Most of what almost everyone knows about Social Security finance is wrong. Not YOU of course, I mean the other guys. But as a favor to me look over my shoulder as I edicate the wrong'uns under the crueller and make sure I don't make mistakes.
Alrighty. The above two graphics form a portion of Table VI.A1 of the 2012 Social Security Report. And from this blizzard of dates, numbers and labels there is a lot to learn. Which in turn will inform discussions of Social Security solvency going forward.
Some lessons: The Social Security Trust Fund, or at least the OAS (Old Age Survivors) Trust Fund, is almost as old as Social Security itself being established on Jan 1, 1940 pursuant to the Social Security Amendments of 1939. The belief that the Trust Fund was an artifact of the 1983 Greenspan Commission is a myth. As is the less wrong belief that the 'prefunding' established in 1983 for Boomer Retirement was an innovation. That too is not true.
Table VI.A1 reports four sources of Social Security income plus total Income, three outflows of Social Security Cost, the difference between Income and Cost in 'Net Increase During Year', the subsequent Trust Fund 'Amount at End of Year', and an expression of that amount in the form of a Trust Fund Ratio. One thing to note right away is that no single column reports total cash flow, nor can it be calculated by simple subtraction, and it certainly doesn't equate to 'Net Increase During Year'. That is both 'Income' as defined and 'Net Increase' include 'Interest' that may or may not come in cash form. Suggesting that focusing on cash flows as a measure of Social Security health is likely to lead one down the wrong path, that just isn't how the Trustees measure 'Solvency', which instead is a technical term of art. Something with a lot of consequences.
That the final column represents 'Trust Fund Ratio' is not an accident, properly considered this is THE number that determines 'Solvency' as defined. For the Trustees of Social Security the Trust Fund is only 'solvent' if its total year end principal equals or exceeds next years actual or projected Cost expressed as a percentage of that Cost with 100=1 year. By this measure Social Security was hugely solvent from 1936 to 1950 with balances exceeding 10 years of next year cost, very solvent from 1950 to 1960 with balances near or above 2 years of next year cost, just solvent from 1960 to 1970 with TF Ratios clinging right to the 1 year/100 level, and actually insolvent from 1971 to the crisis point in 1982 and the impetus for the Greenspan Commission.
It is crucially important to understand that 'Solvency' is not directly related to 'Net Increase in Assets', Social Security can be fully solvent in years where TF balances actually drop (e.g. 1959) while moving from solvency to insolvency in a year when TF balances actually increase in nominal terms (e.g. 1971).
Returning to the Table we can see some truly astronomical Trust Fund Ratios in Social Security's first thirteen years, in fact in 1939 Social Security held assets totally 80 years worth of 1940 actual cost and even in 1950 the TF Ratio equated to 11 years of 1951 Cost. There are two explanations for this. First the Social Security Act of 1935 deliberately pre-funded Title 2, what we know as Social Security today, and originally monthly benefits were not intended to start until 1941. Contrary to popular belief Social Security Insurance NEVER was designed to operate from a negative actuarial position and so-called 'Legacy Costs' are much overstated as a source of current Social Security 'crisis'. That is the 'Backwards Transfer' touted by Hassett and Biggs of AEI is mostly mythical.
Now retirees were not left high in dry in 1935, just told to wait six years for monthly checks, instead the 1935 Act set up a parallel retirement program under Title 1, one that directed funds to State Old Age Pensions. Title 1 was a General Fund financed straight out welfare program and one that paid out more benefits to more recipients right to 1950 than Title 2 FICA funded insurance. Which is the apparent source of another prevalent myth, that FDR intended Social Security to be temporary and to be phased out in favor of private accounts. Instead this is a transferral of the reality that would have General Fund Title 1 be temporary and to be phased out in favor of worker financed Title 2. It is the existence of Title 1 that explains the rapid buildup of Trust Fund Ratios to 1950 and its shrink down to target range by 1965 and rough stabilization for some years after, Title 1 being the invisible pig in the python of Title 2 finance.
But as noted Social Security slipped into technical insolvency in 1971 and its actuarial position got worse and worse with reserves slipping down to just a few months in Reagan's first full year (TF Ratio of 18 in 1981). This was a true existential crisis, by comparison Trust Fund ratios in 2012 were above 300, and the Greenspan Commission was addressing a truly pressing need.
And it is here that myths start creeping back in. The tax increases imposed in the course of the 1983 legislation that emerged from the Commission did not immediately restore Social Security to solvency, nor did they initially generate much in the way of actual cash surplus. Instead they were phased in in such a way that TF Ratios did not get back to target until early 1993 and were only at 40 something with Reagan left the White House.
Meaning there WAS NO REAGAN RAID on Social Security. Under law since 1939 and even in practice before that all Trust Fund balances were mandated to be held in Treasuries or government backed equivalents. And yes this means that any actual cash surpluses in principle (though not really in operation) flow to the General Fund, but it wasn't like either Reagan or Bush 1 had any choice about this, their Administrations simply ,and i might add slowly, were just building up the Trust Fund to its legal MINIMUMS. Moreover on inspection the actual dollar figures were very small, particularly in Reagan's years, since you have to back out interest earnings from 'net increase' to get a measure of funds actually available to fund tax cuts and military buildups. And the 1988 Social Security Report, the last to actually be produced during the Reagan Administration, showed 1987 balances of only $62 billion, up $40 billion from year end 1982, but of which increase some $25 billion came in the form of non-cash interest and tax on upper income retirees while only some $15 billion were "over taxation" or "pre-funding" or "theft" or whatever from worker paychecks. And while $15 billion is real money it pales in comparison with the costs of Reagan's tax cuts and military buildup, the notion that it was 'stolen' or even that the 1983 fix was 'engineered' to facilitate that 'theft' simply falls apart on examination of the actual numbers.
I have a lot more to say on this, but will save that for Comments and later Posts. Over to you all.