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

EconomistMom.com

Oh Yeah, That Social Security Problem…

March 25th, 2010 . by economistmom

nytimes-social-security-deficit-032510

We’ve been so busy with health care reform and trying to figure out how to “bend the health cost curve” that we’ve almost forgotten about the other federal entitlement program that is on an unsustainable (albeit less severe) fiscal path.  An article in today’s New York Times reminds us of that other program, Social Security, with the newsworthy event being a Congressional Budget Office table that reveals the program is expected be in a cash deficit this year–with Social Security benefits paid out expected to exceed Social Security payroll taxes collected.  True, the Social Security program only faces one of the two pressures adversely affecting Medicare spending: just the demographic challenge of a rising elderly population relative to the working-age population.  And as a result the problem is not as large.  Some argue that’s a reason not to worry about it–and not to do anything about it until it’s more certain we’re right at the doorstep of that unsustainable fate.

Not doing anything about Social Security now would be fine if either: (i) we knew what to do with the much bigger challenge of  much more rapidly rising Medicare spending (i.e., we knew how to flatten that health cost curve and were really on the way to doing it), or (ii) we didn’t know what to do to close the much smaller Social Security deficit.  But the fact is that it’s hard to know how to solve the Medicare problem and relatively easy (mathematically and economically) to solve the Social Security problem.  We know how to solve the latter, and I think most Americans in hearing what some of these solutions are (see this NYTimes editorial blog featuring some experts’ ideas), would think they were no big deal.  Why, just this week my (very socially-conscious) 17-year old daughter, Emily, brought up the unsustainable fiscal outlook and said she didn’t understand why we don’t just raise the retirement age–which is what both Bill Gale (of the liberal-leaning Brookings Institution) and Andrew Biggs (of the conservative-leaning American Enterprise Institute… ok, perhaps more than leaning, but more on the David Frum story later) recommend in the NYTimes piece.  Emily spoke of raising the retirement age as a “no brainer”–and not because I’ve brain-washed her but perhaps because she knows her mom will be working forever anyway just to pay for her and her three siblings’ college educations…  ;)

Of course no one likes to work longer, just like no one likes to pay higher taxes or have limits on their subsidized health benefits.  But let’s face it: there will have to be more of these things we don’t like, and to me it’s a simple cost-benefit analysis.  The relative benefit of doing something we know how to do sooner (reducing the Social Security deficit) seems high in that small, phased-in changes now can save a lot of compound interest going forward.  Even if it saves us more money than the size of the (small) Social Security “problem,” that’s a good thing, because chances are really good that our other reforms to the other programs will save us less money than the size of those other (bigger) problems.  And any pain associated with these Social Security fixes–such as a gradual rise in the retirement age–seems relatively low and “easy” compared with those other (currently largely uncertain) ways of solving the long-term fiscal challenge which we’ll still be trying to figure out for many, many years to come, and which will have to result in much more fundamental (and ultimately painful) changes to those other programs.

So yeah, we still have this little problem of Social Security, but it’s completely and pretty easily solvable–and so we should be paying attention to it, now.

45 Responses to “Oh Yeah, That Social Security Problem…”

  1. comment number 1 by: SteveinCH

    Diane,

    I don’t understand the “raise the retirement age” solution relative to the means testing solution.

    Raising the retirement age is both inefficient and ineffective. It’s inefficient because some older folks now not covered by SS will take other government programs instead, meaning that more people will need to be dispossessed of SS per dollar of savings.

    It’s ineffective because it hurts people who arguably have need simply because of their age.

    The argument I always hear on the other side is that universality is critical to sustaining the program but most Federal programs, including those that have been around quite a long time are not universal.

    I’m puzzled to be honest.

  2. comment number 2 by: Brooks

    the other federal entitlement program that is on an unsustainable (albeit less severe) fiscal path.

    Conceptually, it makes no sense to speak of Social Security being “on an unsustainable path” as if that were some actual problem simply because of a gap between currently projected revenues from the dedicated tax and currently projected SS spending. They are just part of the whole — overall revenues and overall spending. There is no Social Security money tree (nor any real “trust fund”). After all, is Defense “on an unsustainable path” because we have projected Defense spending and zero projected dedicated revenues? We certainly wouldn’t say that because it’s obviously meaningless.

    The decision we face is how much we want to spend on Social Security, period. Currently projected dedicated revenues (FICA SS + “repayment” of the “trust fund” balances) are irrelevant as anything but a bookkeeping matter unless we choose to make it a substantive matter. If the amount we wish to spend over time exceeds the dedicated revenues, we can raise that dedicated taxation or supplement it with general fund revenues (it’s not carved in stone that we must fund SS — or for that matter Medicare — through payroll taxes alone if that’s not best from an economic and/or “fairness” perspective). And we can choose however much we want those revenues to be incremental (higher taxation) vs. “funded” via lower spending on something else, just as would be the case for anything else we wished to fund — the existence of a dedicated tax for a program (and any related “gap” vs. projected spending) has no meaningful bearing on any of these decisions. It only matters because there is so much conceptual confusion that the politics are heavily affected by all this nonsensical talk of the significance of the degree or duration of Social Security “solvency” and the size of projected “gaps”, etc.

    So yeah, I think should look at policy changes to reduce projected spending on Social Security (and Medicare), but that’s because of projected OVERALL deficits and my preferences as to how to reduce them. It has absolutely nothing to do with the amount of projected SS spending relative to projected dedicates SS revenues, which is essentially nothing but insignificant bookkeeping. After all, we could eliminate any “gap” in SS “funding” without any net tax increase or reduction in SS spending at all if we just increased FICA SS taxation enough to fund projected spending and offset that tax increase with a cut in other taxes producing revenue-neutrality, and — presto! — “gap” eliminated! But of course, projected OVERALL (unified budget) deficits haven’t changed one bit; we would have just reduced the SS “gap” and increased on-budget deficits, both of which result from bookkeeping rather than reflecting any substantive change in the nation’s finances.

  3. comment number 3 by: AMTbuff

    I can’t see why anyone would be interested in the SS shortfall when we are obviously rich enough to have just bought into a huge new entitlement using borrowed money.

    The fiscal cliff is approaching faster now than ever, and we’ll be going over it soon enough no matter what we do, so why not just enjoy the ride until then?

    It’ll be far easier to fix these programs after the crash, when meat axe cutbacks will finally be politically possible. It’s a waste of effort to attempt to fix anything before then.

  4. comment number 4 by: John Bailey

    We need to deal with Social Security, Federal civilian and military retirement, state and local pension funds, the Pension Benefit Guaranty Corporation, Medicare, Medicaid, state and Federal retiree health benefits, private health insurance, the tax code, Federal, state and local spending, unemployment, underwater mortgages, Fannie, Freddie, et.al., bank failures, the FDIC, and the Federal debt.

    I think that the first step is to agree that the people who have to bear the burdens make the choices. Tax, spending, and debt levels should be set directly by the citizens/taxpayers. They should also make the decisions on major policy questions. Everyone will lose something. Everyone also needs to have the opportunity to participate in the decisions that impact their lives.

  5. comment number 5 by: rjs

    at 62, i agree with emily, raise the age…there’s no reason someone my age should be on the dole…

    & does the “cash flow” graph include interest currently being erned on prior years surpluses now invested in Treasuries?

  6. comment number 6 by: dWj

    I have seen in polls that cutting social security benefits is more popular (viz. less unpopular) than raising the age. As long as we extend people’s ability to choose to retire sooner than par for a larger monthly sum or later than par for a smaller monthly sum, the two are, of course, equivalent, except for framing; framing should be done in whatever way is most politically expedient.

  7. comment number 7 by: economistmom

    rjs: No, the cash or “primary” deficit or surplus in Social Security doesn’t include any interest earned on trust fund balances; it just compares annual Social Security benefits to the taxes dedicated to Social Security.

  8. comment number 8 by: Jim

    The last sentence of the second paragraph states “Emily spoke of raising the retirement age as a “no brainer”–and not because I’ve brain-washed her but perhaps because she knows her mom will be working forever anyway just to pay for her and her three siblings’ college educations.”

    I have never understood why parents are responsible to pay for a college education. College students consider themselves grown up and mature, so why should they have to pay their own bills? There are many ways that they could pay for their education, get scholarships, get a job, join the Army.

    I suppose I am bias. My brother and I paid for our college educations. My mom paid for hers while she was working a full time job and taking care of my brother and me.

  9. comment number 9 by: BillSmith

    Is there someplace to look where one could plug in some numbers and see the effect of a change in the retirement age? Say for people currently under 55, add 1 month to their retirement age for each year under 55? So 35 years olds would retire 20 months later than they currently could now?

  10. comment number 10 by: Curly

    Oh Yeah, That Social Security Problem… And with the 5 billion that then new health care law will take out of Social Security it is even shorter than stated here.

  11. comment number 11 by: Hoppy

    And what jobs are people going to work at longer than they do now? Unless they’re planning to relocate to Shanghai or New Delhi, where should they apply?

  12. comment number 12 by: B Davis

    SteveinCH wrote:

    The argument I always hear on the other side is that universality is critical to sustaining the program but most Federal programs, including those that have been around quite a long time are not universal.

    Social Security is something of a combination of a mandatory retirement/annuity program and a safety net. As you may know, up to 85% of benefits are taxed depending on the beneficiary’s income. In addition, low-wage earners get a higher return on their contributions than do high-wage earners. You can see a description of the income-based formula at this link. One could certainly argue that the retirement and safety-net portions of Social Security should be split into their own programs. As you suggested, a worthwhile safety-net should be able to stand its own. Still, Social Security is what it is and its basic structure is unlikely to change anytime soon.

    Hence, even if benefits for high-income earners are pared back, all same-aged retirees will likely continue to receive benefits starting at the same age. Given that life-expectancy is increasing, it does seem like a no-brainer that people work for a portion of their additional healthy years. If some medical advance could increase my life-expectancy by ten years, I’d be more than happy to work for five of them.

    Of course, people should have the freedom to retire earlier if they are willing to accept lower benefits. One little known fact is that, despite the fact that the age of full retirement is increasing from 65 (for those born in 1937 or earlier) to 67 (for those born in 1960 or later), the early retirement age is remaining at 62. According to this link, a person retiring at 62 will receive a smaller and smaller percentage of the normal retirement benefits as the normal retirement age increases. The reduction will increase from 20% for those born in or before 1937 to 30% for those born in or after 1960.

    I suspect that a proposal to increase the normal age of retirement would receive less opposition if it were made clear that early retirement would remain an option. Proponents of such a proposal would do well to also specify what the change will be to the early retirement age. In addition, they can make clear that the program will be designed such that the lowest level of benefits received will be sufficient to continue providing a minimal safety-net.

  13. comment number 13 by: Brooks

    B Davis,

    Fair point that progressive taxation of benefits is a form of means testing, but don’t forget that means testing arguably should be based in part (or even mostly) on wealth rather than current income.

  14. comment number 14 by: SteveinCH

    BDavis,

    Thanks. It remains my opinion that providing benefits and taxing them back is a very inefficient approach to passing out money.

  15. comment number 15 by: B Davis

    Brooks wrote:

    Fair point that progressive taxation of benefits is a form of means testing, but don’t forget that means testing arguably should be based in part (or even mostly) on wealth rather than current income.

    Agreed. However, it may be difficult to ascertain someone’s wealth. I believe that the main cases in which it is currently ascertained is for estate taxes and for taxes on certain property whose existence would be difficult to hide (such as houses and cars). That may be much of the reason why it seems difficult to get good data on the distribution of wealth. The best source that I’ve come across is the the Survey of Consumer Finances that is put out by the Federal Reserve every three years or so. I’ve posted some data on the 2004 Survey at this link.

    Speaking of estate taxes, I hope that Congress can address that issue now the Health Care Reform bill has passed. As I’m sure you know, the estate tax was repealed in 2010 but is currently scheduled to be reinstated in 2011 with the top rate of 55% that existed in 2001. That’s why Paul Krugman referred to this part of the 2001 Bush tax cut as the “Throw Momma From the Train Act of 2001″. The reasonable thing to do would have been to continue the 2009 rates (top rate of 45% with an exemption of $3.5 million) until they could settle on a permanent solution. The longer they delay, the more likely that people will believe that the repeal will stand (with no retroactive reinstatement) and the more likely that some Momma truly may be “thrown from the train”. They need to address this issue as soon as possible.

  16. comment number 16 by: BillSmith

    Brooks

    Wealth tax, eh? How much is my house worth?

  17. comment number 17 by: Brooks

    B Davis and BillSmith,

    I realize there are difficulties and potential practical problems with factoring in wealth for means testing (measuring it; potentially adverse incentives regarding saving; etc.). I’m just pointing out the theoretical argument for it and suggesting the possibility of actually including it. I don’t know enough about the details of implementation and effects to say whether or not it would work as intended and be desirable on balance, but I do consider it plausible.

  18. comment number 18 by: Brooks

    Oops, I forgot that more than two links causes flag for moderation, so I’ll re-post.

    FYI All,

    At http://krugman.blogs.nytimes.com/2010/03/27/dealing-with-the-debt-a-brief-note/ Krugman persists with his sharp disagreement with the Krugman of 2003, saying with regard to our long-term fiscal outlook that “there are real worries — but the math per se isn’t very hard…The numbers aren’t that bad”.

    For elaboration on the 2003 Krugman vs. the 2009/2010 Krugman (if anyone here hasn’t already seen this discussion), see
    http://economistmom.com/2010/02/why-paygo-isnt-enough-but-why-well-take-it/#comment-6741 and [see next comment for last link].

  19. comment number 19 by: Brooks

    [continued from last comment]
    Here’s that last link http://economistmom.com/2010/02/why-paygo-isnt-enough-but-why-well-take-it/#comment-6807

  20. comment number 20 by: Brooks

    I was surprised to see the paragraph below on PolitiFact in assessing Boehner’s claim that “our national debt ($12.7 trillion today) is on track to exceed the size of our entire economy (about $15 trillion) in just two more years”. From Politico:

    Some economists prefer to use public debt rather than gross federal debt, but one measure “isn’t more ‘right’ than the other – they are just looking at different things,” said Marc Goldwein, policy director for the Committee for a Responsible Federal Budget, a middle-of-the-road budget-hawk group. “Boehner may be cherry-picking, but I don’t think he’s misrepresenting in any way.”

    Either the above is misleading editing or it’s surprising to me that someone with CRFB would make such a statement. Trust fund “debt” is somewhere between having little substantive meaning and being completely meaningless, so what matters is debt held by the public, not gross debt.

    Gross debt is almost not meaningful at all as a measure of our nation’s current and projected finances, financial condition and all related problems and risks. Only debt held by the public affects our federal government’s expenditures to service debt, our credit-worthiness and interest rates and also potential monetization of our debt (the Fed buying Treasuries to finance our debt by “printing money”) to a degree that would cause harmful levels of inflation. The “trust fund” balances (which are added to that publicly-held debt to sum to the gross debt) at most represent a minimum we must eventually spend on programs related to those “trust funds”, and those figures are, for practical purposes, meaningless. To illustrate, the largest balances are the total of $2.5 trillion in the Social Security “trust funds”. All that means, at most, is that we must eventually spend $2.5 trillion on Social Security benefits. But no one thinks we would spend anywhere near that little cumulatively in coming decades on Social Security benefits, so it is, in effect, a meaningless minimum spending amount. It’s like if we had some obligation to spend at least $10 billion on Defense over the next ten years — it means nothing because its inconceivable that we would spend anywhere near that little.

  21. comment number 21 by: Brooks

    The Politico link http://politifact.com/truth-o-meter/statements/2010/mar/26/john-boehner/boehner-says-federal-debt-will-equal-gdp-two-years/

  22. comment number 22 by: SteveinCH

    I’m no longer so sure Brooks. As you point out, a good chunk of the nonpublic debt is the SS trust fund I think. Clearly that’s going to get converted to public debt at some point in the next couple of decades unless we suddenly start running an on-budget surplus (not something I’m holding my breath for). If we see the glide path, I’m not sure why it shouldn’t count the same since it’s going to happen.

    I understand the accounting treatment but I think the whole discussion of debt is disingenuous since we should be looking at the actuarial picture rather than the single year picture. By that light, our debt is understated, not overstated.

  23. comment number 23 by: Brooks

    Steve,

    [darn, I did it again -- included 3 links. I'll split up the comment here]

    Steve,

    Again, if the SS trust fund balances represent a minimum that we must eventually spend on SS*, and if that minimum is way below any conceivable amount we will be spending anyway, then it’s a minimum without any significance. Again, it’s like if we had some obligation to spend a minimum of $10 billion on Defense over the next ten years, an amount that is far below any conceivable amount that we will spend anyway. It’s meaningless.

    If it helps, here are a couple of comments with elaboration. The first one has my “Joe the father” illustration. http://economistmom.com/2009/03/another-debt-documentary-ten-trillion-and-counting/#comment-2688
    http://economistmom.com/2009/03/another-debt-documentary-ten-trillion-and-counting/

    * Jim Glass has argued, somewhat plausibly in my view, that it doesn’t even represent that much (that obligation).

  24. comment number 24 by: Brooks

    Steve,

    Here’s a different explanation http://www.pbs.org/wgbh/pages/frontline/tentrillion/etc/ednote.html (the guy from OMB is right; Frontline’s response doesn’t refute or detract from the validity of his point).

  25. comment number 25 by: Jim Glass

    Jim Glass has argued, somewhat plausibly in my view, that it doesn’t even represent that much (that [SS Trust fund] obligation).

    Here’s what I say, plausibly or not…

    There are two meaningful definitions of the “national debt” reported by the Treasury. They are both important, but clearly different, and so shouldn’t be confused or conflated.

    1) The “debt held by the public”. This incurs the cash cost — which means tax cost — of debt service (interest). It is guaranteed by the Constitution, so there is no getting out of it. Today just over $8 trillion.

    This is of first, prime, current importance. The degree to which the govt can afford it determines the govt’s credit rating, the interest rate it must pay, how much in taxes the govt must collect to service it (and thus the associated deadweight cost of taxes on the economy), etc.

    2) “Accrued net liabilities”. These are legally accrued obligations to make payments in the future, in excess of legally accrued projected future revenue. The big items here are Medicare, Social Security, and federal/military pensions, with net accrued liabilities totalling $51.2 trillion at present value at the end of 2009 (up $3 trillion from 2008 — add that to the officially reported deficit to get the real deficit for 2009, by the accounting rules the private sector uses).

    These items incur no current cash cost, thus no tax cost, and they aren’t guaranteed by the Constitution. Congress can reduce/eliminate them at any time by changing the law.

    (E.g., Congress can reduce SS benefits so it never need pay on the SS trust fund bonds. It could even simply cancel the bonds with no default, because the bonds are issued by the govt to itself. Since nobody else holds them, who could sue for default? Nobody. One can’t default on a debt to oneself. If you doubt this, try to do it and see!)

    Of course, these items do roll over into the debt held by the public over time, so their amount gives us our picture of the future state of the debt held by the public on current law.

    E.g., with Boomer entitlements beginning to roll over in a big way in the 2020s, S&P has projected the credit rating of the US as falling from AAA to “junk” over the period 2017-2027 on current (actually pre-2008) policy.

    This shows us how much we have to change policy to keep future debt held by the public at a level that will preserve the US govt’s credit rating. Which is important and meaningful information.

    Now, as to the very commonly cited “gross debt”, or “public debt” (as distinct from “debt held by the public”) it entirely illogically includes all the debt held by the public plus about $4 trillion of accrued liabilities — or about 8% of their total — only because intra-governmental bonds have been issued for them.

    But this $4 trillion is indistinguishable from the rest of the $51 trillion — it just represents promises Congress can always change, which carry no current tax cost, exactly like the other $47 trillion of it — while it is completely different from the $8 trillion of debt held by the public, which is Constitutionally guaranteed and incurring a tax carrying cost now.

    Ergo, the figure of “gross debt, $12 trillion” is neither fish nor fowl and is grossly misleading for any given purpose: It overstates the real, Constitutionally-guaranteed, tax-cost incurring debt by 50% … and understates the total spending promises the govt has made, and that people are counting on, by 80%!

    So it should be ignored, it is worse than worthless, it is misleading-to-deceiving.

    An recent example of this: Over a McArlde’s a couple days ago a commenter was strongly arguing, “so what that $2.5 trillion of SS bonds will be rolled over in the debt held by the public. It makes no difference! None at all! The official public debt of the US, the gross debt, will be totally unchagned, to the penny! So it can’t possibly hurt the national finances, because that debt is already incurred! Other European nations that haven’t saved with such intra-governmental debt are going to have to increase their total national debt with new debt to pay their retirees, so they are going to be hurt, not us! ”

    Hah. The SS actuaries project a 6% interest rate on US debt, long-term. And 6% of $2.5 trillion is $150 billion in taxes annually. It would take about a 15% across-the-board income tax increase to pay that interest. If you think an extra $150 billion of tax cost annually is “no difference”, well…

    That was an intelligent person who had been misled by the “gross debt”.

    Examples of those doing the misleading are defenders of entitlements who say “look, it is only ‘debt held by the public’ that really counts”, because they don’t want to deal with the cost of accrued entitlements at all — not even the mere 8% of it included in the gross debt.

    And yes, this surely includes Krugman, who used to say the cost of entitlements was causing the “looming threat to the solvency of the federal government”, and now just keeps saying, “Look how small the debt held by the public is. (Entitlements? What entitlements? Oh, those things, we’ll just reduce them later)”. BTW, the man doesn’t stop — he just did it all over again!

    So, my advice, use measures #1 and #2 for their appropriate purposes.

    Forget #3, it is bogus and deceptive. Remember, the Social Security trust fund and all the bonds in it were created by accident — there was no intention to create any such thing in the SS law changes of 1983.

    If that accident hadn’t occurred, measure #3 wouldn’t exist. So forget it, it shouldn’t exist.

  26. comment number 26 by: Brooks

    Jim,

    Seems we agree that gross debt is a meaningless and misleading metric.

    As I’ve explained to you before (starting at http://economistmom.com/2009/10/how-true-fiscal-conservatives-talk-about-tax-policy/#comment-4066 ), you are wrong in the significance you attach to “unfunded liabilities”. What matters is the gap between projected total revenues and projected total spending, not the gap between some arbitrary chunks of each simply because some are entitlements that have dedicated taxation. Medicaid is an entitlement without dedicated taxation, so projected spending on it is excluded from “unfunded liabilities”. Why? What makes projected spending on Medicare — let alone the gap between the arbitrarily selected category of Medicare spending less the arbitrarily selected categordy of dedicated Medicare tax revenues — worth including in some measure of our fiscal outlook but Medicaid not? Ditto for Defense spending and other discretionary spending, and for all other tax revenues? Why focus on just a couple of arbitrarily selected programs, and then double down on the arbitrariness by focusing on the gap between spending on these programs and their dedicated tax revenues?

    I’ll paste below what I recently wrote on another blog:

    The “unfunded liabilities” figure is not really analytically sound as a measure of our long-term fiscal imbalance or even as a good indicator thereof or of changes in this imbalance, because it is a function of an arbitrary carve-out of a portion of revenues from total revenues and of a portion of spending from total spending. It’s no more a metric of our long-term fiscal imbalance than the present value of the gap between any other pairing of projections for any one type of revenue and one type of spending — say, between Defense spending and corporate income tax revenue, or any other chosen pair. What matters is the imbalance between TOTAL revenues and TOTAL spending, and if one wishes, the present value thereof, not some arbitrary breakout of components (plus the debt held by the public).

    There is a conceptual fiction that Social Security and Medicare are somehow truly islands apart from the overall fiscal picture, as if taxation to fund them comes from an entirely different source than the source that is taxed to fund all other federal spending. In reality, we could always supplement the dedicated tax revenues for those programs with tax revenues that go into the general fund or more directly, for a given level of revenues, just shift taxation in a revenue-neutral way (i.e., increase payroll taxation and offset it by reducing other tax rates). Similarly, even if Social Security or Medicare were fully “solvent” forever, we could still reduce our overall fiscal imbalance by spending less on them, reducing the dedicated taxation for those programs, and offsetting those tax cuts with increases in other taxes, resulting in in lower projected spending, no change in projected revenues, and thus lower projected deficits. It’s all ultimately fungible, so these break-outs are ultimately arbitrary and not particularly meaningful.

    What the term “unfunded liabilities” refers to is the amount of incremental dedicated tax revenues for the chosen programs that we would need to:

    (1) avoid reductions in projected spending on those programs,
    AND
    (2) avoid an increase in projected deficits between all other revenues and all other spending (while also avoiding higher tax revenues from other taxes and reductions in other spending).

    Well, how meaningful is that? Not very. It’s like saying:
    1. (A-B) + (C-D) = E
    2. A-B is huge.

    My concern is E, so what does it tell me for someone to point to A-B or to changes in A-B as an indicator of the magnitude of E? It’s only meaningful if they also tell me something about C-D, in which case, what’s the point of focusing only on A-B in the first place?

    Well, the A-B in the algebra above is this metric of “unfunded liabilities”. It’s the gap between projected dedicated tax revenues associated with those particular programs (plus the “balances” of “trust funds” for those programs, which are their own fiction and source of conceptual misunderstanding) less the projected spending on those programs. C-D represents the gap between all other revenues and all other spending. And E represents total deficits, which are what we care about, since again, everything is fungible going forward. Ultimately we can choose how much we want to spend on what and how much to tax via which taxes.

    To illustrate this point, suppose we did away with dedicated taxation altogether, and simply funded Medicare and Social Security from the general fund henceforth. Well, I suppose one who sees the “unfunded liabilities” metric as very analytically meaningful would have a dilemma on his hands. He could either now view our “unfunded liabilities” for those programs as much, much higher, since now there are zero dollars of dedicated taxation associated with those programs, so ALL projected spending on those programs would be considered “unfunded liabilities”, OR he could now see these programs the same way he sees other projected spending that is not on programs that have an associated dedicated tax, such as Defense, and say that our “unfunded liabilities” have disappeared completely! But regardless of which he chooses, has our projected overall long-term fiscal imbalance changed at all? Of course not. But one would sure think so if he considers the size of our “unfunded liabilities” a very useful and analytically sound metric for the the size of our long-term fiscal imbalance.

    We have an existing debt held by the public. And we have projected (total) deficits. If one wishes, one can take the present value of those deficits and add them to the current debt, which would make more sense than just slicing out the gap between particular pairs of taxes and spending on particular programs. But, since we really want to put this in the context of some measure of our ability to pay, it’s probably misleading to express such a figure relative to today’s GDP, and better to express the current debt relative to today’s GDP and to do the same for projections.

    As just a note, I think ultimately an even better measure would include the context of our public and private net worth as well as income (GDP), since both net worth and income measure a creditor’s ability to service debt.

    [end paste] ———————————————-

    I would also note that the alternative to honoring our debt held by the public is default (another way out, in effect, could theoretically be wild monetization, and even that might not really be a viable means of avoiding default, not to mention all the other harmful effects). By contrast, the alternative to “honoring” the “unfunded liabilities” is simply to change fiscal policy to spend less on those programs, which at most would represent broken political promises, not anything that would make it impossible or much more expensive for the federal government to borrow money (as would default or heavy monetization, or a perceived increase in the threat of either).

    Jim — if you still don’t see my point and you don’t want to get into it again, no problem. I explained throughout a long exchange previously and you didn’t get my point (or at least wouldn’t acknowledge its validity), so I’m fine with not going over it again.

  27. comment number 27 by: SteveinCH

    Brooks,

    I don’t know if I got involved in the last foray but I’m with Jim on this one. I completely agree that projected total deficits taken at present value is a better solution (if such a number actually existed). The problem is that there is no so number really.

    If you look at the OMBs or CBOs long term projections, they basically assume flat discretionary spending. Discretionary spending (particularly nondefense discretionary) hasn’t been flat in a long time. So, said differently, the question is what you assume for discretionary spending.

    So a current law projection has to make an assumption on discretionary spending. That assumption has a huge impact on the number you get. By contrast, the deficits computed on SS and Medicare do not suffer from this same problem as both the revenues and expenses are fixed.

    Yes, it’s a less complete picture but a more accurate one.

  28. comment number 28 by: Brooks

    Steve,

    Projected SS and Medicare spending are not fixed. They can be changed as a matter of policy (eligibility; benefit levels), or could change based on other factors (most notably the relevant medical inflation rate, which can be affected somewhat by policy changes), just as we can change policies related to the projected level of discretionary spending.

    Moreover, you haven’t said anything that diminishes my argument that the “unfunded liabilities” metric has little/no validity as a measure of our long-term fiscal imbalance. I don’t know if you read through all of my comment, but I think I demonstrate my point pretty clearly conceptually, algebraically, and via illustration. Perhaps the illustrations make it clearest for most people, so I’ll paste those paragraphs below:

    1. What matters is the gap between projected total revenues and projected total spending, not the gap between some arbitrary chunks of each simply because some are entitlements that have dedicated taxation. Medicaid is an entitlement without dedicated taxation, so projected spending on it is excluded from “unfunded liabilities”. Why? What makes projected spending on Medicare — let alone the gap between the arbitrarily selected category of Medicare spending less the arbitrarily selected category of dedicated Medicare tax revenues — worth including in some measure of our fiscal outlook but Medicaid not? Ditto for Defense spending and other discretionary spending, and for all other tax revenues? Why focus on just a couple of arbitrarily selected programs, and then double down on the arbitrariness by focusing on the gap between spending on these programs and their dedicated tax revenues?

    2. To illustrate this point, suppose we did away with dedicated taxation altogether, and simply funded Medicare and Social Security from the general fund henceforth. Well, I suppose one who sees the “unfunded liabilities” metric as very analytically meaningful would have a dilemma on his hands. He could either now view our “unfunded liabilities” for those programs as much, much higher, since now there are zero dollars of dedicated taxation associated with those programs, so ALL projected spending on those programs would be considered “unfunded liabilities”, OR he could now see these programs the same way he sees other projected spending that is not on programs that have an associated dedicated tax, such as Defense, and say that our “unfunded liabilities” have disappeared completely! But regardless of which he chooses, has our projected overall long-term fiscal imbalance changed at all? Of course not. But one would sure think so if he considers the size of our “unfunded liabilities” a very useful and analytically sound metric for the the size of our long-term fiscal imbalance.

    If you think you can show me why the above don’t illustrate pretty clearly why “unfunded liabilities” is an artificial, arbitrary, invalid metric for measuring our long-term fiscal imbalance, go ahead.

    It simply makes no sense to arbitrarily cherry-pick “gaps” in “funding” for programs that happen to be funded via dedicated taxes and add those “gaps” to our debt held by the public (with or without also adding trust fund balances) and call that some measure of our long-term fiscal imbalance. As I explained algebraically:
    If E is the long-term fiscal imbalance — what I really care about.
    And A-B represents “unfunded liabilities” (the “gap” between projected dedicated revenues for particular programs ["A"] and projected spending on those programs ["B"].
    Then E = (A-B) + (C-D), with C-D being the “gap” between all other revenues (other than those dedicated revenues for those particular programs) and all other spending (other than spending on those programs).
    And the only way A-B (i.e., “unfunded liabilities”) tells me anything about E is if I also know C-D, in which case, why bother with A-B as if that tells me something about E on its own?

  29. comment number 29 by: Brooks

    Steve,

    Here’s another illustration, in case it helps:

    Frank projects he’ll make $X each month. But he has arranged for his employer to put part of his monthly paycheck, $A (which is a subset of $X), into a checking account that is dedicated to paying his rent. His projected rent amount is $B. The part of his paycheck that the employer will not deposit into that checking account, the employer will pay in cash to Frank. That amount, which is X-A, will be $C. And all Frank’s other projected spending is $D.

    Now, we want to have a measure of Frank’s projected total surplus or deficit, which we’ll call $E.

    You tell me that you have a good metric for E, and your metric is A-B, i.e., the gap between the amount that is projected to be deposited in that checking account and the amount of rent. It so happens that the projected rent amount is higher, so there is a Rent Program deficit – i.e., an “unfunded liability”, and you calculate that “unfunded liability” and present it as a measure of the magnitude of Frank’s projected fiscal imbalance (”E”).

    I ask “But what does that tell me about the size of E, which is what I care about?”

    You answer “Well, it tells us how much E would be if C-D = 0, i.e., if all other (non-Rent Program) revenues equal all other (non-Rent Program) spending, and it tells us the minimum E can be if C is less than D (i.e., if there is a deficit between all other revenues and all other spending).”

    To which I reply, “Then why bother with A-B as a metric. If I have to know C-D in order for A-B to have any meaning as a metric for E, why bother breaking it out. It’s as meaningless as if I used A-D or C-B or any other arbitrary pairing of a particular category or subset of revenue vs. a particular category or subset of spending. It’s just part of the whole (total revenues and total spending) and tells me nothing about the whole unless I also know the balance for everything else, so it really tells me nothing about E.”

  30. comment number 30 by: Brooks

    Steve,

    Yet another way to see why “unfunded liabilities” is not really a valid metric for our long-term fiscal imbalance:

    Suppose we shifted tax rates around in a revenue-neutral way, increasing payroll taxes enough to completely eliminate the “gap” in “funding” for SS, Medicare, and whatever else is lumped into “unfunded liabilities, and reducing other taxes (again, with no net effect on total revenues). Presto! There would no longer be ANY “unfunded liabilities”. ZERO “unfunded liabilities”. But would the size of our overall fiscal imbalance be any different? Of course not. Doesn’t that demonstrate how arbitrary and thus insignificant the “unfunded liabilities” metric is as a measure of our overall fiscal imbalance?

  31. comment number 31 by: SteveinCH

    Brooks,

    I accept every comment above but my question is how do you measure your preferred metric. If it cannot be measured accurately, it’s not worth much.

    My point is, I prefer an accurate but limited metric (number) to a metric that can be manipulated via assumptions in a very obvious way.

    I suppose another way to look at it would be to calculate the “remaining funding” level of the government. That is, take all receipts and just entitlement spending and express the difference as spending per year (nominal or percent of GDP). That probably solves for a holistic view although it’s not very intuitive as a measurement.

  32. comment number 32 by: Brooks

    Steve,

    how do you measure your preferred metric

    I assume that present value of future deficits would be calculated the same way that present value of “unfunded” liabilities are calculated. The difference is that the former is looking at what matters — the whole — and the latter just arbitrarily selects a subset of programs and on a meaningless basis counts the “gap” between a subset of revenues (the dedicated revenues of those arbitrarily selected programs) and the spending on those programs. The latter is not a “limited” metric as a measurement of our overall fiscal imbalance; it’s an invalid metric. It says nothing unless we also know or make an assumption about the difference between all other revenues and all other spending, in which case, why bother with the metric at all. The best that can be said of the “unfunded liabilities” metric is, as I mentioned, that if we can safely assume that all other spending (everything excluded by that metric) is “underfunded” by all other revenues, then “unfunded liabilities” represents the minimum that our projected overall imbalance can be, just as, if we know that all Frank’s income not going into his rent-dedicated checking account is less than all his non-rent expenses, we can consider his “unfunded rent liability” to be the minimum that his overall underfunding can be. But if we can calculate that “all other” underfunding, we may as well just measure the whole, and if for some reason we couldn’t measure the “all other” but somehow know it’s underfunded, we still don’t know the magnitude of overall fiscal imbalance (underfunding) and can’t measure changes in it over time by measuring changes in the magnitude of “unfunded liabilities” unless we make a further assumption that the magnitude of the “all other” imbalance hasn’t changed — which would mean that we are able to measure that “all other” underfunding after all.

    Re: your last paragraph, I don’t see what the point would be of calculating “remaining funding” after projected spending on entitlements, unless one wishes to say “If we don’t cut projected entitlement spending and if we want balanced budgets, we have to limit discretionary spending to $X”.

  33. comment number 33 by: SteveinCH

    Brooks,

    My point is that there is no way to measure your preferred metric without making an assumption about the level of discretionary spending. Said assumption will move the analysis substantially given the timeline in question. I know we can make the assumption but given how sensitive the analysis would be to the assumption, it is easy to game the analysis by gaming the assumptions. While this already happens to a degree on macroeconomic assumptions, there is at least a body of independent estimations on which one can fall back.

    The point therefore of the metric I suggested is that it does not rely on an assumption about future levels of discretionary spending to provide a metric but uses a breakeven analysis.

    At least when I’ve done it before in a business context, running the breakeven is often a better analytical approach than making a very debatable assumption.

  34. comment number 34 by: Brooks

    Steve,

    If we make no assumption regarding discretionary spending, the “unfunded liabilities” metric means absolutely nothing as a measure of our overall fiscal imbalance. Again, it’s like if Frank me the dollar amount of his “unfunded rent liability” (the difference between the amount of his income he projects he will carve out as “dedicated” to his rent vs. the projected spending on rent) and someone asked me what the size of his overall fiscal imbalance is. I’d have to say that I have no idea, because I know nothing about the the dollar amount of the rest of Frank’s income and the dollar amount of all Frank’s other spending. Knowing the difference between a particular, arbitrarily selected category of his spending (rent) vs. the arbitrary portion of his income that he chooses to “dedicate” to that category of spending (rent) tells me nothing about his projected overall balance or imbalance. All I know is that Frank cherry picked his “Rent Program” for some reason and then for some other reason chose to calculate the gap between his spending on rent and the arbitrary amount he carves out of his overall income to “dedicate” to rent. It’s all completely arbitrary. He could just “dedicate” a larger portion of his income to rent — enough to eliminate the “funding” gap for his rent — and then that “unfunded liability” would disappear completely, but his overall fiscal imbalance would not change one penny (he would just have an equal size reduction in surplus or increase in deficit for everything other than rent), so what can that “unfunded rent liability” figure possibly tell me about his overall fiscal imbalance if it can come out to a large amount, zero, or anywhere in between simply by shuffling around the same amount of revenue?

  35. comment number 35 by: SteveinCH

    Right Brooks but any assumption we make is just an assumption and the output is extremely sensitive to the assumption. Without discounting, a point of GDP in discretionary spending (well within the range of error) adds probably on the order of $30 trillion over a 75 year forecast. Even with discounting, it’s a $15 trillion swing.

    That’s why I’m advocating doing the breakeven analysis. At least then you can say.

    1. We have X in revenues according to current law

    2. We have Y in committed outlays (mandatory programs).

    3. We have a countercyclical spending assumption of Z (let’s say 3 to 5% of GDP for 2 years, once a decade).

    Now if we want the whole thing to balance, we can only spend this much as a percent of GDP on all other.

    Now granted you can relax the first 2 assumptions but only by changing the law. Assumption 3 is debatable but probably realistic.

    At least to me, this is preferable analytically to saying, we assumed that discretionary spending would be X% or Y%. There’s no basis for that assumption other than to drive the results of the analysis.

  36. comment number 36 by: Brooks

    Steve,

    Earlier I said:

    I don’t see what the point would be of calculating “remaining funding” after projected spending on entitlements, unless one wishes to say “If we don’t cut projected entitlement spending and if we want balanced budgets, we have to limit discretionary spending to $X”.

    It seems that that is exactly what you are saying. Well, ok, if that’s a way you wish to present our fiscal policy choices, fine. But it says no more about the magnitude of our long-term fiscal imbalance than any other arbitrarily chosen subsets of projected overall revenues and projected overall spending. All it says is “If we take projected spending for this category or subset of overall spending as a given, and if we take projected revenues as a given, and if we want balanced budgets, this is how much we have left over for all other spending”.

    Of course, we need not have balanced budgets even on average (e.g., over the course of economic cycles) and might not want balanced budgets, given the trade-offs. But for whatever level of deficits one considers acceptable, if he opposes policies that will produce higher than projected revenues, and if he is stating a preference for keeping projected spending on some subset of programs as is, then your framing can be an appropriate way to present the trade-off he is implicitly forcing, because there’s only so much left over to spend on everything else.

    And just to be clear (to any others reading this; I think you realize it), calculating the above and using it accordingly has nothing to do with the “unfunded liabilities” metric, which is not the calculation you are discussing, but rather a calculation of the difference between projected spending on particular programs and projected “dedicated” tax revenues associated with those programs.

  37. comment number 37 by: B Davis

    Brooks wrote:

    I realize there are difficulties and potential practical problems with factoring in wealth for means testing (measuring it; potentially adverse incentives regarding saving; etc.). I’m just pointing out the theoretical argument for it and suggesting the possibility of actually including it. I don’t know enough about the details of implementation and effects to say whether or not it would work as intended and be desirable on balance, but I do consider it plausible.

    I agree with factoring in wealth for means testing just as I believe that there is a valid rationale for taxing wealth. A rationale for taxing income is that various government services are necessary to maintain an environment in which people can earn that income. Likewise, a rationale for taxing wealth is that government services are also necessary to maintain an environment in which people can maintain their wealth. On that topic, estate taxes are one of most efficient ways of implementing such a tax. When the estate has more than one beneficiary and/or creditors, there needs to be an accounting of its assets anyhow. This is therefore the ideal time to levy a tax on wealth.

    Because there is a valid rationale for taxing wealth, the argument that the estate tax represents “double taxation” (because the wealth has already been taxed as income) is obviously flawed. It is based on the faulty assumption that only income should be taxed. Hence, the estate tax should not have been repealed in 2010 and there is no valid argument for letting that repeal continue. For this and the reasons mentioned in my prior post, Congress needs to address this issue as soon as possible.

  38. comment number 38 by: SteveinCH

    B Davis,

    By your logic, I can tax anything because government action is necessary to maintain an environment where people can do anything. As such, it’s not a logic for a wealth tax per se.

    As to your contention on double taxation, I profoundly disagree. The estate tax is double taxation as would be a wealth tax. You are free to argue that we shouldn’t care as there are many other forms of double taxation (mostly excise taxes) but to argue it isn’t double taxation seems quite inconsistent with the meaning of the words.

    As to the merits of a wealth tax, I think enforcement would be quite an issue. Unlike a means test which could be configured around particular break points (e.g., over or under 500k) and the extra spending from cheating is likely low if you assume that many people can’t “cheat’ enough for it to matter, a wealth tax would require a precise estimation of wealth increasing dramatically the deadweight loss from “cheating”

    In addition, I’m not sure a wealth tax is constitutional but that’s an entirely different thread.

  39. comment number 39 by: AMTbuff

    B Davis didn’t deny that a wealth tax is double taxation. He said it was justifiable double taxation.

    The problem with taxing both income and wealth is that it penalizes saving over spending. Consume now and you avoid the wealth tax. Defer consumption and you risk paying a double tax. A tax code that discourages savings is more economically destructive than it needs to be to raise a given amount of revenue. However it might still be more politically viable, since it panders to envy.

    A wealth tax is not allowable under the letter of the Constitution, but I wouldn’t expect that alone to block it. Besides, if the government switched to the Haig-Simons measure of income, we’d effectively have a wealth tax, albeit with single taxation only and with no taxation of wealth accumulated to date.

  40. comment number 40 by: B Davis

    SteveinCH wrote:

    By your logic, I can tax anything because government action is necessary to maintain an environment where people can do anything. As such, it’s not a logic for a wealth tax per se.

    In fact, we do have a number of usage taxes which serve, at least in part, to pay for the government action that is necessary to make the usage possible. However, there are a number of government services (such as the courts, the police, and the armed forces) which do not lend themselves to a usage tax. The only general things that can be taxed to pay for these services are income and wealth. By what rationale should the burden be placed wholly on the income earner? Both the income earner and the holder of wealth benefit from the stable environment created by government services. As you can see from the second graph at this link, the great majority of federal tax revenue comes from income tax. And yet we have a great pounding of fists and gnashing of teeth over the tiny sliver of tax revenue that comes from estate taxes.

    Of course, any level of taxation must be set at a level which does not unduly burden the activity being taxed. In addition, it needs to be enforceable at a reasonable cost. Placing an annual tax on all wealth would likely be unfeasible since it is much more difficult to track wealth than it is to track income. Likewise, it would likely be difficult to have a means test that depended on an annual measure of wealth. Hence, as I mentioned, estate taxes are one of most efficient ways of implementing a wealth tax. When the estate has more than one beneficiary and/or creditors, there needs to be an accounting of its assets anyhow.

    As to “double taxation”, a large number of estates end up benefiting from “zero taxation”. At death, stocks receive a step-up in basis such that their increase in value up to that point is never taxed. As explained, at this link, there will continue to be $1.3 million of this step-up allowed for nonspousal heirs and and additional $3 million allowed for the husband or wife. Those lost taxes will have to be made up by income-earners.

    In any event, there is no excuse for continuing the “Throw Momma From the Train Act of 2001″ whereby estate taxes are repealed this year but come back in full force next year. Congress should reinstate the estate tax at the relatively low 2009 level until they can come up with a long-term policy. Until then, we have a law that is creating perverse incentives that making it impossible for people to plan their estates.

  41. comment number 41 by: SteveinCH

    If you object to the step up, then get rid of it and have people pay cap gains as appropriate. That’s not an argument for the estate tax. By the way, a cap gains tax is also a form of double taxation but again a subject for a different day.

    My point was your logic for a wealth tax justifies a tax on anything, thus, it has nothing to do with justifying the estate tax.

    As AMT points out, wealth taxes promote consumption at the expense of savings. It’s not at all clear to me this is a good idea.

    Finally, if you are in favor of wealth taxes, you should also be in favor of reducing benefits to seniors with wealth. After all, what is the point of giving money to people to create larger estates (than they would have had without the transfer) that you will then tax through the estate tax.

    It certainly makes more sense to stop the transfer to the wealthy than it does to tax them after the transfer.

  42. comment number 42 by: B Davis

    SteveinCH wrote:

    If you object to the step up, then get rid of it and have people pay cap gains as appropriate.

    Good luck on finding a politician who would support that. It would increase taxes for the great majority of estates as they are under the minimum taxable size for estates (which was $3.5 million in 2009) but would lose the step up. In any case, a big part of the reason for the step up is that beneficiaries who inherit a stock often don’t have records as to it’s cost basis.

    That’s not an argument for the estate tax. By the way, a cap gains tax is also a form of double taxation but again a subject for a different day.

    I have to profoundly disagree. A major reason why people work is to earn income and increase their wealth. Similarly, a major reason why people buy stocks is to earn capital gains and likewise increase their wealth. Why should one form of increasing one’s wealth be taxed and another not be taxed?

  43. comment number 43 by: SteveinCH

    I didn’t think we were debating political reality but if you’d like to, most of what’s posted on this site isn’t realistic politically, including your idea of a wealth tax.

    Because two activities have a similar purpose, does not mean they should be treated similarly. Cars and bicycles are both designed to get you from point A to point B but aren’t treated similarly in a bunch of ways.

    Cap gains is double taxation. Where did the money for the investment come from? Oh yeah, that would be after tax income thus meaning the money has already been taxed. As I said above, you may or may not consider double taxation relevant but you can’t argue it isn’t double taxation credibly.

    My point was your initial argument could be applied to anything, not just wealth. I stand by that argument. Society allows many things to happen but not all of these are taxed so that in an of itself is not a justification for a wealth tax.

    In the end, I don’t think it matters. You believe we should have an estate tax. I agree with the provision that we stop paying people money to artificially sustain their estates prior to taxing them. You seem to believe there should be a wealth tax if we could figure out how to administer it. I disagree and, if you want to talk about the politically impossible, go with that plan.

  44. comment number 44 by: Jim Glass

    “By the way, a cap gains tax is also a form of double taxation but again a subject for a different day.”

    Well, then consider this posted tomorrow.

    I have to profoundly disagree.

    The capital value of an asset = the future income from it discounted to present value.

    Thus, if the income from an investment (interest, dividends) is taxed, and the capital value of the investment (the current value of a bond, stock share) is also taxed, double taxation of the income from the investment results, unambiguously.

    Anyone in doubt about this can look at the BEA’s national income accounts. There is no entry in them for “capital gain income” because it would double count income.

    Similarly, in the early days of the income tax the US Supreme Court held (four times, IIRC) that capital gain was not income and so was not subject to income tax. (Following, of course, the advice of the leading economists of the era). For instance…

    “The mere fact that property has advanced in value between the date of its acquisition and sale does not authorize the imposition of a tax on the amount of the advance. Mere advance in value in no sense constitutes the gains, profits, or income specified by the statute. It constitutes and can be treated merely as increase of capital.”

    Lynch v Turrish (1918) approvingly quoting its earlier decision Gray v. Darlington.

    “Enrichment through increase in value of capital investment is not income in any proper meaning of the term.”

    Eisner v Macomber (1920)

    Then Congress changed the law, specifically making capital gains taxable as “income”, whether income or not.

    So it’s pretty clear that a tax on capital value (via capital gain tax, wealth tax, estate tax, wahtever) does double-tax income.

    More disturbingly, the doubled tax lands on investments, economic savings.

    The nation’s future wealth is dependent on savings and investment — especially today, when the savings rate has been far too low for too long, and we have a tsumami of retiree obligations coming due soon. We need all the investment we can get. The last thing need is to have the politicians tax investment to consume it … paying for COLA handouts to retirees to make up for deflation, more bombs for Afghanistan, whatever.

    Which is why economists generally (often strongly) favor consumption taxes over our kind of income tax — since it taxes consumption but doesn’t tax investment … much less double tax investment!

    “As long as an inheritance tax remains a true inheritance tax, it always involves a conversion of capital into income, hence an act of economic waste which is damaging to all.”
    – Joseph Schumpeter.

    Say Barron Hilton dies and leaves his fortune divided between his two grandchildren Paris and Scrooge. The grandkids each liquidate what they inherit at the same rate of $X million a year — which is taxable capital gain to them (from the stepped-up basis level, of course).

    * Paris consumes her capital gains on yachts, champaign, gigolos, and Faberge Egg-throwing parties.

    * Scrooge lives ascetically in a basement re-investing every dollar of his gains, financing the growth industries of the 21st Century (which will employ our kids, whose taxes will pay for your and my Social Security and Medicare).

    Should Paris and Scrooge pay the exact same taxes?

    Should Scrooge pay much more in taxes than Paris, because of all the extra wealth he creates? (Thus reducing how much he can create.)

    Should Barron’s estate be taxed like Paris on the capital value he left in it as the result of his saving, investing and business-building like Scrooge (thus reducing the nation’s stock of capital investment)?

  45. comment number 45 by: B Davis

    SteveinCH wrote:

    I didn’t think we were debating political reality but if you’d like to, most of what’s posted on this site isn’t realistic politically, including your idea of a wealth tax.

    If you reread my posts in this thread, the only kind of general wealth tax that I have supported is the estate tax. That has been realistic politically since 1916 in the United States as it has existed since then.

    Cap gains is double taxation. Where did the money for the investment come from? Oh yeah, that would be after tax income thus meaning the money has already been taxed. As I said above, you may or may not consider double taxation relevant but you can’t argue it isn’t double taxation credibly.

    And where did that after tax income come from? In most cases, it came from money received in exchange for products or services. And where where did that money come from? Oh yeah, that would be from after tax income that has already been taxed! Your definition of double taxation counts only money that is earned via labor and resets the clock, so to speak, each time that the money changes hands. And yet, when the money changes hands via an estate, you DO count that as double taxation. You are free to use any definition of “double taxation” that you wish. But that does not mean that others must accept your definition or that any definition should be used to set tax law.

    My point was your initial argument could be applied to anything, not just wealth. I stand by that argument. Society allows many things to happen but not all of these are taxed so that in an of itself is not a justification for a wealth tax.

    In fact, the only possible sources for a tax revenue is income or wealth. You can’t obtain tax revenue from someone who has neither. I believe that it is justifiable to tax an increase in wealth, whether that increase is obtained through income or through investments. Both benefit from the existence of government services (such as the courts, police, and armed forces). Why do you believe that it is OK to tax someone who obtains wealth through income but not OK to tax someone who obtains wealth through investment?

    In the end, I don’t think it matters. You believe we should have an estate tax. I agree with the provision that we stop paying people money to artificially sustain their estates prior to taxing them. You seem to believe there should be a wealth tax if we could figure out how to administer it. I disagree and, if you want to talk about the politically impossible, go with that plan.

    As I said, I have not proposed any new wealth taxes. In fact, I consider myself relatively conservative on government policy matters. I think that the great majority of current policies are better than the great majority of policies being suggested by think tanks and various pundits. Current policies have at least gone through some test in the real world. That’s not to say we should not be constantly attempting to improve current policies. Just that we should be very careful and attempt to achieve as much of a consensus as possible before we do change current policies. Among those current policies (at least until this year), I would include the tax on capital gains and the estate tax.