…because I’m an economist and a mom–that’s why!

Useful Lessons from CBO’s New Report on Social Security

August 22nd, 2008 . by economistmom

Yesterday CBO released an updated analysis of the Social Security program, with lots of interesting ways of characterizing the range of uncertainty in the projections of benefits and taxes.  The cover graphic (which is also Figure 1 on page 9) tells us the “bottom-line” story:  it’s almost certain that Social Security revenues will start to fall short of scheduled Social Security benefits in 10-20 years.  The report and the CBO Director’s Blog highlights several useful lessons that come out of what they see as their most interesting numbers.

First, while a shortfall in Social Security is almost certain, a shortfall larger than 2 percent of GDP is unlikely:

[T]he likelihood that outlays will exceed revenues in 2030 is about 97 percent, CBO projects, and there is almost a 50 percent chance that the gap will be larger than 1 percentage point of GDP; the chance of its being 2 percentage points (or more) of GDP is only 6 percent.

(This is what makes Social Security a relatively “small” problem compared with the challenges facing Medicare.)

Second, the shortfall threatens the net benefit of the program to younger generations, not older ones:

According to CBO’s projections, the 1940s cohort, for example, is virtually certain to receive all of its scheduled first-year benefit. The 1990s cohort has only a 32 percent chance of receiving all of its scheduled first-year benefit but an 84 percent chance of receiving at least 70 percent of that benefit.

So one could interpret this intergenerational distribution as the Social Security program providing less “insurance” to younger generations. 

But third, what is considered a “fair return” to those younger generations depends on what measure of benefits one considers most relevant–which is a value judgment.  So this is where many of us have (and will continue to have) disagreements… 

  • Based on the first-year benefit highlighted above (see Figure 5 on page 13), the “unfairness” can be characterized as younger generations facing a larger divergence of payable benefits compared with scheduled benefits.  On the other hand, first-year payable benefits, in thousands of real (2008) dollars, are still higher for younger generations than older generations. 
  • Based on replacement rates (or how much of career-average annual earnings are “replaced” with first-year benefits–see Figure 7 on page 15), the shortfall in Social Security significantly reduces the (payable) replacement rate to younger generations below the scheduled benefits that would otherwise provide a replacement rate similar to that given to older cohorts.
  • Based on lifetime benefits (see Figure 9 on page 17), even with the shortfall under the payable benefits scenario, median lifetime Social Security benefits to younger generations are still higher in real dollar terms than the median lifetime benefits to older cohorts.  And there’s really no uncertainty about this in CBO’s analysis (based on Figure 12 on page 20).
  • But based on the ratio of lifetime benefits to lifetime taxes (see Figure 13 on page 21), younger generations are likely to get a lower “lifetime return” on the taxes they paid into the system, compared with older cohorts.  For example, the shortfall makes it more likely than not that middle-income individuals born in 1990-2000 (i.e., many of our kids and grandkids) will receive lifetime Social Security benefits that fall short of their lifetime Social Security taxes paid (providing negative lifetime return).

A fourth useful lesson, featured in Box 1 on page 3 of the report, is that the Bush income tax cuts worsen the outlook for the Social Security program–and not just the outlook for general revenues.  Extending the tax cuts would reduce projected revenues from the taxation of Social Security benefits and hence increase the Social Security 75-year shortfall, to -0.47% of GDP or -1.30% of taxable payroll, compared with -0.38% of GDP or -1.06% of taxable payroll under the “extended baseline” used in the main projections.  Yet another reason why we ought to think carefully before extending any of the Bush tax cuts and why our mental starting point ought to be the level of revenues achieved under the current-law baseline.

I’m sure they’re many other useful lessons from this CBO report, which I’m likely to focus on with greater detail in future posts, especially with input and feedback from some of you readers.  And others will no doubt take away different highlights and different lessons.  (Andrew Biggs has already commented on the report with a couple different posts on his blog.) 

22 Responses to “Useful Lessons from CBO’s New Report on Social Security”

  1. comment number 1 by: Andrew Biggs

    A good post (as usual!). A couple quick thoughts:
    First, I think what matters for individual treatment is the ratio of benefits to taxes (or the IRR or something similar). The problem is that there’s not much we can do about it: the treatment of future beneficiaries is a function of future demographic and economic outcomes, which we have only limited control over, and the large net benefits paid to early participants in the program (the ‘legacy debt’), which is a sunk cost. We can alter the pattern of benefit/tax ratios by having current cohorts pay more, but we’re pretty much stuck with the average treatment going forward.

    Second, unless I’ve misread the CBO report (which is possible…) I think the note on the effect of the Bush tax cuts isn’t actually right. The Social Security Trustees and actuaries project that future income tax revenues will remain roughly constant relative to GDP, since the tax/GDP ratio has been fairly constant historically. CBO instead uses a literal projection of ‘current law’, which implies higher revenues. But we should be clear about what the new CBO tax baseline implies: from 1957 through today, individual income tax revenues have have averaged 8.2% of GDP. Under the CBO extended baseline, by 2082 income tax revenues have reached 15.9% of GDP, almost double. That’s due to the effects of real bracket creep (not indexing tax brackets for real growth of incomes). Even if the Bush tax cuts were made permanent, revenues would still increase (to around 12% of GDP). The real effect of the Bush tax cuts, if projected out 75 years, would be around half the difference between the new CBO numbers and the Trustees’ projections. More broadly, CBO’s baseline is what literally would happen if we never changed current law, but it’s also pretty unrealistic relative to historical norms.

  2. comment number 2 by: economistmom

    Andrew: Not sure what you mean by the note on the tax cuts being not actually right, but both my reference and CBO’s box are comparing CBO projections using the “extended baseline” scenario (where the tax cuts expire as scheduled) to CBO projections using the “alternative fiscal” scenario (where the tax cuts and AMT relief are extended). All I’m saying (and all CBO is pointing out) is that the projected outlook for Social Security is worsened if the Bush tax cuts are extended, compared with if they are not extended. And yes, you’re right that even if we extend all the Bush tax cuts and AMT relief, that revenues as a share of GDP would still go up over time–the natural result of real income growth interacted with a progressive income tax system.

  3. comment number 3 by: Andrew Biggs

    EconMom: There are basically three scenarios: CBO’s extended baseline, where the tax cuts expire; CBO’s alternative baseline, where the tax cuts are made permanent and AMT extended; and the Trustees baseline, where income tax revenue remains constant vs. GDP. Roughly speaking, under the first income taxes would reach 16% of GDP by 2080, under the second they’d be around 12%, and under the third they’d be around 9% (the current level). The 0.24% of payroll difference CBO cites is between the first and third scenarios; my guess is that between the first and second the difference would be around half that much. Not a huge deal, but I thought it was worth pointing out. Thanks!

  4. comment number 4 by: Bruce Webb

    Andrew’s post at ‘Notes’ didn’t allow comments so I would like to thank him for the link here.

    I put up the link and some excerpts at Angry Bear with a focus on contrasting CBO outcomes with those of the Trustees. I didn’t editorialize but just pointed out that a back of the envelope calculation suggests an immediate payroll gap of about 1.2% under CBO projections. Which would seem to translate to a similar 25% cut in PV unfunded liability. But as I hasten to point out often I am just a number pointer and not a number cruncher. Any corrections/additions/comments are of course welcome.

  5. comment number 5 by: economistmom

    Andrew: No, the numbers in that CBO Box 1 compare the two CBO scenarios, regardless of what the Trustees assume/project.

    Here are 75-year shortfall numbers for the three scenarios you describe, from that Box 1 of the CBO report (page 3) and from Appendix B of the report (page 33) which compares CBO (baseline extended) and Trustees projections:

    shortfall under CBO baseline extended: 1.06% of taxable payroll (or 0.38% of GDP);

    shortfall under CBO “alternative fiscal scenario” (tax cuts extended): 1.30% of taxable payroll (or 0.47% of GDP);

    and shortfall under Trustees: 1.70% of taxable payroll (or 0.61% of GDP).

  6. comment number 6 by: Andrew Biggs

    I stand VERY corrected on that — you’re right. I was reading too quickly and apologize. So the difference regarding income tax assumptions between CBO and Trustees accounts for more than 0.24 percentage points of the difference between the two forecasts. My bad!

    I think I’d still hold that even the alternative fiscal baseline is pretty generous in terms of revenue from benefit taxation. It’s assuming income taxes/GDP over 40% above current/historical levels. Again, it’s not literally incorrect but the result of extending an incompletely indexed tax code over long periods.

  7. comment number 7 by: Jim Glass

    … third, what is considered a “fair return” to those younger generations depends on what measure of benefits one considers most relevant–which is a value judgment.

    Well, one might remember that FDR and the founders of Social Security made explicit promises about this. E.g., FDR’s head of SS, Arthur Altmeyer,”selling” SS to the public in 1936

    “Moreover, every worker eligible …will receive a monthly retirement benefit upon reaching the age of 65 larger than he could purchase from any private insurance company with the taxes he will have paid.”

    That’s a pretty objective measure. It was based on SS returning basically the federal bond rate on contributions for every annual cohort, which is what FDR’s original SS Act provided for. (Not only was that a better rate than commercially available annuities, but of course the great majority of workers then couldn’t purchase an annuity or pension in the market at all).

    But as soon as the SS payroll tax money started coming in 1939, Congress began changing the law to front-load benefits and back-load taxes to pay for them, increasing the “return” to early workers to way above the bond rate, and correspondingly reducing the return to later workers (who weren’t voting yet).

    In 1944 FDR vetoed such changes, and Altmeyer urged Congress to uphold the veto, saying of the changes Congress was creating…

    “It is a mathematical certainty that [they] will eventually necessitate raising employee’s contributions rate later to a point where future beneficiaries will be obliged to pay more for their benefits than they would if they obtained this same insurance from a private insurance company.

    “I say it is inequitable to compel them to pay under this system more than they would have to pay to a private insurance company, and I think Congress would be confronted with that embarrassing situation…”…


    “If we should let a situation develop whereby it eventually becomes necessary to charge future beneficiaries rates in excess of the actuarial cost of the protection afforded them, we would be guilty of gross inequity and gross financial mismanagement, bound to imperil our social insurance system….” (SSA).

    That’s a pretty clear and specific definition of “fair return” coming from FDR and Altmeyer — a tax cost of benefits not exceeding their actual actuarial value, or the cost of commercially available pension equivalents, lest we “be guilty of gross inequity and gross financial mismanagement, bound to imperil our social insurance system….”.

    It’s also a definition that has been pretty much lost down the memory hole. With today’s and future workers required to take a $15 trillion loss, present value, relative to the bond rate, one can see why.

    But Altmeyer’s warning ….“bound to imperil our social insurance system.” … is exactly on the mark. The real danger to SS as an institution in the future isn’t that the trust fund is going to “run out” of bonds in 2040 or whenever, or that it is underfunded by 1.x% of payroll, or anything like that. The real danger is that the political future of SS is going to be the reverse of the past.

    Among past retirees SS was hugely popular because it made them $15 trillion richer as a group, above the market return, in a world when most of them couldn’t even get a market-return investment. Who wasn’t going to love that?

    But future participants are going to be made $15 trillion poorer as a group, below the market return, when they all have superior market returns readily at hand — even if only by buying T-bonds directly through Treasury Direct. That’s a $30 trillion swing down in return. Who’s going to be so happy with that? Politically, how are future voters going to line up on that?

  8. comment number 8 by: Brooks

    First, while a shortfall in Social Security is almost certain, a shortfall larger than 2 percent of GDP is unlikely…(This is what makes Social Security a relatively “small” problem compared with the challenges facing Medicare.)


    I must point out that I disagree with the conceptual framework implied in that statement. What makes Medicare a bigger problem than Social Security is that Medicare spending is projected to be greater, not the size of any Social Security shortfall (projected SS FICA revenues plus repayment of Trust Fund bonds and interest less Social Security spending). Even if Social Security were projected to be infinitely, fully “solvent” (zero shortfall ever), it would still be contributing to our fiscal imbalance by exactly the amount we spent on it. ALL spending contributes to the fiscal imbalance, just as ALL revenues reduce it. The fact that Social Security has a dedicated tax that is carved out from other taxation is irrelevant to the big picture, which is our OVERALL fiscal imbalance (TOTAL revenues less TOTAL spending).

  9. comment number 9 by: economistmom

    Brooks, I don’t know… I still think that Social Security is not “as big” a problem in terms of fiscal sustainability as Medicare is– because we pretty much have a dedicated tax base to fund Social Security with a steady stream of revenues which does not show signs of being too widely divergent from the forecasted stream of benefits (program costs). With health care spending, it’s a different matter entirely, because the path of outlays is projected to rise so steeply and so obviously faster than the rise in anything we currently tax–or any form of our income (public or private) more generally. That’s not to say that we won’t have to tap into the tax base currently devoted to Social Security to help fund health care and other government spending, but it’s just to say that the fiscal condition of Social Security on its own is a much more manageable situation than is Medicare (or health spending in general), especially if we take corrective action sooner rather than later.

  10. comment number 10 by: Brooks


    Well, by that logic, wouldn’t one say that if Social Security were projected to be fully solvent forever (i.e., no shortfall of revenues), then Social Security spending does not contribute at all to our overall fiscal imbalance, and that therefore cutting Social Security spending could not reduce the overall fiscal imbalance?

    That was a rhetorical question, my point being that the dedicated tax is irrelevant to the big picture — our overall fiscal imbalance (total revenues less total spending). Obviously we could reduce overall deficits by reducing Social Security spending even in the infinite, full solvency scenario above. We could just reduce SS FICA rates (to avoid unnecessary Social Security surpluses), and offset those tax cuts with increases in other taxes (so no net tax cut; i.e., revenue-neutral), and the result would be lower overall projected spending, unchanged projected revenues, and therefore lower projected deficits.

    I think the conceptual error comes from confusing a dedicated tax with a self-financing program. Social Security is not a self-financing program. It is financed by taxes on the same total population as other taxes, only instead of the revenues going into general revenue and then being taken out of general revenue to be spent, it is dedicated to Social Security (even if the surplus is “borrowed” for other spending). That’s just a matter of which pocket we put how much of our income into. It’s still part of the whole.

    By contrast, I had a client years ago that was a self-financing quasi-governmental program. It started as a state government program to serve a venture capital function (to attract and grow high tech businesses and increase high tech jobs in the state). Initially it was funded with bonds, but then transitioned to self-financing via returns on its equity investments. Totally different story from Social Security.

    I’ll take the liberty of pasting my “Defense Tax Illustration”:

    Let’s assume, arguendo, that you favor cutting Defense spending as part of the solution to our long-term fiscal imbalance. Now let’s assume that tomorrow a “Defense Tax” is put in effect, with revenues dedicated to the Defense budget, projected to fully provide for a continuation of the current level of Defense spending, and let’s assume that some other taxes that go into the general fund are lowered such that all the changes end up revenue-neutral. Now, all of a sudden, presto! — Defense is “solvent”. Would you now say “Well, it doesn’t make sense to look at cutting Defense spending as part of the solution to our fiscal imbalance, because Defense is “solvent”, and therefore it does not contribute at all to our overall fiscal imbalance? I assume you would not say that, because that would be nonsensical. We can adjust the Defense Tax rate up or down as we wish, in accordance to any increases or decreases we choose to make in Defense spending, and if we choose to reduce projected Defense spending, we can reduce “Defense taxation” and offset that revenue loss with increases in other taxes, thus lowering projected overall spending, keeping projected overall revenues unchanged, and thus reducing the projected overall fiscal imbalance. And the same applies to Social Security.

    Spending on Social Security should be viewed in the same way as any other spending, regardless of whether it’s financed via a dedicated tax or not. Every dollar of Social Security spending contributes exactly the same amount to our fiscal imbalance as every dollar spent on anything else: exactly one dollar. We should look at Social Security spending and revenues in the context of overall spending and overall revenues, not in a vacuum, and we should decide how much to spend on Social Security based on economics and on our values and priorities, just like any other fiscal policy decision. The relationship between projected spending and projected revenues per SS FICA as currently structured is not, in itself, relevant to that decision, except as a administrative matter.

    Please see my post with a fuller explanation at

  11. comment number 11 by: Bruce Webb

    I spent some time this morning {Aug 3rd actually} with the Historical Tables (IV.A2 and IV.A4 specifically) and did the simple exercise of adding up actual benefits paid out.
    OAS 1941-1956 $25.6 billion
    OASDI 1957-1968 $181 billion
    OASDI 1969-1980 $793.1 billion
    OASDI 1981-1992 $2.476.5 trillion
    OASDI 1993-2000 $2.829.3 trillion
    OASDI 2001-2007 $6.330.9 trillion

    For a grand total of $12.636.4 trillion. In light of this Jim’s insistance that “Among past retirees SS was hugely popular because it made them $15 trillion richer as a group, above the market return,” is simply wrong on its face and is the result of a misreading of Table IV.B7. About $10 trillion of the amound he is claiming for “past retirees” is actually the projected gap between income and cost for the years 2083 to 2108. ‘current participant’ simply doesn’t mean what Jim’s formulation would require.

    You simply cannot extract an EXTRA $15 trillion in benefits for past retirees out of a total $12 trillion paid out to date, with fully half of that in the last 7 years.

    Jim’s ‘past retirees’ in fact includes everyone who is 15 and older in 2008 (the definition of ‘current participant’ for the purposes of Table IV.B7). There is a fatal confusion of past and future going on here.

  12. comment number 12 by: Andrew Biggs

    You’re right that Social Security spending (like any spending) contributes to the fiscal gap. But Social Security as a program — taxes and benefits — contributes only to the degree that benefits exceed taxes. When most people talk of Social Security and Medicare they’re thinking of the net cost, not the gross.

    Bruce, I don’t know exactly how you’re calculating your numbers, but I’m pretty sure those benefits are expressed in nominal or real dollars while Jim Glass’s $15 trillion is a present value. These are incompatible unless converted. You’re right that the net benefit paid to past/current participants includes people who have just begun working, and so the time frame extend out 100 years. That said, the $15 trillion (or whatever) net benefit paid to past/current participants is almost entirely a function of net benefits paid to past participants. Most current and future retirees will receive internal rates of return below the trust fund bond rate, which means that those cohorts would actually reduce the $15 trillion. In other words, in present value terms, past participants likely received a net transfer of more than $15 trillion.


  13. comment number 13 by: Brooks


    I think you may be missing my conceptual point. By your logic, if Social Security were projected to be infinitely, fully “solvent”, you would say that Social Security, as a program, does not contribute at all to our overall fiscal gap, which would mean that cuts in Social Security spending, combined with equivalent cuts in SS FICA taxation, could not reduce the overall fiscal gap, which, as I’ve explained, would not be a valid statement (we could cut Social Security spending, cut SS FICA accordingly, and offset that tax cut with tax increases elsewhere, resulting in lower projected overall spending, unchanged projected revenues, and therefore lower projected deficits).

    Similarly, please see my “Defense Tax Illustration”. By your logic, if instead of funding Defense via general fund tax revenues we created a dedicated “Defense Tax” with a tax rate sufficient to cover the current level of Defense spending in every future year, you would say that the “Defense Program” is not contributing at all to the overall fiscal gap, and again, the corollary would be that reducing Defense spending (and reducing the Defense Tax accordingly) could not reduce the overall fiscal gap. Obviously such a statement would not be valid.

    Unless a program is truly self-financing (in the sense of the example I gave of my client, the quasi-governmental entity that financed itself on an ongoing basis from returns on its equity investments, which were continuously reinvested, etc.), it is part of the whole, period, except for administrative purposes. Funding via a dedicated tax does not change that fact; it just carves out part of the overall taxation to pay for a particular program that is part of overall spending. We can increase or decrease spending on that program, and we can increase or decrease taxation for that program. The dedicated tax financing structure doesn’t matter from the standpoint of how much that program contributes to our overall fiscal gap (it contributes exactly the amount of the spending), nor from the standpoint of decision-making (other than administrative) regarding how much to spend on that program.

  14. comment number 14 by: Bruce Webb

    Andrew I know where Jim is getting his figures, both his $17 trillion and his $15 trillion, because he made a point of linking me to the table where he got them. Which is to say Table IV.B7. If you read the text associated with the table you can see that it defines ‘current participant’ as anyone 15 or older in 2008 and that both the $17 trillion and the $15 trillion represent costs projected forward not covered by projected income. There is in fact no connection between these numbers and benefits paid out in the past. None, zip, nada.

    There may be some value in calculating out the constant dollar value of past benefits, I don’t see what that would be because those benefits were paid out in then current dollars but if someone wants to try out the exercise fine. It might even be that the number approximates $15 trillion (I rather doubt it). But Jim told me directly (and claims he has told me before) that the $15 trillion in excess benefits to past retirees is clearly shown in Table IV.B7. It isn’t.

    We have an unfunded liability over the next 75 years that under IC projections is $4.3 trillion dollars per Table IV.B6. Some small portion of that represents scheduled benefits for current retirees over their remaining lifetime. Some more represents accrued benefits for people already in the workforce. But most of it is to pay projected benefits for people who were born after 1993 (2082-89=1993) and so will not be hitting the workforce before 2011 at the earliest.

    Of the $15.2 trillion that Jim cites $9.9 trillion of it represents unfunded liability for scheduled benefits payable in the years 2083 to 2108. No current retirees will still be drawing benefits then and very, very few current workers. To project that back is simply a reading mistake. Anyone can make a mistake, certainly I have in the past which is why I always to supply a link to my data source. But the evidence is very clear that Jim simply misinterpreted the meaning of the term ‘current participant’ to mean ‘past retiree’. This isn’t some disagreement about nominal vs real, this is about Jim misreading a Table and its associated text.
    Table IV.B7.—Present Values of OASDI Cost Less Tax Revenue and Unfunded Obligations for Program Participants[Present values as of January 1, 2008; dollar amounts in trillions]

  15. comment number 15 by: Jim Glass

    Andrew Biggs wrote:
    I’m pretty sure those benefits are expressed in nominal or real dollars while Jim Glass’s $15 trillion is a present value. These are incompatible unless converted.

    That’s the way it looks to me.

    Bruce Web wrote:
    I know where Jim is getting his figures, both his $17 trillion and his $15 trillion, because he made a point of linking me to the table where he got them. Which is to say Table IV.B7. If you read the text associated with the table you can see that it defines ‘current participant’ as anyone 15 or older in 2008

    The actual text from the Trustees Report reads…

    “…Subtracting the current value of the trust fund (the accumulated value of past OASDI taxes less cost) gives a closed group (excluding all future participants) unfunded obligation of $15.2 trillion. This value represents the shortfall of lifetime contributions for all past and current participants relative to the lifetime costs associated with their generations.”

    And SSA also tells us that workers born after around 1970 are getting less than the bond rate — so that gives us a pretty clear picture of the line break between the *past and current* participants getting the $ 15 trillion, and the *current and future ones* getting the $-15 trillion.

    But the evidence is very clear that Jim simply misinterpreted the meaning of the term ‘current participant’ to mean ‘past retiree’

    While Bruce wants to eliminate *past* participants from “past and current participants”, and include in in current participants getting more than the bond rate *all* of them — as young as age 15(!) — even though he knows most will be getting *less* than the bond rate, and thus taking the loss.

    Bruce also chooses to forget the real rates of return on contributions by generation provided by SSA, that paint a pretty clear picture of who got more than they paid in, and who’s getting less …

    []36% real annually for the very first “Ida May Fuller”retirees, to

    []12% for those retiring in the 1960s (when Paul Samuelson praised Social Security for being an actuarially unsound Ponzi scheme that works), to

    []5% for the retirees of the 1980s, when Congress protected benefits for then-seniors by cutting benefits for, and increasing taxes on, the then-young, to

    []Under 1.9% and falling (less than the projected “risk free” rate on Treasury bonds of 2.9%, and thus a real economic loss) for today’s young workers born in the 1970s and later.

    Bruce also is going to have explain to the Treasury that it doesn’t understand things any better than I do…

    “Why must the system increase net receipts by $13.6 trillion if it is already requiring current and future workers to pay in more than they will receive? The answer relates to the system’s generosity to early birth cohorts — generations of workers now either retired or deceased.

    “Social Security paid these previous cohorts benefits that exceeded their lifetime contributions by more than $13.6 trillion.” [earlier year dollars]

    And he’s going to have to explain to Krugman that he’s wrong too… (and PK has a rep for not taking criticism lightly!)

    “Social Security … has turned out to be strongly redistributionist, but only because of its Ponzi game aspect, in which each generation takes more out than it put in.

    “Well, the Ponzi game will soon be over, thanks to changing demographics, so that the typical recipient henceforth will get only about as much as he or she put in (and today’s young may well get less than they put in).”

    The comment system spam trap won’t let me provide the links to all of these, but Bruce has seen them all before so it doesn’t really matter.

  16. comment number 16 by: Bruce Webb

    Jim play around with words and bond rates all you want. Your conclusion that past retirees took $15 trillion more out of the system above and beyond their expected rate of return is simply not justified by the Table and its associated text.

    “With today’s and future workers required to take a $15 trillion loss, present value, relative to the bond rate, one can see why.”

    Well no. Almost all of that $15 trillion dollar ‘loss’ represents payouts to todays’ and future workers form todays’ and future workers and has nothing to do with past checks. Because you simply seemed to have missed the import of the language you site:

    ““…Subtracting the current value of the trust fund (the accumulated value of past OASDI taxes less cost) gives a closed group (excluding all future participants) unfunded obligation of $15.2 trillion. This value represents the shortfall of lifetime contributions for all past and current participants relative to the lifetime costs associated with their generations.”

    ‘Current participants’ are defined as: ‘For this purpose, current participants are defined as individuals who attain age 15 or older in 2008.” (A sentence that BTW immediately precedes the text you cite.) Which is to say you are including people who won’t be beginning to draw benefits until 2060 in with your ‘past retirees’. And yes I am separating out ‘past’ from ‘current’ because the numbers show that almost all of that cost is forward in time and not past and indeed 2/3rds of it coming in the 25 years between 2083 and 2108.

    Past beneficiaries collected a total of $12 trillion dollars in benefits with right under a trillion of that paid out by 1980 at which point all future retirees would have been paying in their entire work careers. (1980-1936 = 44 + entry to work force at 21 = age 65). From 1980 to 1983 the TF went to near zero meaning all past checks had been financed out of past contributions. Since then workers have been paying in more than an overlapping group of retirees have been taking out with the result we have $2.2 trillion dollars credited to SS to pay for future benefits. To try to flip this around and insist that somehow these people extracted some $15 trillion in extra value over and beyond their contribution needs a simpler explanation than you are presently delivering.

    Under Intermediate Cost the actuarial balance between 2008 and 2032 is +.38% Table IV.B4.—Components of Summarized Income Rates and Cost Rates, Calendar Years 2008-82[As a percentage of taxable payroll] at which point the youngest retiree eligible for full benefits will be 91 years old.

    Sorry you are trying to put a $15 trillion load in a $1 trillion (payouts to 1980) or a $12 trillion (payouts to 1997) sock not even accounting for the fact that we have surpluses projected forwards to 2017.

    And don’t worry I’ll tell Krugman why he is wrong next time he brings it up.

    Try this another way. Unfunded obligation for past and current participants.
    2006: $13.4 trillion
    2007: $14.4 trillion
    2008: $15.2 trillion

    Are you suggesting that between 2006 and 2008 past participants managed to collect an extra $1.8 trillion putting current and future participants on the hook for that much more? A neat trick from a bunch of dead to simply elderly folk.

  17. comment number 17 by: johnchx

    Ah Brooks…you simply cannot seem to contemplate the possibility that people aren’t confused — we just disagree with you.

    Maybe it will help if I tell you that my answers (I can’t speak for our host, of course) to your rhetorical questions are exactly the opposite of what you seem to think they must be. First, I would have no difficulty at all agreeing that, were Social Security actuarially fully funded, then cuts in SS spending would not contribute to closing the fiscal gap. That, in fact, has been more or less my position all along - and isn’t, by the way, contingent on the actuarial status of the program.

    And yes, if Congress enacted a dedicated tax to fully finance defense spending, I would regard defense cuts as “off the table” as a gap-closing measure.

    At bottom, the fiscal gap is not a financial problem. It’s a political problem. It is the gap between our political system’s ability to “will the ends” (authorize spending) and “will the means” (authorize corresponding taxation). With regard to Social Security, the political system has willed means more-or-less commensurate with the ends. If it were to do the same with defense, I would view defense is essentially the same light.

    Here’s the thing: if the problem is going to be solved, that is how it is going to happen. We will look at defense and say, “How the heck are we going to pay for this?” And we’ll find a way. And farm subsidies. And health care. And the National Science Foundation. And so on.

    Saying, “Let’s insist on looking at the gap as if the commitments we’ve made in the past don’t count anymore,” makes the process of assembling solutions harder, not easier. Because you can’t cut a deal if everyone knows that the deal can be undone — retroactively! — tomorrow. Respecting the commitments we made twenty-five years ago makes future deals possible.

  18. comment number 18 by: Brooks


    You write:
    And yes, if Congress enacted a dedicated tax to fully finance defense spending, I would regard defense cuts as “off the table” as a gap-closing measure.

    I guess that says it all. If I’m understanding you correctly (i.e., if you are really answering my question in that illustration), then you are saying that if Defense were funded via a dedicated tax and if it were fully “solvent” (projected dedicated tax revenues covering projected spending), you would think that cutting Defense spending (and cutting the Defense tax and offsetting that tax cut with increases in other taxes) could not be one way to reduce the overall fiscal gap. And you offer this to illustrate that you are not confused, but simply disagree with me. No, if the above is your contention, you are confused. It’s not a matter of opinion; it’s a matter of algebra. Doing so would lower projected overall spending while leaving projected revenue unchanged, thus reducing projected deficits.

    Now, if you’re not really answering my question and saying that cutting Defense spending in that scenario couldn’t reduce the overall fiscal gap, but only that you would consider cutting Defense spending “off the table” (i.e., not even something that should be considered) as one way to reduce the overall fiscal gap, simply by virtue of it being funded via a dedicated tax, then I’d have to ask why. Imagine we had a filibuster-proof Republican Congress this year and they doubled the Defense budget and carved out of general fund taxation a dedicated Defense tax that was high enough to fully fund that level of spending for the next 50 years. In other words, they didn’t change the level of overall general fund taxation, but just carved out a dedicated tax and reduced other general fund taxes accordingly for revenue-neutrality. Now, if next year we had a Democratic president and Congress and cutting Defense were possible, are you seriously telling me that you would say that, because Defense is fully “solvent”, they should not even consider cutting Defense as one way to reduce our overall fiscal gap (along with the aforementioned tax shift — cutting the Defense tax and offsetting it with increases in other taxes)?? If you are saying that, please tell me why? And if you are not saying that, what are you saying?

    Oh, and as for:
    Saying, “Let’s insist on looking at the gap as if the commitments we’ve made in the past don’t count anymore,” makes the process of assembling solutions harder, not easier.

    Please don’t put words in my mouth. I did not say that those commitments “don’t count”. First, to be clear, I’m not talking about defaulting on Trust Fund bonds. Second, as for promised benefits, reducing projected Social Security spending would not necessarily require reneging on past promises, but could change the benefit calculation and eligibility applied to SS FICA paid going forward. Third, if reducing projected Social Security spending were to involve reneging on commitments of promised benefit levels and eligibility to some extent, I’m not saying that in such a case, commitments “don’t count”, only that they are not inviolable in difficult circumstances such as a looming fiscal and economic crisis. Just to take an extreme for illustration, if the choice is between violating the promise of Social Security benefits to Bill Gates vs. substantially raising taxes on working class folks, I wouldn’t say that we should not even consider breaking that promise to Mr. Gates.

  19. comment number 19 by: johnchx


    It wasn’t my intention to put words in your mouth. But I don’t believe that I am: disregarding the existing commitment to expend the proceeds of the Social Security payroll tax only for Social Security benefits, and for no other purpose, is implicit in your choice of the unified budget deficit as the policy variable to target.

    If you would extend to me, and others, the courtesy of assuming that I can add, you might suspect that — unlike yourself — I do not accept the unified deficit as the proper or most useful measure of the fiscal “gap.” That is, we have a disagreement about what the policy-relevant variable is.

    Now, I have my reasons for thinking that the unified budget deficit is a highly misleading indicator of the state of the national finances, and for preferring other indicators of the scale of the “gap.” But I recognize that others might not be persuaded — and that if I want to persuade them, I have a responsibility to offer some reasoning supporting my point of view. Because it is a point of view, not “algebra.”

  20. comment number 20 by: Brooks


    What are your answers to my questions re: the “Defense Tax” illustration?

    1) Are you saying that, in that cutting Defense spending (in conjunction with the aforementioned tax shift) could not reduce the overall fiscal imbalance?

    2) Are you saying that, in my hypothetical in which a Republican Congress and president doubled the Defense budget for this year, and set up a dedicated tax to fully fund Defense spending at that level for the next 50 years (see my previous comment), you would oppose a subsequent Congress and president even considering reducing Defense spending (i.e., it should be “off the table”) as one way to reduce the overall fiscal imbalance (in conjunction with the aforementioned tax shifts), on the basis that Defense was fully “solvent”?

    As for the unified deficit vs. the on-budget deficit, they each tell us different things about our finances, and they both have value. But one thing we shouldn’t do is double-count, which would occur if we include Trust Fund balances in our debt figure (i.e, using $9.6 trillion as the starting debt figure) and add future unified deficits (gap between total revenues and total spending) to that debt figure. When I speak of our long-term fiscal imbalance, I’m speaking of the next several decades and I’m looking at projected revenues, projected spending (including Social Security), and projected deficits (yes, unified deficits), as well as the publicly-held debt as a percent of GDP (because it’s the publicly-held debt on which we we must pay interest expense, whereas any interest on Social Security Trust Fund bonds are really just like earmarks for part of future tax revenues — i.e., a minimum we must spend on Social Security in the future).

    And you won’t find any quarrel from me on the view that the unified deficit masks the degree of fiscal irresponsibility of current policies. It does.

    In any case, hopefully we can agree that, if tax revenues are kept at their current level as a percent of GDP and spending proceeds as projected over the next several decades, there will be huge, persistent and growing deficits as a percent of GDP, and unsustainable growth in our interest expense and publicly-held debt-to-GDP. Right? So, with that “fiscal gap” in mind, please answer questions #1 and #2 above.

    Thanks in advance.

  21. comment number 21 by: johnchx

    What are your answers to my questions re: the “Defense Tax” illustration?

    Well, my first answer is that when I appear as a witness, you may cross examine me…but until then, let’s try to converse with one another as peers.

    Now, can we please dispense with the fiction that I can’t add? Surely, if we define Social Security spending as an addition to the unified budget deficit, then we are agreed that a reduction to Social Security spending constitutes a reduction in the unified budget deficit. It is the implication that our disagreement arises from my not understanding this that I’m objecting to.

    Where we actually disagree is on the identification of “the” problem, the reason that the problem is significant, and on the “ground rules” within which the problem should be solved. I think I understand your position, and I even agree with some elements of it. (And, conveniently, the policies I’d advocate would turn out to fix most of what I imagine you would say is the problem.) But your position is a position, not self evident truth or an accounting identity. And it is a position with which it is possible to disagree.

    To unpack the disagreement, let me try to summarize what I understand to be your position: (a) the problem is the negative cash flow of the U.S. government as a whole (i.e. the unified budget deficit); (b) the problem is important because the unified budget deficit is projected to grow large over the next several decades; and (c) any policy change which improves the unified deficit should be “on the table”, i.e. “no preconditions”. (Have I got that more or less right?)

    My view is different: (a) the problem is difference between tax revenues earmarked for current services and the cost of current services; (b) the problem is important because is bespeaks a failure of our political system to align means (taxes) with ends (spending); and (c) politicians solve problems piecemeal, so proposals that would re-open “closed” or “solved” problems should generally be “off the table.”

    Now, I think there are arguments for both of these positions. (I happen to think mine is more persuasive, but that’s neither here nor there.) The point is that there actually are live differences of informed opinion about how the situation should be viewed and approached. And it’s more productive to address the live issues than to dwell on dead ones.

  22. comment number 22 by: Brooks

    your position is a position, not self evident truth or an accounting identity. And it is a position with which it is possible to disagree.

    I do have a position, but what I’ve discussed on this thread does not constitute any position – no advocacy of policy — whatsoever. I’ve said nothing on this thread of what we “should” do. I have simply pointed out a matter of very basic algebra, not opinion or preferences.

    We have a projected long-term overall fiscal gap (deficits resulting from the difference between total revenues and total spending) which, when added to the existing debt on which we incur interest expense (the publicly-held debt), is projected (based on continuation of current policies) to ultimately lead to interest expense as a percent of GDP that is so large that we can no longer pay it (i.e., service our debt), and we’d have to default (or perhaps inflate away our debt, but I’m fuzzy on that scenario – suffice to say I assume it would be awful in a different way). That — and all adverse consequences (impact on our standard of living / quality of life) on the path to that disaster point, either directly or resulting from efforts to correct our course — is the ultimate, big-picture problem. And every dollar of spending contributes exactly one dollar to the fiscal gap and thus every dollar contributes equally to the problem regardless of how it’s spent (net of dynamic effects, but that’s beside the point here). So even if Social Security were projected to be fully solvent, it would still be contributing to our long-term fiscal imbalance, just as would any other form of spending, and how much it contributes to this imbalance is equal to how much we spend on it (net of dynamic effects and factoring in timing). So, it is simply wrong for someone to suggest that, if Social Security were projected to be infinitely fully solvent, then Social Security is not contributing to the overall fiscal imbalance and that cutting Social Security spending could not be one way to reduce this overall fiscal imbalance. Similarly, it is simply wrong to measure the degree to which Social Security will contribute to our projected overall fiscal imbalance by the size of the Social Security shortfall — the gap between full solvency and projected revenues under current SS FICA tax rates and applicable income (which would be zero in the infinite full solvency scenario).

    Whether or not we should cut projected Social Security spending is an entirely different question, and one I haven’t addressed at all on this thread. And that matter is, of course, ultimately a matter of opinion and preferences. Like any fiscal policy choice, economics should inform us of what trade-offs relate to alternative policies, and we should then choose among policies – in effect choosing among those trade-offs — based on our values and priorities. And yes, we should also consider the behavior of politicians in assessing any differences between theoretical trade-offs and more likely trade-offs in the real world.

    As for your (a, b, c) view, as I’ve said previously, I agree that use of Social Security surpluses for current spending and common usage of the unified deficit in reporting of “the” deficit masks the true level of fiscal irresponsibility of current policy (and that of the last couple of decades). I don’t think we have a disagreement there.

    Lastly, I once again request that you answer my questions regarding the Defense Tax illustration. I don’t think it’s at all inappropriate for me to repeat a couple of very relevant, clear questions that you have not answered. It implies no assumption of superiority or authority over you.