Social Security and the Debt Clock (why I get headaches sometimes)
by Bruce Webb
The Treasury Department maintains a handy web application called Debt to the Penny. As the name suggests it will give you total Public Debt to the penny for specific dates in the past or for a range of dates. It is this total Public Debt figure that is generally cited in news coverage, resulting in reporting that debt under Bush went from $5,727,776,738,304.64 on Jan 19, 2001 to $10,626,877,048,913.08 on Jan 20, 2009. If theoretically Bush had continued Clinton era fiscal policies and had Clinton era outcomes would the the final Total Debt figure be higher that $5.7 trillion? Or lower? or About the same?
The obvious, intuitive answer would be ‘lower’. That is even small General Fund surpluses serve to lower overall debt and that effect combined with the higher rate of real wages experienced in Clinton’s terms means higher revenue and an even greater surplus in Social Security, so General Fund surplus plus Social Security surplus = lower debt. Right?
Hmm, well, no, not necessarily. To see why you can follow this below the fold. Just be sure to bring your headache medication of choice.
Because a closer look at Debt to the Penny shows that Treasury tracks two different debt categories: Debt held by the Public and Intragovernmental Holdings and combines them to get total Public Debt. And more than half of Intragovernmental Holdings are made up of the Special Treasuries in the Social Security Trust Funds. The Bush years did not put any serious holes in long-term possibilities for Social Security solvency, results approaching total system solvency being by my calculations more probable than not. That doesn’t mean the TF came through unscathed. Per the 2001 Report’s Intermediate Cost projections toward year end 2008 TF balance was projected to be $2.808 trillion up from $1.049 trillion. Table II.D1.- Abbreviated Operations of the Combined OASI and DI Trust Funds, Calendar Years 2000-10 [Amounts in billions] Instead it ended up at $2.4 trillion.
So if we return to our theoretical scenario we would have Clinton era small General Fund surpluses combined presumedly with Intermediate Cost SS surpluses. But our equation changes. Now we have General Fund surplus MINUS Social Security surplus = total debt. Whether the resultant is higher or lower than the original $5.7 trillion in total debt is difficult to calculate in precise terms but it seems likely that the growth in the TF by $1.75 trillion (2001 projection ) would have likely been significantly more than the cumulative GF surplus meaning that the Debt Clock would have continued to tick.
If we examine the Social Security Reports from 1997 to 2007 we see that the optimistic Low Cost alternative projects what would seem to be a Goldilocks outcome, with the porridge being neither too hot or too cold. After all what is it about fully funded benefits with no needed changes in taxation or retirement age is there not to like? Well nothing really. Until you start looking at issues of intergenerational equity. Because how does Low Cost translate to the Debt Clock in future years?
Table VI.F8.—Operations of the Combined OASI and DI Trust Funds, in Current Dollars, Calendar Years 2008-85 [In billions]
2040 $11.7 trillion; 2060 $28.3 trillion; 2080 $88.0 trillion and all of that scoring as debt on the Debt Clock. That is why I get headaches, fully funded Social Security translating to ever mounting total Public Debt.
Which only compound when I consider the flipping point. Under Low Cost assumptions Social Security relies on interest earned on the TF to fill the gap between revenue from taxes and total cost for every year from 2023-2064. But the actual principal in the TF is entirely due to excess contributions made by workers from 1983-2023. In 2065 Low Cost projections would have tax revenues once again exceeding cost without needing to tap the interest on the principal which at that point would total $37.0 trillion. Who has the moral claim to that money? In 2065 there will be substantial numbers of retirees who were in the work force prior to 2023. But only that fraction of them who owe tax on benefits are still contributing anything and their whole cohort has secure benefits going forward, they can be made whole by eliminating tax on benefits. So what claim do those workers who entered the work force after 2023 but before 2065 really have? Sure their contributions served to largely fund the retirement of people before them, but only ‘largely’ because a substantial part of that cost was picked up by interest on surplus payroll contributions they never had to pay, to some degree they have had a subsidized ride and have been made whole as is. And the worker entering the workforce in 2065 doesn’t have too much of a claim on the balance given that he has yet to pay anything at all and is projected to get full benefits.
The simplest answer is to first simply write down the TF in 2065 from $37 trillion to $7.5 trillion (one year of reserves) and then reduce FICA to the level where it plus interest on the remaining $7.5 trillion will continue to meet total cost. And then secondly commit to rebating to surviving retirees any lifetime tax on benefits paid. In this scenario nobody gets seriously screwed, they paid their insurance premium in the form of payroll tax, everyone gets full benefits. But viewed from another light it is just a $29.5 trillion dollar gift to those people who were on the hook previously, i.e. high income earners who just end up with a huge liability lifted. Meaning that the people who borrowed all that money from 1983 to 2023, or at least their heirs, get a windfall while still have being able to put their boots in the sides of Boomers and Gen-Xers (who themselves would have been fooled into blaming Boomers all along).
This is not a simple story, which is why it gives me headaches to explain. But it goes to show why Social Security surpluses are not an unalloyed good. Instead they can serve to create a mental picture of total debt that is disconnected with economic reality. If we end up achieving Low Cost outcomes, or outcomes close to that, and as a result at some point in the next fifty years we end up writing down the TF by $29 trillion, is that portion of the principal really debt that should be scored on the Debt Clock in the meantime? Certainly it was a liability to the degree that interest had to be paid on at least a portion of it (in the mid-thirties this approaches 50% of accrued interest needing to be tapped in any given year, that percentage drops over time). Well it is hard to say.
Moral? Don’t let the Pete G. Peterson people get a backdoors victory by trumpeting Total Debt while slyly denying that the Trust Fund is real. Because if it is not real to the degree that Peterson et al are never going to have to pay the borrowed money back, either because the economy grows fast enough to mean it is not needed in the end or because they will just get us to accept lower benefits, then it ends up being no more debt than ‘unfunded liability’ means an actual liability. They are calling ‘debt’ to scare you into actions that remove that debt. From their ledgers. Because Peterson would still have workers be screwed in the form of lower benefits.
(BTW these calculations show why Buffpilot’s claim that Clinton didn’t reduce debt is not quite right. First of all some of that ‘debt’ was really SS surpluses, and second he ignores inflation by using nominal and not real numbers.)