"House rich & cash poor": Why Social Security can’t be Raided(Part 2)

(Still cross posted from dKos. But since coberly and I have an extended colloquy there maybe not a bad place to start)
Part 1 was kind of a set-up in both senses of the word in that it didn’t really deliver on the post title. But I think a necessary set-up and so lets resume.

When we left off we had Social Security after having a long period of positive cash flow from 1936 to 1956 and so a lot of pre-funding, leveling off in terms of Trust Fund Ratio through the 60’s, only to go into some decline in the 70s. And if we return to Table VI.A2.— Operations of the OASI Trust Fund, Calendar Years 1937-2010 we can see a system that was by any measure you like very sick by 1981, and much sicker than today. Whereas the year end balance for OAS in 2010 still represented 4 full years of 2011 cost (TF Ratio of 400) the corresponding balance in 1981 was less than a fifth of a year (TF Ratio of 18). Action was imperative and the motivation was not Reagan’s desire to tap into worker pocketbooks to fund tax cuts, as far as the Trust Fund was concerned it was the farthest thing from a piggy bank. Which gets us to our second set-up point: the Myth of the Reagan Raid. Onward and lowward (i.e. below the fold).

The Trustees measure the health of the Trust Funds in terms of Actuarial Balance. A Trust Fund in annual balance ends the year with a TF Ratio of 100 or above. But the Trustees have a longer horizon than just the next year, instead they consider the Trust Funds to be in Short Term Actuarial Balance if they project to have TF Ratios of 100 or more in each of the next 10 years, while they are deemed in Long Term Actuarial Balance is they project to have those levels of TF Ratio in each of the next 75 years.

Now the vast majority of Trust Fund assets since inception of the program have been in Special Issues of Treasuries at times with an admixture of regular Treasuries (but at no point I am aware of ever representing more than a fraction of total assets). The restriction to Treasuries is a direct consequence of language in the 1939 Amendments to the Social Security Act of 1935 mandating that all funds credited to the Trust Fund and not needed for short term benefit payments has to be held in “instruments fully guaranteed as to interest and principal by the federal government”. Meaning it would take a change in the law to have them in any other asset class, and except for short term payment purposes even in cash. While some people see something nefarious in this practice it has served the system well over the years, for example large balances built up over the 40s and 50s were successfully used to bridge gaps between income and costs through the 60s and especially the 70s. The historical record shows that every obligation was honored down to the next to the last penny (and for DI starting again in 2005). Because in the eyes of the U.S. Treasury those Special Issues are just as good as cash, in fact they pay interest on them just as the Federal Reserve does on reserves deposited with it by member banks.

Which brings us to the first point here. There can be no ‘raid’ of Social Security in any year that has a TF Ratio less than 100. Instead the federal government via the Trustees has a positive legal obligation to buy Treasuries to build up that ratio back to a minimum of 100. And if we look at Table A2 again we see that Reagan inherited a Trust Fund that has fallen out of actuarial balance in 1971 (TF Ratio of 94) and never recovered. Indeed without a temporary loan of $17.5 billion from DI to OAS in 1982 checks might have been delayed for the first time ever. But between the loan and the fix installed in 1983 during the process that included the Greenspan Commission, the Trust Fund began to recover. But only just. In fact by the end of Reagan’s second term the TF Ratio was only up to 41 or less than half the target, if anything the argument would run that Reagan should have taxed workers even MORE so that the Trustees could buy even MORE Treasuries. Instead the fix was deliberately phased in with a ten year time table with remarkable success with the TFs finally get back over 100 in the course of 1993. Which in respect to Social Security takes both Reagan and Bush 1 off the hook, when examined dispassionately all they did was restore the Trust Funds to its minimum required reserve. Now as it turns out that reserve was mandated by law to be in the form of Treasuries which by definition meant cash going to Treasury, that is what buying a bond does, you give the government money, they spend it on what they want and give you a promise in return. There is nothing ‘Phony’ about that process, not at all. And certainly no ‘raid’, instead we have the Trustees fulfilling their mandate to restore actuarial balance with any cash flowing to the General Fund being a simple byproduct of that legal requirement.

Okay that settled, back to the topic of the posts. Over the course of the mid to late 90s the solvency of the Trust Fund continued to improve by every measure: TF Ratio, year end balance, and increase in assets. But even then this risks exaggeration of actual cash flow, instead you get that by netting out interest which came only in the form of Special Issues and so were not financed out of the outside economy. In fact in Clinton’s last year 2000 the actual cash flow was $74 billion out of total SS Surplus of $132 billion. And for the most part it never got better than that, though assets in the OAS Trust Fund grew from $931 billion in that year to $2.4 trillion today (in both cases excluding DI) a growing piece of that was simply interest credited to the Trust Fund.

And here is the key point. That interest is an obligation to the General Fund, it cannot by its nature be borrowed. It can and does score as income for both Trust Fund and overall Budget accounting, for those purposes it is considered as real as real. But it can’t be tapped for any other purpose even in theory, in the metaphor expressed in the post Title it is home equity and not cash, and in this case not subject to refinance to extract that cash, which in any case wouldn’t serve to increase net wealth as such.

No it is only actual cash flow above and beyond cost that is available for ‘raiding’ in any sense and examination of our table shows that cash flow stabilized and then dropped sharply after 2008 to the point that even though OAS returned a total $92 billion surplus in 2010 this represented negative cash flow for that program alone of $16 billion. And OAS is by far the healthier of the two OASDI programs, comparable numbers show that DI went cash flow negative in 2005 and actually started cashing in its principal in 2008. This contrasts to OAS whose total Income including Interest exceeds Cost at a rate that will keep its total balance growing until 2023 with assets sufficient to bridge its cost gap until 2038. On the other hand DI’s income imbalance is such that it projects to get totally out of assets by 2017. (Combine those two and you get the 2036 date reported for THE Trust Fund.)

You can’t raid Interest and even though cash flow for OAS was projected to go back positive briefly in 2013 it was never projected to be that much, to run out by 2017 anyway, and at this point probably won’t show up at all.

The history of Trust Fund operations from 1936 to 1982 shows us clearly that Trust Fund Assets are exactly that, assets available to the Trust Fund. But only in years of positive cash flow are those assets ever tappable for other purposes, (for example setting up personal accounts alongside Social Security on which fees can be charged by Wall Street). And the days of strong cash flows are over. Not a crisis, this outcome was fully projected in the 90s, and on balance things turned out better than expected on net. In 1997 the date of Trust Fund Depletion on a combined basis was set to be 2029, we actually gained 7 years of projected solvency since. So there was nothing wrong with the planning, this is the way the Trust Funds were designed to operate. As noted earlier balances built up from 1936 to 1956 were tapped starting in 1957 to bridge the Income Excluding Interest and Cost gap until funds ran short in 1982. And we are just repeating that process today with balances built up from 1983 through now started being tapped by DI in 2005 and OAS in 2010. The difference being that we still have a TF Ratio of 400 in 2010 compared to the 22 TF Ratio of 1982. The Trust Funds are real, they are just no longer stealable. The dollars flows were such starting in 1996 that raiding to fund private accounts almost penciled out, but that window slammed shut by 2004. (The plans floated by Bush in 2005 having seriously cooked the books to make the numbers run, subject for another post.)

Social Security: house rich, cash poor, and unable to be burglarized. Not because the walls are so secure, though they are pretty good, there just isn’t anything inside the house that you can actually sell to a fence.