Soc Sec XXII: What if? Low Cost as Fairy Tale

Although a lot of people are aware that ‘Low Cost is Out There’ it is difficult to impossible to find anyone who takes it seriously. Some experts simply dismiss it out of hand (Samwick), others point to the Stochastic Projections and simply deem it unlikely (Biggs), then again others reject it on grounds that run from legitimate ($130/barrel oil; $8 bushel corn futures) to ridiculous (fertility {a long story to explain why}). But the fact remains that unlikely does not mean impossible, we may very well see outcomes approaching Low Cost. And if we do the result will rock our world, because the Brave New World of Low Cost has real world implications that range well beyond retirement security.

So I propose we throw probability considerations to the wind and undertake a guided tour through LowCostLand. To switch literary references those that dare can follow me through the Looking Glass.

This figure shows in graphic form the outcomes of Intermediate Cost (II) vs High Cost (III) vs Low Cost (I)

Figure II.D6 from the 2008 Report shows us the contrast between Intermediate Cost (II) and Low Cost (I) at a glance. The chart shows Trust Fund ratios over time under all three alternatives provided by the Trustees. Alternative II or IC shows the standard and familiar narrative, the growth in the Trust Fund ratio slows after 2014 stalls in 2017 and then shrinks more or less rapidly until the Trust Fund is exhausted in 2041. (We should note that the stall in the growth of the ratio does not immediately show as an absolute shinkage in the dollar balance, that actually continues to grow until 2023, the distinction between TF ratio and cash balance will be important later on). Alternative III or High Cost shows much the same picture only accelerated. But we are here to examine Alternative I, that of Low Cost.

Here we see the TF ratio continuing to increase all of the way to 2023 and a level of 450, then a decline down to 390 by 2041, then about a fifteen year period of relative stability until around 2055 which is followed by a marked upward movement through the rest of the 75 year actuarial window. For the purposes of this post I intend to ignore the period past mid-century, it has interesting implications of its own, ones that I have already covered in my old post from 2006 Interest on Interest: a Threat. Instead I want to look at the medium term consequences of Low Cost between 2008 and 2041.

Okay, lets look at first order effects. If we flash forward to 2041 we see some pretty self-satisfied Boomers, (including that cranky old guy waving his cane in the air yelling “I told them rat bastards!! Right back in Ought Eight!!!” Don’t worry about old Mr. Webb, buy him a glass of beer and he quiets right down). Under Low Cost outcomes they ended up with 100% of scheduled benefits, in their case the equation runs 100% of 160% = 160%. Moreover their kids either already in retirement or nearing it also look to get full benefits without having to endure an increase in retirement age. And just as important the grandkids were able to pay for it all at the same 12.4% rate the Boomers paid. The question of intergenerational equity never ends up getting raised. At this point we could wrap up the fairy tale in traditional fashion with “And they all lived happily ever after”. But instead we are going to invoke that incredibly annoying TV pitch man Billy Whatisname and cry out “But wait!! There’s more!!!”.

Because there is. What happens if we translate that curve into dollars? The Trustees present future outcomes in both constant 2008 dollars and in then current dollars. To keep relative tax burdens steady I propose to use the former and so use Table VI.F7.—Operations of the Combined OASI and DI Trust Funds, in Constant 2008 Dollars, Calendar Years 2008-85 [In billions]. First thing to note is that Low Cost does not mean No Cost, that dip in the ratio between 2023 and 2041 represents transfers from the General Fund to Social Security. For example in 2025 the gap between Social Security tax revenue and Cost is $68 billion. Okay a substantial sum of money but only 24% of the actual interest earned that year. In 2035 we are looking at a bill of $159 billion or roughly the same as the amount we are apparently raising effortlessly for this years stimulus package and still only 48% of the interest accrued that year. But by 2045 the needed transfer is down to $104 billion. In 2055 down to $40 billion and finally in 2065 Low Cost has Social Security back with a cash surplus of $18 billion.

Which tells us what? Well a couple of things. First the fact that we never need to pay more than half of the interest owed in any given year means that the effective interest rate on the Special Treasuries is much lower than the nominal rate during the payback period from 2023 to 2060, in fact at that latter date we will be paying interest at a rate of .15% or about 40 times less than the nominal 5%. But it is the second fact that is the most mind-blowing. Under Low Cost there is never any need to pay back any of the principal. None, zip, nada. If you look at the numbers of Intermediate Cost you see the functional equivalent of a fully amortized loan, that is interest only to 2023 and then totally paid down by 2041. Under Low Cost that loan in translated into interest only from about 2027 to 2061, at ever dropping rates, and then can if we choose simply be written off. Now whether the people living in the years after mid-century choose to take this course or not is not under my control, I won’t be caring either way by that point, but that future outcome has some real world implications for the year 2008.

Back to Table VI.F7. When we talk about current and projected debt at levels around $9 trillion and rising we normally concede at some point that we will have to pay it all back, which leads people to cluck their tongues about the Present Value of all that debt. But at this point in time a full 24% of that outstanding debt is represented by the Special Treasuries in the Trust Fund, and that percentage is scheduled to rise sharply over time. Under Intermediate Cost assumptions Intragovernmental debt for Social Security rises to over $5.5 trillion in non-inflation adjusted dollars by 2023 (about $5 trillion in constant dollars) and all has to be paid off with additional cost for interest over an eighteen year period. And almost all of this debt repayment occurs within the current term of the 30 year bond. Which means that to one degree or another it is already being priced in with the expectation that Treasury will have to go hat in hand to the public (and the Chinese Central Bank) to borrow that $5 trillion plus interest. Unless if doesn’t have to, under Low Cost all that is required is a much smaller amount of borrowing to pay off only a portion of the interest owed on the Trust Funds.

I’ll leave it to better arithmetic minds to calculate the discount. But a lot of our thinking about current accounts, the affordability of tax cuts, the strain represented by future medical costs is colored by the prospect of having to pay down trillions in debt to Social Security over the next three decades and still coming out with a benefit cut in 2041. Under Low Cost the need for that pay down almost vanishes, trillions and trillions of dollars we have mentally committed to Social Security simply moves to the other side of the ledger. In the Brave New World of Low Cost things are possible that we haven’t dared imagine.

Over to you.