Grandma Off the Train" year), and then comes back in full force in 2011 with exemptions restored to the level as of 2000.

So, whatever distributional affects attributable to the 2001 tax bill show up in 2002 tax returns do not reflect the entirety of the bill. Just the tip of the iceberg.

And, we'll see another big part of the iceberg when we get to see 2004 results, where there were some capital gains to report, and where capital gains tax rates were cut by a quarter and dividend tax rates by more than half.

[alajac]

Inflation being a de facto tax on wealth, they don't HAVE to do anything but print the money as needed. Which is what Bushco is trying to head off.

[J]

I would note that the SS Trust Fund is only able to buy Government bonds because bonds are being floated in the first place by a President and Congress who are incapable of aligning income and expenses.

If the government had continued its policy in 2000 of running a surplus to pay down prior debt, the SS Trust Fund would have no bonds to buy, and would have been forced into diversification of either other bonds of other currencies (a good hedge against devaluation) or equities (as most state Pension plans do.)