Social Security 2027: A date for action?
I spend a good deal of time talking about ‘Nothing’ as a plan for Social Security. You can make an excellent numeric case for ‘Nothing’ in the short run VIII: Calculating the Cost of Inactivity and you can make a reasonable economic case for ‘Nothing’ in the long run II: The Shape of Low Cost. For that matter you can make a pretty good case that even if the Trust Fund does run out at some time in the future that it may not be that big a deal. For example the CBO just released their most recent projection last week in Updated Long-Term Projections for Social Security and tell us (bolding mine)
CBO’s projections indicate that future Social Security beneficiaries will receive larger benefits in retirement— and will have paid higher payroll taxes—than current beneficiaries do, even after adjustments have been made for inflation and even if the scheduled payments are reduced because the trust funds are exhausted.
This is because the schedule of benefits is set up so that initial benefits track the overall growth of real wages over your work life, in this case the rising tide DOES lift all boats.
But I have to admit that ‘Nothing’ is not a particularly satisfying plan. It is all well and good to say ‘So what 78% (Trustees) or 84% (CBO) of the scheduled benefit is better in real terms than what my Mom gets today’ but somehow that leaves something lacking. So I want to propose an action plan based on real numbers and set a time when we might want to move on Social Security in an effort to either eliminate the projected benefit cut or smooth and extend it out. Details and numbers under the fold.
I attempted to make the case in XXXVIII: Financing Shortfall that the doom and gloomers who insist that paying back the Trust Fund will require massive tax increases or huge program cuts or unsustainable borrowing are simply ignoring the numbers, in inflation adjusted terms they never approach typical Bush II level unified deficits:
all these numbers represent transfers from the General Fund. Numbers are in billions.
Year:::::Current:::::Constant 2008 dollars
Now there is no doubt that given reasonable control on the General Fund side that we could afford to finance these numbers. On the other hand we have other societal needs. So I suggest we pick a target figure of $200 billion in 2008 dollars as the limit for what we will willingly want to transfer from the General Fund (or roughly the one year cost of the current wars). Which suggests a target date for implementation right around 2026-2027. Which suggests firing up the policy shops around 2023 (my scheduled date for retirement, maybe I can land a consultant gig).
Now lets turn around and see what current law requires. The Trustees use two different official measures to track the health of Social Security. One is Long Term Actuarial Balance which is the gap between future income and future cost reported as the amount of payroll tax needed to deliver 100% of the scheduled benefit, or that amount expressed as a percentage of GDP, or as a Trust Fund ratio all over a 75 year window. The official target is a TF ratio of 100 which is to say a TF balance equal to one year of projected cost in each of the next 75 years. The second measure is Short Term Actuarial Balance which is always reported in terms of TF ratio over a ten-year window. If the Trust Fund is projected to have a TF ratio of 100 or more in each of the next ten years then the overall system is deemed in Short Term Actuarial Balance. If the TF fund is projected to fail that test the Trustees are mandated to urge Congress to act to bring it back in. Which is to say that Short Term Actuarial Balance is the canary in the coal mine.
Currently the Trust Funds individually and combined are in Short Term Balance so under the law we can (and in my view should) wait and see what eventuates. But under current projections when would the system fail the actuarial test?
Table IV.B3.—Estimated Trust Fund Ratios, Calendar Years 2008-85[In percent]
Well a little interpolation shows us going below a TF ratio of 100 in 2036 meaning that if outcomes do in fact come in in line with Intermediate Cost that the Trust Funds will fall out of Short Term Actuarial Balance in 2027, year ten being the trigger.
So both from the financial-economic side and the legal-technical side we end up with an action date under IC projections of 2027.
Now this date is obviously fluid and has been moving out in time. (And if we examine Low Cost we see that the system never fails the test and in fact may need to have its taxes cut back a little to keep the TF ratio at least under 400 and ideally back towards a sustainable 100 over the long term.) But at least it is something to hang our hat on. When anybody asks ‘What is YOUR plan for Social Security’ the answer is ‘We are targetting the official measure of Short Term Actuarial Balance. If it looks to be failing that test four years out we will start planning with an implementation date of the year the system falls out of actuarial balance. Meanwhile we will simply monitor the numbers as they come out.’
Sure a little wonkish but it would show that supporters of Social Security are not just working on a hope and a prayer but instead imposing realistic tests on the system. My personal belief is that the date the system is projected to fall out of actuarial balance will continue to recede, but belief doesn’t buy you much. On the other the numbers are speaking to us and they are telling us we need to do nothing about Social Security before 2027 and plan nothing before about 2023. Meanwhile we have plenty of serious problems in other areas to worry about.