Martin Feldstein: Social Security Compromise

Do we really have to have this debate all over again? There has been a noticeable push to revive the Social Security issue in recent weeks, but I doubt it will have any political traction:

Social Security Compromise, by Martin Feldstein, Commentary, WSJ: A recent proposal by Rep. Rahm Emanuel, chairman of the House Democratic Caucus, may point the way to a bipartisan compromise on Social Security. The essence of Mr. Emmanuel's proposal ... is to create universal personal retirement accounts funded by equal 1%-of-earnings contributions from employers and employees. Although enrollment would be automatic, participation would be voluntary since each individual could choose to opt out in any year. ... A subsidy for low-income individuals, comparable to the tax benefit for savers in IRAs and 401(k) plans, would guarantee their participation. ...

As a candidate in 2000, George Bush called for personal retirement accounts to supplement the traditional tax-financed Social Security benefits. ... Unfortunately, Democratic critics argued that individual accounts would be "gambling" with the retirement savings of working men and women. Moreover, congressional Democrats refused to begin a dialogue with the administration, even when the White House insisted that every aspect of its plan was negotiable.

The partisan character of the attack was ironic since President Bill Clinton had ... spoken favorably about the potential role of individual accounts. The Clinton White House and Treasury developed a detailed plan based on such accounts. ...

The Rahm Emanuel proposal, coming from the chairman of the House Democratic Caucus, looks like an important signal that a compromise may now be possible between Republicans and Democrats along the general lines originally proposed by President Bush. There would of course be many details to be negotiated. But the basic idea of combining traditional tax-financed Social Security with investment-based personal retirement accounts looks like it can be on the table.

As both President Bush and Mr. Emanuel emphasized, personal retirement accounts would increase individuals' confidence in their retirement incomes and would raise our national savings rate, contributing to faster economic growth and less dependence on capital inflows from abroad. The creation of personal retirement accounts can also be a first step toward correcting the fiscal problem of Social Security.

As everyone now recognizes, the current 12.4% Social Security employer-employee payroll tax will not be enough to finance the benefits specified in current law as the population ages. Continuing to finance those benefits with a pure tax-financed system would require raising the payroll tax rate to more than 18%, or finding other ways to raise tax revenue.

The historic returns on stocks and bonds, however, imply that a 2% contribution to personal retirement accounts would eventually generate enough retirement income to maintain the monthly benefits specified in current law, without any increase in taxes. Such is the power of investments in stocks and bonds that 2% of payroll contributions to personal retirement accounts would eventually do the work of a 6% tax-rate increase.

Reforming the tax-financed Social Security benefits to avoid a future fiscal shortfall is a bridge that need not be crossed now. The important point is to start the personal retirement accounts as soon as possible. The rapid aging of the population and the imminent start of the baby-boom generation's retirement make it important to avoid delay. The White House and congressional Republicans should reach out to Mr. Emanuel to see if the compromise needed to save Social Security and increase savings can now be achieved.

As Dean Baker is always ready to point out, the solvency argument is not very compelling:

The Congressional Budget Office's projections show that the program can pay all benefits, with no changes whatsoever, through the year 2046... The projected shortfall over the whole 75-year planning period is 0.4 percent of GDP, approximately 30 percent of the current cost of the war in Iraq...

But even if you do believe there are problems on the horizon, when you remember that support for reform from many on the right is not because they want to save the current Social Security system, but rather because they see this is a first step toward eliminating it, the reluctance of Democrats to enter into negotiations is understandable. Democrats distrust Republicans on this issue for good reason.

Comments

[Gabriel M.]

I'd rather be with Jose Pinera on this one, rather than with... Bismark?

[Meh]

The reason "personal accounts" are forever being pushed is because it's a handy earner for various Wall Street institutions to get hold of the commissions and transaction charges for these accounts.

So it won't go away, ever.

[ilsm]

The issue is not so much SS solvency, as liquidity.

In ten years SS will be a net drain to liquidity. Some one is going to have to have money to lend to the federal government.

May as well set up a set of funds for SIV's to sell to.

The SS fund will be selling t bill in competition with the derivative markets............

[bakho]

"Continuing to finance those benefits with a pure tax-financed system would require raising the payroll tax rate to more than 18%, or finding other ways to raise tax revenue."

The 18% payroll tax is the OMG scare. The payroll tax is currently split equally between employer and employee.

"Finding other ways to raise tax revenue" of course is the solution and as Dean Baker points out a relatively small cost compared to total budget.

The elephant in the room continues to be spending on health care. Uncontrolled rising health care costs and insurance premiums are a far greater "tax" on employers that provide insurance benefits and employees that "co-pay" than SS.

[pgl]

Here we go again. Private accounts are magic fairy dust with all sorts of free lunches. Over at EconoSpeak, I rebut this argument. Feldstein should be ashamed of himself.

[johnchx]

Stock answer to anyone who want to "solve" a financial crisis which will materialize in 20+ years:

"That's very interesting. But before we start talking about how we're going to pay our expenses in 2020, why don't you give me your plan to pay for last year? From that, we can gauge your seriousness and your understanding of the basic facts of federal finance."

There's really no point in talking about any "giant fiscal hole in the sky" (real or imagined) with somebody who can't articulate an arithmetically sound solution to today's comparatively modest shortfall.

[howard]

i was going to ignore this altogether - feldstein having a track record of dishonesty on this matter - but then i decided to skim his remarks, and waddya know?

the track record of dishonesty turns out to be a perfect predictor: "As everyone now recognizes, the current 12.4% Social Security employer-employee payroll tax will not be enough to finance the benefits specified in current law as the population ages" is the kind of remark that disqualifies you from discussing social security.

it does, however, qualify you to write for the wsj editorial page, since dishonesty is job one over there.

[robertdfeinman]

A dumb question:
Given that the trend is away from publicly held companies and towards absorption by private firms where are all those new dollars being saved for retirement going to be invested?

I claim that much of the rise in the stock market over the past 30 years was as a result of the fresh money flowing into it from 401K plans. Not only was there fresh money from the monthly contributions, but the increasing adoption of such plans provided additional funds. These funds needed to be invested someplace and the bidding for a limited pool of suitable stocks caused them to rise.

The trend has now tapered off. The monthly contributions may continue but the rush to dump defined benefit plans is winding down. I suppose a libertarian argument will be "given enough buyers a market of sellers will emerge". Unfortunately the quality of recent offerings doesn't look promising. There are no new General Electrics being offered. The closest might be Google and that is primed to be a bubble.

What one sees instead are such sterling firms as Enron being created just to tap into the money looking for a home. This is not promising.

[pgl]

Howard - one of my EconoSpeak colleagues followed up my post tracking the history of Feldstein goofs on this topic. Alas when I go to our website, it truncates both of our posts. Hope you have better luck as our blog master needs to fix this glitch.

[calmo]

Excellent dumb question rdf...seriously, "This is not promising." is the expression used to contain the apprehension --that pre-verbalized thought, that is delivered by the nose...that the howling baby's diaper needs to be changed.
So were you a great diaper-changer in your day?
I was Ace.

[The Baron]

johnychx take a look at ilsm's comment. SS is the solution to how we are paying for our shortfalls today. In 20 years, actually far less, it not only stops being a major part of the deficit 'solution', but turns to becoming a major part of the deficit 'problem'. This is why, all due respects to our host, I disagree with Mark's take, that we don't need to worry about SS running out of money. Yes, SS is fully funded, but the majority of that funding is in T-bills, whose purchase has been financing our deficit spending. Shortly, those T-bills not only won't be being purchased by the magic money machine, but will start to be redeemed, and not renewed, at an ever increasing rate. This change in economic flow must be compensated for elsewhere in the economy. As Mark says, SS will still get paid out... the consequences of such are what get left out all too often.
As bahko states, and Mark has pushed in the past, the looming elephant is the rising Medicaid/Medicare bill, it is a current part of the problem rapidly growing, however, the obvious 'solution' (in D.C. speak), further deficit spending, would also have to rely on a large, government mandated source of T-bill buyers. The largest of those, SS, is about to leave the market.

[Denis Drew]

Per capita GDP grows 4 times as fast as population -- why Social Security was never a fundamental problem. Decades from now the payroll tax may go up 50% but incomes should have gone up 100% (SS is projected by the CBO to grow from 4% to 6-7% of GDP by around 2050 and stay there as far as the eye can see).

The cost of medical care grows 5 times as fast as our ability to pay for it (20 times as fast as population growth IOW). Ergo, 80 years from now medical expenses may grow to 300% of today's share GDP (20 X 15%), but we should have 400% of today's per capita GDP to pay for it (per capita GDP usually doubles every 40 years -- not even counting technological deflation, which it shouldn't).

Perhaps our biggest long run economic goal should be to get medical spending down to 200% of today's per capita GDP. :-)

In the meantime re-raising the non-essential SS issue is just a giant distraction from today's truly gigantic economic needs -- foremost being re-unionizing the USA according to first-world standards, i.e., institutionalizing sector-wide collective bargaining; second being universal medical care, etc.

[Bruce Webb]

I know I am largely preaching to the choir here but at least lets get some correct figures on the table.
Moreover, congressional Democrats refused to begin a dialogue with the administration, even when the White House insisted that every aspect of its plan was negotiable.
I cry bullshit on this one. Bush made it perfectly clear that he would not accept a straight out fix by payroll tax. Not everything goes down the memory hole.
Continuing to finance those benefits with a pure tax-financed system would require raising the payroll tax rate to more than 18%, or finding other ways to raise tax revenue.
No it wouldn't. The current payroll tax gap is 1.95%. A tax fix implemented tomorrow would have an increase to 14.35%. As those madcaps called the Trustees of Social Security put it;
For example, payroll taxes could be raised to finance scheduled benefits fully in every year starting in 2041. In this case, the payroll tax would be increased to 16.41 percent at the point of trust fund exhaustion in 2041 and continue rising to 17.60 percent in 2081. Feldstein managed to pre-date (and misstate) that '18%' by 74 years.
And this is simple stupidity
The historic returns on stocks and bonds, however, imply that a 2% contribution to personal retirement accounts would eventually generate enough retirement income to maintain the monthly benefits specified in current law, without any increase in taxes.
They can call it what they want. When the government requires an additional 2% of payroll from you and your employer that is a 'tax'. And did I mention that the currrent payroll gap was 1.95%? It doesn't have a thing to do directly with "historic returns" under Intermediate Cost assumptions a increase in payroll of 2% will "maintain the monthly benefits specified in current law". But all private accounts would do is to add administrative costs.

I am all for capturing additional yield, I have suggested diversifying the Trust Fund to other asset classes for a long time Invest or Divert?: Lets do Both But people are trying to give private accounts credit where it doesn't exist. Yes stocks can yield more than bonds, which is why CALPERS has a diversified portfolio.

You expect this kind of dishonest drivel from the WSJ Opinion page. But surely Rahm E has someone on staff who can actually read and understand a table. For that matter you don't have to master every detail of Tables V.A1, V.B1 and V.B2 (though I suggest people do), just getting through the 15 page summary would be enough fact checking to thoroughly debunk this piece of economic junk writing.
2007 OASDI Trustees Report: Table of Contents
http://www.ssa.gov/OACT/TR/TR07/trTOC.html

[Tapen Sinha]

If I may, I would like to plug the new book I have edited with Steve Kay of the Fed-Atlanta to be published by Oxford shortly. It discusses what we can learn from many countries in the region.

http://www.oup.com/uk/catalogue/?ci=9780199226801

[Bruce Webb]

The Baron
SS is the solution to how we are paying for our shortfalls today. In 20 years, actually far less, it not only stops being a major part of the deficit 'solution', but turns to becoming a major part of the deficit 'problem'.
Well except that it doesn't. You can grant them every assumption they claim and the dollar figures just are not that big. In Constant 2007 dollars the gap under Intermediate Cost assumptions starts at $30 billion in 2017 and rises to about $365 billion in 2040 and then actually goes to zero in 2042 as benefits get readjusted to current income (and still representing a 20% better real result than current retireees receive).
http://www.ssa.gov/OACT/TR/TR07/VI_OASDHI_dollars.html#wp140103

Moreover every bit of GDP growth above 2% reduces that. The reality is that for about a six year period we might be temporarily returned to Reagan era deficits, (but probably not). We could finance that with no particular problem, if that is we could get a handle on health care and defense budgets. Social Security never becomes a "major part of the deficit", in fact under trend growth (which Low Cost is a good approximate) the gap between Income excluding Interest and Cost remains about $125 billion per year in constant dollars througout the latter part of the 75 year window. By 2081 Low Cost would have us tapping the General fund for $122 billion of the $1 trillion in interest earned on a $19 trillion dollar portfolio. (In current dollars $440 billion on $4.1 trillion in interest on a $75 trillion dollar portfolio). In effect we are borrowing money at about a nominal 5% rate but with a real rate about 12% of that or .6%.

It is worth repeating: 2.8% Real GDP and Social Security 'Crisis' vanishes before your eyes.

[johnchx]

The Baron writes: johnychx take a look at ilsm's comment. SS is the solution to how we are paying for our shortfalls today. In 20 years, actually far less, it not only stops being a major part of the deficit 'solution', but turns to becoming a major part of the deficit 'problem'.

Indeed. That's why I prefer to quote my "fiscal gap" figure (in round numbers $700 - $800 billion in 2006) rather than the more widely reported unified or on-budget deficits.

Closing this gap would have the following effects:

(1) Debt held by the public would decline by an amount equal to (using the $700 billion gap) or greater than (using the $800 billion gap) the increase in debt held in government accounts.

(2) Unfunded liabilities for civilian and military retirement benefits, post-retirement medical care, and veterans benefits would not increase. (We wouldn't make any progress towards reducing the $3.5 trillion unfunded liability, but it wouldn't grow.)

Together, this would mean that the federal government's overall financial position would improve by an amount equal to the surpluses run by Social Security (other government operations being, as it were, fully funded).

[Bruce Webb]

Actually someone who really knows numbers (as opposed to a number pointer like me) might want to calculate the real interest rate under Intermediate and Low Cost assumptions. Under Intermediate Cost we have been borrowing money every year since the reform in 1983 and are scheduled to do so until 2017, whereupon we engage in a 34 year payoff. Now you can amortize that. But under Low Cost assumptions we never end up repaying principal and only a portion of the interest. Here is the kicker. Under Low Cost assumptions Income excluding Interest falls behind Cost in 2023. At which point the actual utility of those 40 years of surplus begins to fade away. Yet the Interest bill never does. Its exactly like paying a minimum on a credit card, after awhile that "must have" big screen starts costing you more and more, even after it breaks down.

Well a tension ends up building between generations. A guy born in 2001 and entering the workforce in 2023 never would have paid any excess tax in, yet when he takes retirement in 2068 some 12% of his check is being paid on interest on a bond portfolio he never contributed to. That is under relatively modest Low Cost assumptions. Plug 3% Real GDP into the model and all hell breaks loose. Because how much is too much?

Under Low Cost the Trust Fund Ratio (reserves as a function of time) peaks right under 500 in 2023 and sinks to 400 in 2040 and more or less retains that level through the 75 year window (Figure II.D7 http://www.ssa.gov/OACT/TR/TR07/II_project.html#wp106217). But what happens if we get a return to 90's level growth? Well the Trust Fund Ratio goes through the roof. At some point workers will ask why they are continuing to pay 12.4% into the system plus interest when there is six, seven, ten years in reserve? With most of that consisting of interest on interest on excess contributions paid not by currrent retirees but instead by now dead Boomers?

A sustained period of 3.0% growth sets up a perfect economic storm around med-century. Because oddly enough an overfunded Trust Fund presents more risk than an underfunded one. Because you can always cut benefits to meet current income. But what do you do if you have nominal trillions in the bank at multiples of current cost? There is no easy way out of that box.

In 2006 $102 billion of the $189 billion surplus was interest earned, it in turn immediately started earning interest. If you don't catch it in time the threat of interest on interest literally compounds. That is the more likely challenge in the next decade or so, and pretty much no one sees it coming.

[Peter Schaeffer]

The real argument for personal savings accounts isn’t the future solvency or insolvency of Social Security. The real issue is the decline of corporate pension plans. Historically, lower income workers typically relied on Social Security which replaced a reasonable percentage of their pre-retirement earnings. Because of the progressive benefit formula, Social Security has never provided comparable benefits (percentage of pre-retirement income) to higher income workers.

These workers, however, tended to benefit from defined benefit (DB) pension plans in both the private and public sectors. In the private sector, DB pension plans are in rapid decline.

As a consequence, a new mechanism is needed to provide retirement income for middle and higher income workers. Keogh’s and 401Ks are a step in the right direction, but not nearly pervasive enough. Some type of PSA system is needed to address these issues. In my opinion, it will have to be mandatory and the contribution rates are likely to be much higher than 2%.

Note that such a system is not needed as a replacement for Social Security but in addition to it.

One statistical point. According to the Social Security Trustees Report

“The projected 75-year actuarial deficit in the combined Old-Age and Survivors and Disability Insurance (OASDI) Trust Fund is 1.95 percent of taxable payroll, down from 2.02 percent in last year's report.”

That looks like more than 0.4% of GDP to me. However, I don’t have enough of the methodology at hand to be sure.

[robertdfeinman]

Once again the fundamental lie about Social Security - it's not a trust fund, it has no reserves.

Social Security is a pay-as-you-go system. The adjustment in rates promoted by Greenspan was a fraud. It has been taking in more than it pays out for decades. The extra revenue has gone to fund normal government operations.

It receives a special kind of IOU for this from the Treasury. There is no other place for the excess revenue to go, the government can't "invest" in itself.

As some point in the future it will start to pay out more than it takes in. It will then cash in it's IOU's. How will this be done? The treasury will need to raise additional funds. This can be done by increasing taxes or increasing the deficit or cutting back on benefits.

So the only open question is when will the annual payout requirements start to exceed tax collections? Apparently that date seems to be 40-50 years from now depending on various assumptions. So why are we worried about funding the shortfall now? Two reasons:
1. To discredit the idea of Social Security by making people think it is on the verge of going bust.
2. To raise more money for current purposes by, once again, tweaking the collection and payout formulas.

I understand why pols don't want to tell the truth, Gore's Lock Box was just as much baloney as the what the GOP was spewing, but why can't economists do a better job of explaining the situation.

In fact the picture can be extended to all aspects of government. Everything is pay as you go, even if it means printing money.

[ilsm]

I have posted this link at angrybear: http://www.treasurydirect.gov/govt/reports/pd/feddebt/feddebt_aug07.pdf

On Nov 5 the Sep 30 end of fiscal year 2007 debt total will be published, these will be "audited" so it should tell us how much debt instruments in "privately held" and "intragovernmental" accounts made up the national debt.

The site will tell how much interest was paid out in cash to private holders, and how much was credited to the intragovernmental accounts.

In a few months GAO will do their thing and submit an audit report on the national debt.

Anyone who suggests that OASDI payroll taxes need to be raised to continue to maintain the credit balance in the intragovernmental debt is suspect.

That seems to be the tack, raise payroll taxes to keep from liquidating 2 plus trillion in intragovernmental funds for social security.

The second largest holder of intragovernmental debt is: civil service retirement and disability fund (OPM), followed by medicare (HHS), Federal Disability Insurance (SSA) and fifth, military retirement fund.

Should we not worry about paying out the second half of the intragovernmental funds all of which will be in the same situation as SSTF?

There is a debt, deficit liquidity issue and it is not just SS.

[Peter Schaeffer]

robertdfeinman,

“There is no other place for the excess revenue to go, the government can't "invest" in itself.”

Other countries invest in private sector assets at home and (usually) abroad. Singapore would be one of many examples.

“So the only open question is when will the annual payout requirements start to exceed tax collections?”

According to the SSTR, the first year outgo exceeds income excluding interest is 2017 (for OASDI). The First year outgo exceeds income including interest is 2027 (for OASDI).

[johnchx]

robertfeinman writes:Once again the fundamental lie about Social Security - it's not a trust fund, it has no reserves.

Actually, I think there is a good and important reason to treat the "trust fund" as a real entity. The "trust fund" records the following crucial fact: that payers of (progressive, all-source) income taxes owe $2 trillion to payers of (regressive, wage-only) payroll taxes.

Which is another way of saying that high-income earners and recipients of capital-derived income owe a great deal of money to ordinary workers.

To simply view the "trust fund" as a debt from "one part of the government" to "another part of the government" is to overlook big differences between who pays for those two different "parts."

I think the debt ought to be (a) taken seriously and (b) paid.

And that leads back to my rather tedious refrain: $700 - $800 billion in additional annual income tax collection (or spending cuts, or any combination you like) is the minimum necessary to bring us to a point of actually paying for the government we have.

[johnchx]

ilsm writes:Should we not worry about paying out the second half of the intragovernmental funds all of which will be in the same situation as SSTF?

Yep. And we should be mindful of the fact that most of those assets, about $1 trillion, are held to pay pensions, veterans benefits, etc. with a net present value of about $4.7 trillion.

[anne]

"As Dean Baker is always ready to point out, the solvency argument is not very compelling:

"'The Congressional Budget Office's projections show that the program can pay all benefits, with no changes whatsoever, through the year 2046... The projected shortfall over the whole 75-year planning period is 0.4 percent of GDP, approximately 30 percent of the current cost of the war in Iraq...'"

There is simply no problem with Social Security.

[anne]

"Once again the fundamental lie about Social Security - it's not a trust fund, it has no reserves."

"Once again the fundamental lie about Treasury bonds - they're not bonds."

"Once again the fundamental lie about Treasury bonds - they're not payable."

"Once again the fundamental lie about Treasury bonds - they're not securely payable."

"Once again the fundamental lie about Treasury bonds - they're not securely payable indefinitely."

[anne]

"There is a debt, deficit liquidity issue and it is not just SS."

As long as the economy grows remotely close to the growth in debt there is no problem, and even beyond there is minimal problem now and little problem that can not be solved by reasonable tax increases or (shudder) military spending cuts.

[johnchx]

anne writes: There is simply no problem with Social Security.

I agree, with the added caveat that this means that there is a huge problem with financing the rest of the government. (Approximately eight times the annual cost of the Iraq war; more than three times the annual cost of the Bush tax cuts.)

[howard]

i like to note, every so often in this kind of discussion, that the payroll tax, like the highway trust fund tax, is a special purpose tax.

When a case came to the supreme court about the fungibility of highway trust fund moneys, the decision was that the money wasn't fungible: it had to ultimately be used for the stated purpose.

Which is to say that johnchx has it just right: social security financing is fine. the rest of the government's financing: nicht so gut....

[Bruce Webb]

Just as important as the timing of the gap between Income excluding Interest and Cost is the amount. The gap in 2017 is by budget standards minimal, around $30 billion dollars. This amount builds over time but never gets you back to Reagan level deficits in real terms.

Which is why I think it is wrong to formulate this as rdf does, not because it is wrong, but because it comes without appropriate context:
"This can be done by increasing taxes or increasing the deficit or cutting back on benefits."

Yeah well by how much? The current payroll gap is 1.95%, which for a median income household ($50k) works out to $975 or $2.67 per day. For a worker making $8 per hour the daily bill for a tax-based fix is 89 cents. Moreover the cost of the fix has been shrinking over time, in 1997 the gap was 2.23% and there is good reason to believe the improvement will continue. This reality does not get captured well by formulating it as "increasing taxes" which implies some heavy burden.

Same thing with "increasing the deficit". That sounds a lot scarier than "initially $30 billion and ramping up to $260 billion in inflation adjusted dollars then to zero in 2042". Anyone who thinks that deficits of that level are particularly problematic need to revisit some Reagan/Bush budget numbers.

While we are at it lets throw in "cutting back on benefits". Under current projections Social Security in 2041 has to slash benefits by 25%. Which looks like a lot but that is a function of the schedule of benefits that would have the check be in real terms 160% of a comparable one today. And as Professor Rosser always points out 75% of 160% equals 120%.

Which is to say that there is much adoo about not much at all here. Yeah in 2017 they have to start paying back some of what they borrowed. Just like they will have to pay off the Chinese Central Bank for any Treasuries coming due. But in context that is not scary, that's called governmental finance, and we should resist any formulation that suggests that this is unusual.

Because lost in this whole kerfluffle is the fact that despite the Trustees talking about an 'Unfunded Liability" of $4.7 trillion over the 75 year window or $14 trillion over the Infinite Future, the real 'Unfunded Liability' is Zero. Once the General Fund gets done paying back what it borrowed, with interest, under current law its obligation to Social Security is zero. If FICA can only pay 75% of the Schedule well tough. I'll be 84 and probably not caring. The real focus for the generation coming behind the Boomers should be the economy. If you can grow Real GDP at a rate of 2.8 the entire problem vanishes. Instead of talking about tax increases (however modest they would need to be) or raising the cap (and pissing off the upper middle class) or raising retirement age (most people take early retirement today) Gens X, Y and I suppose now Z need to get off the couch and grow the economy like we did.

2.8% Real GDP is not some crazy level of growth. Far from it. For that matter 2.4% would close about half the gap and we haven't had a five year period less than that back to 1960. When you put this in numeric terms and focus year by year on the real numbers most of the discussion simply terms into a big game of smoke and mirrors.

http://www.ssa.gov/OACT/TR/TR07/V_economic.html#wp192078

[robertdfeinman]

Bruce Webb:
First of all I don't think it is necessary for me to provide a context to what is a theoretical point. It may be a practical matter as to how much taxes will be raised at some point in the future, but that is a debatable issue in itself. I'm not interested in getting into (hypothetical) numbers.

I'm not trying to scare anyone into thinking that SS is in trouble as Anne seems to have assumed as well. I don't think it is either.

However your preferred solution is also based upon a hypothetical - the future growth rate. This is another case of what I claim are the unexamined axioms of contemporary economic thought.

I claim that there is a reasonable probability that growth, especially in the industrialized countries, will not continue much longer. The use of historical data is irrelevant. Things are the same until they're not.

I know that capitalists like to postulate growth, but we are faced with issues that have never occurred on a global scale before. The most important are overpopulation, resource depletion and climate change. Calling me a neo-Malthusian, as some do, may make them feel better, but it's not an argument, more like a prayer. One cannot argue by means of historical analogy. It's not like the sun coming up each morning. The future is unknown and changeable. The rotation of the earth, not so much.

To get back to the main point. SS is not in trouble in the near- or mid-term. Efforts to paint it as such are motivated by ideological or corporate opportunistic issues. It may be worthwhile to whack this mole every time it reappears, but can we now go back to the real issue - what do we do when the well (oil and water) runs dry?

[pgl]

islm - as I look at the Treasury's statement, it does say the Trust Fund reserves rose by about $250 billion in just 11 months. So we are talking about a surplus for the year that will be just shy of $300 billion. And it seems total Federal debt is getting really close to $9 trillion. Like I said this weekend - to which CoRev went off on some nonsensical babblespeak to suggest I had no clue. Glad to see the facts - as it seems I got this right. Thanks!

[johnchx]

anne writes: As long as the economy grows remotely close to the growth in debt there is no problem...

Perhaps (although I see no reason why there should be trend -- as opposed to cyclical -- growth in debt at all...why not just pay the bills?).

However, the big picture is this: When Ronald Reagan took office, total debt outstanding was 36-37% of national income. (I use national income, rather than GDP, as it's closer to the real tax base.) Twelve years later, when the reins were handed over to Bill Clinton, debt outstanding stood at about 75% of national income. Put another way, during the Reagan-Bush years, debt grew more than twice as fast as national income.

This trend was reversed to a small degree during the Clinton years; when George W. Bush took office, debt stood at about 66% of national income. And by 2006, it was back up to 73% of national income.

To bring our fiscal situation back into line with the pre-Reagan position, we would have to cut the national debt roughly in half, retiring about $4.3 trillion worth of bonds -- that is, 100% of the marketable Treasury debt held by the public (including debt held by foreign central banks).

[Peter Schaeffer]

All,

The public debt of the Unitd States is around 40% of GDP. Take a look at http://www.aaas.org/spp/rd/debt04b.pdf for some charts. Intergovernmental holdings account for the rest of the debt. However, these are liabilities of one branch of the government to another, not actual debt.

[Bruce Webb]

The adjustment in rates promoted by Greenspan was a fraud. It has been taking in more than it pays out for decades. The extra revenue has gone to fund normal government operations.

Sorry rdf in context that is somewhere between 'overstated' and 'wrong'. Social Security was making a serious run at Depletion in 1983. In fact at one point or another both OAS and DI hit zero which required some fancy borrowing between them and HI (Medicare A)
Historical Operations of the Combine OASDI Trust Funds
At the end of 1983 the Trust Fund ratio was only at 14 which translates to 51 days in reserves. Without the tax increase it would have gone broke, that's hardly 'fraud'.

Which brings up three other points. Given that the Federal Government did in fact finance Social Security through to 1983, redeeming all Special Treasuries, you can say that any borrowing before 1983 was moot. It all got paid back with interest, no harm no foul. Which takes a lot of force out of 'decades'. Note further that there was no net borrowing at all between 1957 and 1966, over this period the Trust Fund balance went from $23 billion to $22.3 billion. Then again there is a sustained period from 1975 to 1981 when the Trust Funds shrank year after year as the General Fund paid down on the Special Treasuries (which apparently were not considered 'phony IOUs' - what changed?).

Now lets take this up a notch. For Social Security the technical definition of 'crisis' is Short Term Actuarial Imbalance. Which is defined as a Trust Fund ratio below 100 in any one of the ten following years. If Social Security were a bank this would represent a Reserve Requirement. That this Reserve was kept in Treasuries is neither here nor there. If we go back to the link we can see that the Greenspan increase had the Trust Fund building at a very slow rate to 41 in 1988 and 96 in 1992. You can and I do argue that there are no excess contributions in any year with a ratio below 100. Meaning Reagan and Bush I are off the hook here. The notion that in any way we financed the Cold War out of Social Security is simply fiction. So under this definition the first time there was possibility of 'borrowing' in any real sense was 1993 when the Trust Fund hit 107 but I don't think there was any reason to think anything was amiss before say 1996 when the Trust Fund Ratio went above 150.

1997 was in fact the last time that Social Security could fairly be thought in 'crisis', the trend lines to Solvency were showing pretty clearly. We could have had an honest discussion about reallocating FICA on a short to medium term basis. But Gingrich and then Bush were not interested in an honest discussion. Any reduction in the flow of payroll dollars to Social Security put their other tax cuts at risk, so they didn't let us go there. But it is hard to look at any year before 1998 where anything was amiss at all. The 1983 Reforms did exactly what they were supposed to do in exactly the right time. The ten years from 1983 to 1993 put Social Security in Short Term Actuarial Balance.

Whatever you want to think about Greenspan, none of this was remotely like fraud. Social Security went out of balance in 1971, hit a dangerous low point in 1982 and 1983, and was restored to balance in 1993. Crisis faced and crisis solved.

Can we put the 'looting' myth to bed, at least for the years before 1998? Because the numbers just are not there to support it.

[anne]

"Why not just pay the bills?" Agreed; but that will take a more robust economy, a political climate in which even allowing tax decreases to revert, and a possibility of limiting military spending become reasonable policies. There is no possibility of dealing with the federal deficit before 2009, but there is no worry for the near term, since the debt is growing reasonably with the economy. As for cutting social benefit programs, I am in no way intetrested.

[ilsm]

Bruce Webb,

Thanks,

[Bruce Webb]

I would agree that the intragovernmental holdings are not debt, mainly because there is a very serious chance that the roughly 50% of that represented by the Social Security Trust Fund will never have to be redeemed by dollars from the public. If the economy hits Low Cost numbers then we never have to repay the principal. Instead the Trust Funds build to $78 trillion by 2085 and only paying out 11% of the accrued interest for that year.

Is there any meaningful sense in which that $78 trillion should be considered as debt? Now the $8.7 trillion that would be needed to generate interest to bridge the $453 billion gap between Income Excluding Interest in 2081 could be reasonably included. The other $69 trillion could simply be wiped off the books without the interest of any then current beneficiary being hurt.

It won't come to that. Under Low Cost the turning point may come in 2015 when Interest earned is first double the difference between Income and cost, at that point you would want to start making some headway on getting principal paid down, which itself can only be done by reducing FICA. Really we should have started examining FICA rates in 2000 and reducing them, but the concept that Social Security could conceivably be overfunded was simply too bizaare for most people to even think about it. And not too much has changed today, people are still talking about tax increases (cap raise) and benefits cuts (retirement age) without bothering to examine the economic assumptions of the models.

If housing doesn't take the economy totally over the cliff we will be having much different conversations in 2010 than most people expect.

[pgl]

Peter - I object. I've paid high payroll taxes most of my adult life and I do expect to have my Soc. Sec. benefits paid until I die. This notion that the government owes itself so none of this exists makes must of us 50-ish hard working folks sick. No, Paris Hilton should pay higher taxes on her income. Hands off my Social Security benefits. Seriously!

If this were some CFO taking funds out of the employee defined benefit program to put more funds in the hands of the shareholders, the CFO would be tossed in jail. Same should hold for Republicans who wish to do the same with respect to the Soc. Sec. Trust Fund.

Shame on you for even thinking about condoning grand larceny!

[2slugbaits]

Maybe Martin Feldstein should go back and reread his Sep 2000 paper "The Distributional Effects of an Investment-Based Social Security System," NBER Working Paper #7492. In that paper Feldstein admits that transitioning to an investment-based system would require that one generation bear the burden of a 3 percentage point hike in FICA contributions. No mention of that in Feldstein's latest offering. But then again, Feldstein has always been something of a Jekyll and Hyde character. The good Feldstein writes responsible academic papers. The bad Feldstein writes hack pieces for WSJ op-ed pages. Back in Sep 2000 Feldstein also recognized that returns to equity would be driven down if we went to an investment-based retirement system. Again, no mention of that this time around.

[johnchx]

Bruce Webb wrote: mainly because there is a very serious chance that the roughly 50% of that represented by the Social Security Trust Fund will never have to be redeemed by dollars from the public.

I have to disagree here. Since when do you get to fail to repay a creditor just because you don't think the creditor "needs the money?"

The "trust fund" represents a debt owed by income tax payers to payroll tax payers. I see no reason that the possibility that the payroll tax may have been set higher than necessary entitles INCOME tax payers to a free lunch.

If it turns out that Social Security is over-funded in the long run, the appropriate alternatives are (a) reducing the payroll tax or (b) increasing benefits.

A multi-trillion dollar free ride for high-earners and recipients of capital income should not, in my opinion, be on the table.

[ilsm]

Peter Scheaffer,

"However, these are liabilities of one branch of the government to another, not actual debt."

Actually, the intragovernmental debt reflects taxes collected for one thing and spent elsewhere.

There are something like 11 different appropriations all enacted by congress each under the US constitution and federal regulations.

Each appropriate establishes a different fiscal and legal entity.

We do owe the government held debt to ourselves as in the US government is us.

I think you need to reflect on the taxation without representation aspect, and the fact the debt ties future decisionmakers hands.

By the way should the FDIC need more than a few billion bucks to bail out insured depositis their part of the intagovernmental debt will have to be turned to cash pretty quick.

Let the inflation taxes roll!!

[Peter Schaeffer]

pgl,

A few quotes from earlier posts

“Once again the fundamental lie about Social Security - it's not a trust fund, it has no reserves.”

“I would agree that the intragovernmental holdings are not debt, mainly because there is a very serious chance that the roughly 50% of that represented by the Social Security Trust Fund will never have to be redeemed by dollars from the public.”

Let me offer a different argument. If Congress repudiated the public debt of the United States tomorrow, the economic consequences would be harrowing. The economy would be reduced to bartering and trading with Euros, Mexican Pesos, and Canadian Dollars overnight (at least that is the conventional wisdom).

Conversely, if Congress canceled the SS Trust Fund who would even notice? Would benefit payments stop? Be reduced? Of course, not.

The real issue is what set of taxes will be used to make SS payments after 2017 and 2028 and/or if benefits will be reduced (beyond the current slow escalation of retirement ages). That’s an eminently legitimate issue worthy of debate. However, it doesn’t constitute Federal debt in the same way that China’s ownership of Treasuries does (by way of example).

I wish I could take credit, but Bruce Webb has pointed out that general revenues were seamlessly used to liquidate the SSTF in the years up to 1983. This may well happen again. However, Congress could raise FICA taxes and defer the 2017, 2028, and 2041 dates arbitrarily into the future. Social Security benefits will be paid in any of the above scenarios.

Let me ask you a question. Let’s assume Bruce Webb is correct and the SSTF will never be redeemed using general tax revenues. Is that a problem for you? Are you insisting that SS taxes be reduced to make sure that the SSTF runs dry?

Let me ask another question. Taking into account the highly progressive benefit formula, are SS taxes more or less progressive than non-SS taxes?

[Peter Schaeffer]

Johnchx,

“The "trust fund" represents a debt owed by income tax payers to payroll tax payers. I see no reason that the possibility that the payroll tax may have been set higher than necessary entitles INCOME tax payers to a free lunch.”

Of course, the Federal government has revenue sources other than SS and the income tax. More generally I would agree with this statement. However, an obligation from one set of taxpayers to another (highly overlapping) set of taxpayers is rather far from a “public debt”.

[johnchx]

Peter Schaeffer writes: However, an obligation from one set of taxpayers to another (highly overlapping) set of taxpayers is rather far from a “public debt”.

Are you sure?

I notice that state and local governments hold about $240 billion in federal debt. That looks a lot like a debt from one group of taxpayers to another. It also looks a lot like "public debt." In particular, I think somebody would definitely object with great vehemence if Congress decided to repudiate this debt.

So it's not the "taxpayers owing taxpayers" that makes the difference. The real difference is in whether the debt is owed to a separate and distinct legal entity.

Now, the social security "trust fund" is not actually a trust fund. Its "trustees" are also not actual trustees. The debt to social security isn't owed to a distinct legal entity with standing to take legal action against expropriation.

But, to me, that's all the more reason to create maximum political pressure on Washington to behave as if it were a "real" debt to a "real" external entity. Because taking that debt seriously was absolutely central to the political bargain that created the social security surplus in the first place. If the "Greenspan plan" had dared to say "...and the projected payroll tax revenues will escheat to the general fund..." the deal would never have gone through.

This is the deal that Greenspan proposed. This is the deal that Reagan and the Congress agreed to. And I don't see any valid reason to accept anything less.

[James Killus]

The only answer that should ever be given to crap like this is "You aren't interested in 'fixing' Social Security, because it doesn't need fixing. You're only interested in looting it and then destroying it. Crawl back under your rock."

This is not to say that I do not admire the substantial information and analysis that is presented here. But remember, the WSJ editorial page deals in disinformation, propaganda, and lies. That should be asserted at every opportunity, and I'm glad to see how many here take the opportunity.

[Bruce Webb]

John at 3:57 misses the point. People can and do convince creditors into granting interest only loans. The federal government has convinced workers that they need to pay in at 12.4%,, yet the needed payout is maybe 12% of the roughly 5% owed. The technical term is 'arbitrage'. You could look it up.

[johnchx]

Bruce Webb writes: People can and do convince creditors into granting interest only loans. The federal government has convinced workers that they need to pay in at 12.4%,, yet the needed payout is maybe 12% of the roughly 5% owed. The technical term is 'arbitrage'. You could look it up.

Huh? I haven't a clue what you're trying to say here, or what it has to do with anything I said. Can you spell it out a little more explicitly?

[Bruce Webb]

Well John one answer is 'beer blogging'.

A more sober response would be that we simply are not having an honest debate about the economic assumptions that are going into these models. There is clear evidence that Low Cost is being shaped to a particular curve and that Intermediate Cost is then being forced to fit underneath that curve. If you read the Social Security Reports in sequence you see economic models under increasing strain with Low Cost consistently looking to be the better predictor. Yet outside my posting there is approximately zero discussion of the consequences of the economy returning Low Cost results. It would be one thing if someone had simply explained why 2.0% ultimate productivity was the top limit of future economic performance or why 1.7% really was the median projection but that never happens. Never and I have been asking for that for at least seven years now.

Low Cost is not only possible, in fact I suggest that it is probable. And if it happens then we see some serious deviations in outcomes starting about four years out. And if we don't take actions on FICA the trend lines end up taking us in some dangerous directions after 2015. Yet no one seems to be willing to engage on the numbers, everything is being held at the conceptual level and lazy assumptions are allowed to remain at the core of the argument.

There has been no case made for accepting Intermediate Cost assumptions. None, zip, zero, nada. It has consistently underpredicted economic outcomes for a decade with the result that depletion has been pushed back from 2029 to 2041. There is every reason to believe that trend will continue and that at a minimum we get some result better than Intermediate Cost. And at some point the numeric result crosses the boundary under which none of the principal in the Trust Fund ever needs to be repaid.

We know that Low Cost yields a result where in 2085 we are only paying out 11% of the interest accrued in that year. At results between Low Cost and Intermediate Cost that percentage increases. But there is no rational reason for accumulating a Trust Fund where total income (FICA, Tax on Benefits, and Interest) exceeds Cost indefinately. Yet that is the path we are on. If people actually examined the numbers.

I am fully prepared to be out thought and out argued on the numeric front. But it is a little irking to simply be ignored. Someone needs to make the case for Intermediate Cost or the case against Low Cost. Because literally trillions of dollars are in the balance. The differences in initial assumptions are relatively tiny, yet the differences in outcomes are huge, even over the medium term. It makes a great deal of difference whether Social Security becomes a net lender or a net borrower in 2017. We have one model where the General Fund has to borrow $30 billion from the public to pay Social Security benefits in that year. We have another model where Social Security is running a surplus of more than $350 billion. The combined effect is the equivalent of China liquidating its entire portfolio of Treasuries. All in one year and within the threshold of the current 10 year bond.

Under Low Cost this huge flow of funds continues for several years on either side of 2017. Wouldn't you think we should have a rational discussion of what the prospective entry of a couple of trillion dollars in the next 15 years might have on the bond market? Even if just in theoretical terms? Well you would think. But where is it?

The question under discussion is whether the Trust Fund balance is a debt or not. Well the real answer is that it depends crucially on economic performance over the next four or five years. Yet it appears that no one is actually willing to engage with the data tables. I have engaged in literally hundreds of Social Security threads over the years. It is astonishing how many of those threads simply cite no actual numbers at all. Well the details matter here. Concepts are fine but the argument is going to be settled at a more granular level. But is anyone except me actually linking back to the data?

Pardon me for being cranky but when it comes to Social Security some otherwise very intelligent people start spouting nonsense. Nonsense that is when measured against the actual numbers. A typical example is the wild rhetoric about 'looting' the Trust Funds, or the idea that huge amounts of dollars were ever there to start with. Well it just isn't so and a simple examination of Table VI.A4.-Historical Operations of the Combined OASI and DI Trust Funds, Calendar Years 1957-2006 would show that clearly. Net flows in and out of the Trust Fund were negligible before 1987 and not really significant at all before the middle Clinton years. And you can't steal what doesn't exist. A little bit of numeric prevention would go a long way here. But no one wants to take their medicine.

[Bruce Webb]

Here is a little thought experiment.

Suppose your doctor tells you you have six months to live. And you respond by refinancing your house with maximum equity extraction and cover it with death insurance. Further you get death coverage for all your eligible credit cards, maximize your credit limits and then limit out the cards. What is your total debt?

Well one answer is to total up your balances, you owe a bunch, plus carrying costs. Another is to take your debt service costs and multiply it by the number of months you have to live. If your average loan rate is 12% then your actual debt is 6% of the cash equivalent of whatever you bought. Wham you just got a 94% discount. And your heirs 100%.

A variation of this is true for Social Security. Its actual debt in any given year is the amount of gap between Income excluding Interest and Cost divided by average interest on the money borrowed. In fact if you don't have to borrow money in any given year there is a real sense in which your debt is zero.

Which is why counting the dollars in the Intragovernmental Trust Funds as debt is misleading. The real debt in any future year is simply the cost gap divided by the interest rate. Under intermediate cost assumptions every bit of the Trust Fund will be paid out by 2041. So the real value of the debt is the face value; $2.1 trillion. Under Low Cost assumptions by 2085 the payout is $450 billion on accrued interest of $4 trillion or 11%. Which suggests that the real value of the current debt under those assumptions is $231 billion.

The nominal value of the debt balance is not nearly as important as the cost of servicing it. Plug 3.0% GDP into your model and arguably that debt goes to zero. No debt service, no debt.

Of course it turns the government from a borrower into a thief, which is why we will need to address the whole thing from the income side. But the issue is not so simple as adding $5 trillion to $4 trillion and imagining that the result has a specific real world meaning.

[Patricia Shannon]

Bruce Webb says...

Same thing with "increasing the deficit". That sounds a lot scarier than "initially $30 billion and ramping up to $260 billion in inflation adjusted dollars then to zero in 2042". Anyone who thinks that deficits of that level are particularly problematic need to revisit some Reagan/Bush budget numbers.

http://www.hightowerlowdown.org/node/1343

This sounds to me like you are saying the Reagan/Bush budget numbers were not problematic. This is the Republican equivalent of those liberals who say - why worry about what happens when we withdraw from Iraq, we withdrew from Vietnam and everything turned out all right, it led to golf resorts. In fact, the Reagan/Bush deficits lead to the recession that caused Bush I to lose in 1992.

[Bruce Webb]

Patricia in 1988 no one imagined we could turn Reagan deficits to surpluses. Yet Clinton did. I just suggest a little less Chicken Little here. Would my priorities be different than Bush's? Boy Howdy. Does that mean current deficits are unsupportable? Nope.

[johnchx]

Bruce Webb:

Here is where we differ. You assume that the payroll tax rate must remain at its current level, even if it generates a long-term surplus, and that this surplus will be available to finance basic (non-Soc Sec) government operations.

This completely violates the terms of the political compromise which allowed the payroll tax to be raised to its present height. I decline to overlook the violation.

The premise of the payroll tax increase was that it was necessary to partially pre-fund the future spending needs of the Social Security system. The promise that went with the tax increase was that the surpluses it generated would be temporarily lent to the government, but would subsequently be drawn down to zero by paying SS benefits. A surplus was necessary in the first place only to stabilize the payroll tax rate itself -- i.e. we raised taxes in the 1980s to avoid raising them by more in the 2000s.

The whole point of the "trust fund" arrangement was to make this promise concrete, visible and enforceable. "Netting away" the "trust fund" surplus violates the promise, and implements by stealth a regressive tax increase to pay for general government.

If the surpluses envisioned by the Low Cost scenario should arise, the premises and promises that underlie the present system imply that we should reduce the payroll tax rate to the point that the surpluses disappear. The system isn't supposed to run a permanent surplus, remember?

Now, imposing a regressive, wage-only tax to help balance the (non-Soc Sec) federal budget is certainly a valid policy option. You are perfectly free to advocate a regressive tax on workers. I don't think it's a very good idea, myself, but it's a free country.

What I do object to is anything that would impose this regressive tax hike on the sly. And "assuming away" the "trust fund" is exactly that.

Maybe I'm off base somewhere. Here's my question to you: is there an argument from policy or fairness for why the payroll tax shouldn't be reduced to bring the "trust fund" into balance (assuming, as you do, that the Low Cost scenario economic assumptions are likely to materialize), or is it simply (a) inertia and/or (b) we need the money?