Social Security and the Infinite Future: From 1997 to Heat Death

by Bruce Webb

Eric Laursen in his post Cato Premieres its Latest Horror Show points us to a new book forthcoming called ‘Social Security; A Fresh Look at Policy Alternatives’ by senior Cato fellow Jagadeesh Gokhale. Gokhale along with his oft-time collaborator Kent Smetters seems to be the father of ‘intergenerational equity’ and lead advocate for measuring that over the ‘Infinite Future Horizon’.

Eric, backed up by people as varied as the American Academy of Actuaries and conservative economist Bruce Bartlett, does a nice job showing why Infinite Future is not a useful tool in this instance, a topic I have taken up from time to time and I urge people to read it. But what interested me is what it revealed about the curious timeline during which Infinite Future even became part of the Social Security Policy discussion. Details below the fold.

The key paragraphs come right up front in Eric’s piece:

In August 2003, Joe Lieberman, then in the early stages of launching a presidential bid, wrote a letter to Treasury Secretary John Snow in which he accused the administration of “stripping out” from its 2004 budget the findings of an internal Treasury paper that Snow’s predecessor Paul O’Neill had ordered up. Attempting to stake out a position as the toughest of the deficit hawks, Lieberman suggested that “this administration is trying to hide the true nature of our financial obligations from the American people in order to advance its agenda of cutting taxes indiscriminately.”

The paper was written by Gokhale, then at the Cleveland Fed, and Kent Smetters, then deputy assistant Treasury secretary for economic policy. Contrary of Lieberman’s claim, Smetters claimed it had been “for internal discussion only,” to try to help O’Neill “think about [the deficit] from an economic perspective.” There was no conspiracy to suppress it, he said; the administration considered including it in the budget but then decided against it. Gokhale and Smetters revised their paper and presented it four times in Washington in May 2003, including to the American Enterprise Institute. They then published it as an AEI monograph.

This suggests without saying so that this ‘internal Treasury paper’ was the work product of Sping 2003 and was suppressed during the preparation of the Budget that Fall. But this can’t be so, as Smetters and Gokhale pointed out in a July Op-Ed similar language had been included in the 2003 Social Security Report, itself signed off by newly installed Secretary Snow.

The 2003 Social Security Report was released on Mar 17, 2003, two weeks earlier than normal and coincidentally or not the week the invasion of Iraq was announced, to say that ‘nobody noticed’ was a gross understatement. In any event it contained this curious (to me) language.

“Even a 75-year period is not long enough to provide a complete picture of Social Security’s financial condition. Figures II.D6 and II.D7 show that the program’s financial condition continues to worsen at the end of the period. Some experts have noted that overemphasis on summary measures for a 75-year period can lead to incorrect perceptions and to policy prescriptions that do not move towards a sustainable system. In order to provide a more complete description of Social Security’s very long-run financial condition this year the Trustees present actuarial estimates over a time period that extends to the infinite horizon. These calculations show that extending the horizon indeed increases the unfunded obligation, indicating that much larger changes would be required to achieve solvency over the infinite future as compared to changes needed according to 75-year period measures.”

Who were ‘some experts’ and why was their advice privileged over say the statutory group of outside advisors Social Security Advisory Board (a group of which Gokhale is the newest advisor).

So whether or not Smetters and Gokhale are the ‘some experts’, clearly this idea was floating around early enough to be incorporated in a March Report, and if that ‘internal report’ had been prepared for Secty O’Neill, it would have had to come into being sometime before O’Neill left office in Dec 31, 2002. Meaning likely over the course of 2002.

Which leaves the lingering question of ‘who had the authority to have this included in the Reports to be signed by the Trustees on March 17?’ More importantly ‘why was this thought to be important or useful?’

Well a trip back to the timeline may be important. Although Bush didn’t run for office in 2000 heavily on the issue of Social Security clearly it was a high domestic priority, he had his CSSS Commission to Strengthen Social Security in place by June 2001, complete with charter and membership with a tight seven month deadline for reporting, by all indications he was ready to take this to the country in spring 2002. Well 9/11 changed his priorities, by the time the Report was released on Dec 21st we were engaged in Afghanistan, and the entirety of 2002 was eaten up building support for a War on Iraq, the time clearly wasn’t ripe to take on Social Security at the same time.

But this left the Report and its recommendations exposed, particularly CSSS Model 2 which was known to be the Administrations preferred plan (Model 1 & 3 mostly being variations on a fundamental theme. And during 2002 the CSSS Model 2 and related plans were subject to attack by places like EPI and CEPR . Because of a different timeline altogether.

From 1997 to 2004 Social Security was under tremendous outside pressure, not because its financial condition was degrading in the face of inactivity, instead it was improving year over year. In 1997 the projected 75 year payroll gap was 2.23% of payroll, by 2001 it was down to 1.86%. In 1997 the date of Trust Fund depletion was set at 2030, by 2001 it was 2039 and on way to being pushed back to 2042 by Report Year 2003. At this point what Baker and Weisbrot had identified in their 1999 book as Social Security: the Phony Crisis was looking pretty phony indeed. And the closer you looked at the internals the more phony it got.

The outlook for Social Security improved so dramatically from 1997 to 2002 for the very same reason that huge and growing Bush I deficits turned to small but growing Clinton General Fund surpluses and large and growing massive Clinton Social Security surpluses were turning into huge projected surpluses, we had just experienced an extended period of low inflation, high employment and significant Real Wage increases, all of which boosted revenues and cut projected cost. Moreover most of our economic elite, including most decidedly Fed Chair Alan Greenspan were insisting that this so-called Great Moderation was permanent, so much so that it justified huge tax cuts for the wealthy.

The problem is that the Great Moderation and Social Security Crisis were fundamentally at odds. Social Security Crisis in the yar 2000 relied on a relatively rapid and permanent slowdown in productivity and Real Wage over the next ten years. For example the 2002 Report (, the closest in time to the Dec 2001 release of the CSSS Report projected that Real Wage would drop from 2.8% in 2001, itself well down from the highs of the late 90’s to 1.1% in 2007 and stay there permanently (Table IV.B1). Likewise the IC model projected that unemployment already up sharply in 2001 to 4.8% would never get better and instead settle out permanently at 5.5% and that real GDP which maybe took a short term hit with the 2001 recession (which could be blamed on the last Administration) would never get back to the 4% plus rates of Clinton’s second term but instead was destined within ten-years to shrink to sub 2% a year FOREVER. (Table IV.B2). To which numbers a reasonable observer might ask: “What the hell is going to happen to this much vaunted Great Moderation?” and “This is the fricking miracle of tax cuts? Permanent 5.5% unemployment, 1.8% Real GDP, and 1.6% productivity?”

Given the outsized promises being made about the effects of tax cuts during the 2001-2002 time frame (i.e. all positive for everyone, capital and labor alike), and the promises that even bigger tax cuts would not materially reduce surpluses going forward, how could the Bush men still maintain Social Security ‘crisis’ with a straight face. Easy, you just redefine it as not being a crisis for this century, i.e. caused by Boomer retirement, but instead as a crisis for the NEXT century and the ones after that.

Thus the introduction of Infinite Future in time for the 2003 Report allowed the substitution of a 3.5% payroll gap for a 1.86% one and so gave cover for reform packages whose combination of benefit cuts and tax increases far exceeded the 75 year gap. For example the Liebman-MacGuineas-Samwick ‘Non-Partisan’ Plan proposes a combined 5.2% fix to a problem then scored at 1.92% (75 year) or 3.5% (infinite future). With a little slight of hand the LMS authors were able to hide 1.5% of the fix in the fine print and convince people that the difference between 3.7% and 3.5% would easily be made up by having an inheritable asset, the so-called ‘Ownership Society’. On close examination that breaks down, most workers wouldn’t end up with any estate out of their PRAs (unless they were lucky enough to drop dead right after retirement), but either way that was a lot easier to sell than the starker 5.2% vs 1.92%.

Now I can’t prove motivation for the switch to Infinite Future, but it is pretty clear that the original plan for the Bush Administration was to push the CSSS ‘Bi-Partisan’ Model 2 plan through Congress in 2002 in much the same way they tried to push the Social Security Crisis Tour through in 2005, or regrettably how Obama is allowing the Rubinistas to jam through cuts via the ‘Bi-Partisan’ Deficits Commission in 2011: assemble a theoretically ‘non-partisan’ group of ‘experts’, use some baked-up and bogus economic models and projections and then create a sense of inevitability around the whole thing. The key is to keep your eye on the numbers and not let them work out of two sets of books, one to predict crisis, and another implicitly to present solutions.

(For some earlier takes on this you might want to consult the first series of:, particularly VI, VII, and XII and then maybe follow up with selected posts from the third series:

The criticism came in mixed and matched form. First when compared to the productivity rates that were firmly being projected by Greenspan and used to argue for tax cuts, numbers in the out years seemed oddly muted. Where was this permanent productivity and GDP growth we were being promised? Second if the Trustees numbers were actually plausible how do you get the return on equities needed to fund PRAs (hence NELB and BDK)? All of which contributed to the biggest challenge of all: how do you put together a set of numbers that funds PRAs and Legacy costs all at a lower rate than a straight out tax fix would? Particularly when year over year the cost of that fix kept dropping?