Social Security In a Time of Recession

by Bruce Webb

A couple of commenters here have inquired about how Social Security would fare in a time of extended recession. And the really short answer is ‘not too badly, mostly it self-adjusts’. But the questions reveal a profound misunderstanding of the real nature of the Trust Funds and their relation to Social Security solvency. So before talking about the self-adjustment we need to step back a bit and consider what the Trust Funds really are.

Most of the confusion comes from thinking that the Trust Funds are an investment instrument. Which confusion is compounded by the belief that this ‘investment instrument’ was created in the course of the 1983 Reform as a temporary measure to at least in part handle the challenge of Boomer retirement. Neither is true.

Taking the latter point first. The Trust Fund was established with the original Social Security Act something that can be easily seen by the heading of the first Report to Congress in 1941.

Board of Trustees of the Federal Old-Age and Survivors Insurance Trust Fund,
Washington, D.C. January 3, 1941

We have the honor to transmit to you the First Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance Trust Fund, in compliance with the provisions of section 201(b) of the Social Security Act, as amended.

The Trust Funds have grown and shrunk over time when expressed as a ratio of reserve to cost and have always earned interest but were never, ever intended to be an investment instrument, not in 1936, not in 1983, and not now. Which makes arguments about ROI irrelevant in context, people are letting the earnings tail wag the Pay-Go dog. To get a better idea of what the function of the Trust Fund really is you need to follow me under the fold.

The Trust Funds are best thought of as two interest earning savings accounts. Each month they are automatically credited with receipts from taxation of wages and benefits, and each month they are debited for benefits paid out, and then are credited again for interest on the balance giving in turn a new balance. And really that is all the regular taxpayer beneficiary needs to know, the money is in the bank and guaranteed by the Federal Government just as your savings are in a FDIC bank. In normal times people need not worry about how the bank handles the money, even if it goes broke your holdings are insured. Well the same is true for Social Security, the Trust Funds operationally being just savings accounts insured by the government.

But of course these are not normal times and banks are failing at a rapid rate. Moreover those that still survive are choking on instruments routinely described as ‘toxic waste’. So it is no wonder people worry about the security of the instruments backing up the Trust Fund. Are the Special Treasuries that actually make the Trust Fund up real? And are they safe?

Well yes and yes. Largely because what is by many seen as a weakness is actually their strength. Special Treasuries are like regular Treasuries notes that have fixed face values and interest rates. Unlike regular Treasuries they are not marketable. Which leads some to claim they are not real ‘assets’ at all. This is to get things backwards. Because the Social Security Trustees always hold these notes to maturity their value is always and exactly the face value and the yield is exactly equal to the stated interest rate. Which means that the existing portfolio is sheltered from fluctuations in the outside bond market. You could sum this up with the phrase ‘No market. No mark to market.’ The Trust Funds as such cannot be wiped out absent the total collapse of Full Faith and Credit of the United States. Which maybe could happen but would leave us with much bigger things to worry about than future retirement security.

This doesn’t mean that the Trust Funds are totally immune to economic downturn. Each year a portion of the Trust Fund portfolio of Special Treasuries matures and if Social Security is in surplus immediately reinvested in new ones at current interest rates. If we examine the actual portfolio of in this case OAS (Old Age Survivors) we will see that bonds that matured in 2008 had interest rates ranging from 3.5% to 7.25% Table VI.A5.—Assets of the OASI Trust Fund, End of Calendar Years 2006 and 2007[In thousands]. To the extent that these had to be replaced by new ones with lower rates than predicted when the year began then Social Security takes a mild future hit. But only on the difference in rates on the relatively small portion of the portfolio that matured last year, none of this has any effect on say the $27,311,591,000 in 2009 bonds that are still yielding 7.25%.

But even that is to obscure the story somewhat. In 2008 Social Security ran a cash surplus of about $80 billion dollars from the excess of receipts from taxation over cost. To which we would add about $100 billion in accrued interest for a total increase in the Trust Fund of $180.9 billion. This compares to a original projection of $196.2 billion. So we have $15.3 billion less in principal than expected plus some new additions to the portfolio with a somewhat less than expected yield on bonds issued over the course of the year. But $15.3 billion amounts to just 0.6% of the existing Trust Fund balance. Even if we do hit 9% unemployment for most of 2009 that would probably mean no more than roughly triple that amount less in projected year end balances. That is against an Intermediate Cost projection of $2.647 trillion we might end up with something closer to $2.59 trillion. Which is still a Trust Fund in overall surplus, something that is maybe not as likely for your 401k over the same period.

As long as Social Security is in cash surplus the variation in earnings on the Trust Fund balances is mostly irrelevant, it has some implications for solvency in around 2040 but really nothing for current recipients to worry about and not really much for people approaching retirement. Even an extended period of slow growth doesn’t change that fact much because Social Security still is what it almost always has been, a Pay-Go system.

Now as to self-adjustment. A normal recession is marked both by low inflation or even deflation and higher levels of unemployment. And the latter might well have an effect on real wage. But these interact in complex ways when it comes to judging solvency in a Pay-Go system.

Unemployment directly reduces revenues in exact proportion to wages not paid, that is 12.4% of payroll dollars not collected.

Flat real wages also reduces revenues when compared to baseline assumptions. But given that initial retirement benefits are set by real wage over the worker’s work life, flat real wages reduce future costs.

Low inflation reduces costs short term in that current benefits are adjusted annually for inflation. It also serves to lower costs long term because it lowers the benefit baseline that will exist at the end of the period of inflation. While it is possible that a sharp and continuing period of growth after recovery would add those costs back in, it might not. For example the strong improvement in Social Security outlook experienced in the late 1990s was the result of a fortuitous combination of high employment and low inflation that put revenues above projected baselines and costs below.

Under normal conditions the various formulas that set initial benefits to real wage and continuing benefits to inflation combine to deliver a pretty steady state outcome for Social Security when taken in terms of income replacement. Social Security recipients now and in the future see their fortunes mostly rising and falling as the economy does, benefitting from or suffering somewhat in much the same way their grandchildren do.

But let me finish with a caveat. The late 90s were great for Social Security-high employment and real wage increases combined with low inflation took an outlook that in 1996 still merited the word ‘crisis’ or at least ‘pending problem’ and mostly erased it for good. On the other hand the late 70s took a Social Security system that had been pretty stable for decades and almost put it into the ditch. This was due to stagflation where the combination of high inflation increasing costs and slow growth decreasing revenues put Social Security right on a path for Trust Fund depletion. If we enter a new period of stagflation triggered by something like another oil shock then all bets are off. But then again your equity investments will probably not be looking rosy either.

So Social Security can well weather even a prolonged recession. But neither it or us will fare well in a perma-depression. A lot depends on whether the various Obama plans for banks/stimulus/universal coverage/energy independence work. Which is where everyone should be focusing their attention and not on some time-wasting ‘Entitlements Summit’.