/* trackback code -- i added this */

Sunday, December 03, 2006

Making Paper Assets Worth Something

In my last post, I commented on M.P. Dunleavey's page-B6 "Basic Instincts" column in yesterday's NYT Business section (entitled "Plan to Retire But Leave Out Social Security"). That post concerned the Chicken-Little silliness of her column's main point ("Omigosh! My Social Security Statement says the trust fund will run out in 2040! Better assume I'll get Zero!!")

In reading Dunleavey's column, I had a couple thoughts that I hadn't had before. A key part of the political attack on Social Security last year was the canard that the bonds it has accrued since the 1983 Greenspan Commission's Report (filled with raving leftists, as you can tell by its having been chaired by Alan Greenspan) recommended that the system "save" for the purposes of dealing with the future costs of the Baby Boomers' retirement. Congress enacted and President Reagan signed legislation that implemented key elements of this report. Most relevantly for this post, the payroll tax was increased so that the system could accrue assets meant to be sufficient to defray the costs of the Boomers' retirement at a time when there would be fewer workers per retiree, and so on. In other words, the Greenspan Commission contemplated exactly the challenges we faced. We find ourselves in our current situation because the total value of bonds that will have accrued from the tax increases recommended by the Commission turned out to be insufficient to meet the program's shortfall through 2056 (which was the long run horizon that the Commission was charged with addressing). As I understand things, this inadequacy is primarily due to an unexpected slowdown in wage growth (benefit levels for current retirees are insensitive to current tax revenues, which are themselves determined by total earnings up to a per-individual cap, which was $90,000 last year).

There is no question that the Social Security system faces a shortfall and that something should be done about it (I tend to favor some version of the Diamond-Orszag plan, about which I will perhaps write in the future). But the sky-is-falling crowd has made an especially large bale of hay out of the supposed vulnerability of the special bonds that the Social Security Trust Funds have been accruing since the Greenspan Commission-induced surpluses commenced. Simply put, here are the basic facts:
  • At present, the Social Security system takes in more in revenues than it pays out in benefits. (This situation has gone on for 2 decades and will continue for the next 11 or so years, even if no changes are made.)
  • By law, the resulting surpluses must be invested in special Treasury bonds. Thus, the Social Security Trust Funds are required to invest in low-yield assets (perhaps I'll discuss the bonds-versus-stocks issue, and its general irrelevance, another time).
  • Those bonds are backed by the usual "full faith and credit of the United States government" promise.
  • Nonetheless, some proponents of structural changes---i.e., partial or full privatization---of Social Security have tried hard to convince the public that "There is no 'trust fund,' just IOUs that I saw firsthand," as President Bush said during a visit to the Bureau of the Public Debt in April 2005.

The worthless-pieces-of-paper PR strategy has probably made headway with some people. And, as Olivia Mitchell of Wharton is rightly quoted as saying in Dunleavey's column, "the Treasury goes ahead and spends that money, so although there is a promise, nobody knows where the money will come from to pay them back."

Mitchell's point is commonly made, and it is certainly correct. Which brings me to the couple of thoughts I had reading Dunleavey's column yesterday:
  1. Suppose we accept, as President Bush and some of his backers claim, that the Social Security Trust Funds' assets are particularly worthless since they are simply promises by one part of the government to pay another. Then, the system can pay promised benefits only if Congress and the President are willing, as the President also said in his visit to the Bureau of the Public Debt, to enact policies so that "future generations will pay [for these IOUs]...either in higher taxes, or reduced benefits, or cuts to other critical government programs." So we are accepting that the system's assets are not really assets at all---not in the sense that a government or corporate bond you or I might own is an asset. With such an asset, we simply sell or cash it in and receive its value. Barring bankruptcy of the issuing entity, we will certainly be paid. By contrast, on the accepted line of reasoning, when the Social Security Trust Fund tries to redeem its "assets" from the U.S. Treasury, the Treasury can simply say, "Ix-nay on the Asset Pay", since presumably the "higher taxes, or reduced benefits, or cuts to other criticial government programs" may simply be too politically unpopular. Thus these assets are more hope than plan.

    But if that is true, then there is not really any Social Security system at all! Social Security is then just another line item in the government's budget. There happens to be a tax "dedicated" to it, but benefits continue regardless of that tax's revenue, anyway. So why shouldn't we equally doubt, say, the Pentagon's ability to get its budget, or the department of transportation, and so on? Hey, why assume that the wealthiest Americans can continue to count on the reduced taxes Congress and the President have bestowed on them? No particular reason, none at all. Social Security's political defenders should start making the case that if there is no Trust Fund as such---if the Greenspan Commission was just a con---then there is no reason to treat Social Security as a special program. If those assets aren't really assets at all, then there really isn't a relevantly separate Social Security program, either. Of course this is really quite silly---of course there is a program, one in whose name quite a bit has been promised. That is why we need a way to enforce the legally binding nature of those bonds. So how to do that?

  2. Quite simply.
    1. First, change the legal status of the Social Security Administration to a private, non-profit corporation that is legally charged with meeting benefits under current policy (including the provision about automatic cuts should full benefit payment make the system insolvent).
    2. Second, require the federal government to direct all relevant payroll tax revenues (i.e., the 12.4 percent part, for you wonks) into the coffers of this private corporation.
    3. Third, have Congress convert all those special bonds into bonds that are legally indistinguishable from ones purchased at public auctions.

    If we took this step, there would no longer be any way to default specifically on bonds held by the Social Security system without defaulting on all sorts of bonds. Those worthless pieces of paper would be indistinguishable from all the other full-faith-and-credit ones held by private individuals and corporations (and foreign governments). I'm not an expert on these details, but I'm betting this plan would pass constitutional muster (see "Reserve System, Federal" for a similar case). It would also have the substantial virtue of (a) making good on the Greenspan Commission's, Congress's, and President Reagan's promises, and (b) forcing politicians to focus on the real issue, which is about benefits and tax revenues, and eliminating the legalistic debate over the status of duly lent money.

2 Comments:

Anonymous Anonymous said...

An interesting idea. Thanks for offering it.

I have a couple of quibbles, if you'll indulge me.

First, you write that the proposed solution would have this effect:
"If we took this step, there would no longer be any way to default specifically on bonds held by the Social Security system without defaulting on all sorts of bonds."

But no one is actually suggesting, despite the frequent allegation to the contrary, that the government is going to "default" on the bonds. In fact, if you look at the actuarial runs even on all of the personal account plans, they all provide for the redemption of all the bonds in the Soc Sec Trust Fund. And most of the sponsors of these plans have taken pains to point out that the government will honor the bonds in the fund (see the 2001 report of the President's Commission as one example.)

The point isn't that anyone is going to "default." One point is simply that redeeming the bonds in the Trust Fund will require -- well, redeeming the bonds in the fund -- and that will cost money.

The second point is that the Trust Fund has not proved to be an effective means of advance-funding Social Security because the money is used to finance current federal spending. A number of scholars have written about this, from Kent Smetters to John Shoven to CBO. So the issue isn't whether the government is going to "default," but whether redemption of the bonds will be a bigger strain than it otherwise would be, because the current system spends rather than saves the money.

Next quibble:

You wrote:

"the 1983 Greenspan Commission's Report (filled with raving leftists, as you can tell by its having been chaired by Alan Greenspan) recommended that the system "save" for the purposes of dealing with the future costs of the Baby Boomers' retirement."

This is conventional wisdom, but it's a myth. One of the unwritten stories of the Soc Sec debate is not only that the Trust Fund has not been an effective means of pre-funding, it was never intended to be. There's a good CRS report on "Lessons from the 1983 amendments" and there's also an online oral interview with the Commission's Exec Director Robert Myers at ssa.gov. Also, see several public comments by the late Senator Moynihan.

The evidence all shows very clearly that neither the Greenspan Commission, nor the committees of jurisdiction that wrote the legislation, ever contemplated building up a big Trust Fund to pre-fund the Baby Boomers' retirements. The only numbers they saw covering the 75 year period were average numbers -- how much each provision did to repair the 75-year actuarial balance, on average. They didn't look at the annual cash flows, and thus didn't see that they were building up big surpluses in the near term to be followed by big deficits in the long term.

In fact, the documentary record is clear that after the legislation was passed and the detailed projections became more apparent, several of the participants came out and said just the opposite, that they never intended it to be this way, and would have done it differently had they realized.

Further proof of this lies in the actuarial methods used by the Greenspan Commission. They didn't use the method we use today, which was adopted in 1988. That method discounts Soc Sec's future shortfall at the rate of interest "earned" by the Trust Fund. That method was adopted in 1988 to reflect the idea that the Trust Fund was supposedly building up and earning interest.

But that's not how the Greenspan Commission defined actuarial balance. They defined is solely in terms of the net balance of surpluses and deficits in the system, as a percentage of taxable wages (the "average cost" method.)

Obviously, if they had intended to advance fund the system through a big Trust Fund, they would have used an actuarial method that would have reflected this viewpoint. But they didn't.

"the tax increases recommended by the Commission turned out to be insufficient to meet the program's shortfall through 2056 (which was the long run horizon that the Commission was charged with addressing). As I understand things, this inadequacy is primarily due to an unexpected slowdown in wage growth"

The Clinton Administration's Soc Sec Advisory Council examined the reasons for the deterioration since 1983. The three biggest were:
-- Unsustainability -- the 1983 fix was never sustainable to begin with, for the reasons above, which the Commission was unaware of.
-- Technical mistakes -- a bunch of them, including treating immigrants as though they were newborns (and thus underestimating the benefits they would get in the near term)
-- Economic estimates -- the Commission used economic growth estimates that were rosier than previous experience, which over-optimism had to be tempered by real data as it rolled in.

This last point should be a warning to us. There's a myth out there that the Trustees are using unduly "conservative" economic estimates and that we should repeat the Greenspan Commission's mistake of wishing a good deal of the problem away by making them more optimistic. But that's a myth; the Trustees are actually projecting faster real wage growth over the long term than we have had on average over the last 40 years. (See http://www.ssa.gov/OACT/TR/TR06/V_economic.html#wp133682).

If you've made it this far, thanks for wading through this.

12/04/2006 7:07 AM  
Blogger Jonah B. Gelbach said...

hi anonymous

thanks much for your comment. i'm short on time, so my reply will be briefer than i'd like.

on your first quibble, i think you're missing my point.

you'll note that i didn't mention (aside from D-O) or criticize any of the reform plans. i'm no expert on the internals of these plans (and you clearly know quite a bit), so i'm happy to stipulate your claim about these plans' authors' insistence on honoring the special bonds.

but much political and media discussion has, in fact, concerned claims that those bonds are worthless IOUs. obviously i agree with you "that redeeming the bonds in the Trust Fund will require -- well, redeeming the bonds in the fund -- and that will cost money" -- i say "obviously" because it's a point i made in my original post.

my overall point here is, let's get away from the silly argument over whether these bonds are bonds that we should expect the govt can or will honor. my idea is geared to suspending the license of commentators who are either unwilling or unable to grasp the points you make.

Regarding your second quibble: I'm well aware of the fact that our representatives have chosen to spend the Trust Fund surpluses. But I think you are too quick to maintain that these surpluses aren't effective in advance-funding Social Security. How do we know that separate accounting (or an absence of OASDI surpluses) would lead to reduced spending or increased taxes elsewhere in the public sector? It seems to me that this counterfactual is nonobvious.

That said, if you are suggesting that in the absence of these surpluses, Congress and the President would indeed be forced to run a tighter ship, then you should really like my idea. Why? Because it would force---or, at least, it could be made to force---the public accounting of the budget to count SSTF surpluses as borrowing, rather than net revenues. For this to work, of course, the SSA would really, honestly have to be a private corporation rather than a govt agency. Which is the crux of my idea.

Regarding your third quibble, you write that the members of the Greenspan Commission

didn't look at the annual cash flows, and thus didn't see that they were building up big surpluses in the near term to be followed by big deficits in the long term.

Yet in the second graf of the "Findings and Recommendations" chapter of their report (http://www.ssa.gov/history/reports/gspan5.html), they write that

The National Commission recognized that, under the intermediate cost estimate, the financial status of the OASDI program in the 1990s and early 2000s will be favorable (i.e., income will significantly exceed outgo) -- see Table 7A in Appendix K.

Moreover, if you look at that table (http://www.ssa.gov/history/reports/k7a.html), you see that they certainly were aware of more than just averages. Whatever the demerits of the actuarial approach they used, they had access to year-specific numbers that implied "big surpluses in the near term to be followed by big deficits in the long term". Furthermore, it's hard for me to believe that a collection of such smart people (and they were such a collection) could seriously fail to realize that any constant tax rate and benefit policy (not constant benefit amt, but rather benefit formula) would yield a buildup followed by a spend-down. Unlike economic figures necessary to forecast the system's health, the demographic facts are pretty easy to get (mostly) right quite far in advance.

In any case, the GC part of my post is basically a rhetorical flourish.
Your points about flawed accounting and technical errors are well-taken, but they don't change the fact that anyone looking at the system the Commission suggested---and that was enacted---could see that the point was advance funding. There's a difference between doing a poor job at a task and not meaning to do that task; your critique suggests the former rather than the latter.

Lastly, I don't know much about the details of the debate over the merits of the intermediate assumptions, so I will gladly take your argument as a cautionary tale (unfortunately, I don't have time at the moment to wade through the very informative link you provided; thanks for including it, though!).

12/04/2006 1:52 PM  

Post a Comment

Links to this post:

Create a Link

<< Home