BW on Soc Sec VIII: Calculating the Cost of Inactivity

The following is a reworked version of a piece originally put up immediately after the release of the 2006 Report in response to what seemed a very worrisome development. After a long series of years of continual improvement in Social Security long term outlook, suddenly progress stalled and actually reversed, the payroll gap went from 1.89% in 2004 to 1.92% in 2005 to 2.02% in 2006. Cause for concern? Absolutely. Cause for action? Well lets do the numbers.

Can we quantify the price of inaction on Social Security? My starting point is this table from EPI Changes in Trustees Projections Over Time.
Note these are not EPI numbers, these are official numbers from the Annual Reports: “Source: Annual Reports of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds, 1996-2004.” With the help of reader Paul in comments I can now present a series that is filled in and longer.
Report Date:::Payroll Gap:::Trust Fund Depletion
1996:::2.19%:::2029
1997:::2.23%:::2029
1998:::2.19%:::2032
1999:::2.07%:::2034
2000:::1.89%:::2037

My fundamental views about Social Security crystalized in the 1999-2000 period. Anyone looking at this number series and drawing the conclusion ‘Man we got to start moving on this crisis, and soon’ is simply innumeric. Maybe something would need to be done down the road but as it stood Trust Fund depletion was retreating at a rate of two years per year, suggesting it might never come. Now the Bush years and a more complicated story.

2001:::1.86%:::2038 (Reader Paul supplied these, and the series after 2004-thanks Paul)
2002:::1.87%:::2041
2003:::1.92%:::2042
2004:::1.89%:::2042

This is where President Bush entered the battle. Clearly progress was being made, Depletion was still being pushed out in date and this even true in 2004 when taken by month, I think there was about a nine month improvement over the course of the year. This progress didn’t show up as much on the payroll gap side, but properly understood this is all the more cause for Inactivity. Doing Nothing in Bush’s first term was the effective equivalent of a tax cut averaging 1.89% in each year, and all from the first dollar and not the last, for wage workers the Cost of Inactivity was money left in their pockets. Absent a better deal than what they were being offered they really had no reason to move. As it turns out Bush’s implied preference, the Posen Plan did not in fact offer that deal, a point that was hammered home by Lee A. Arnold in a clever animation Econolanguage: Social Security now available on YouTube.

2005:::1.92%:::2041
2006:::2.02%:::2040 (this is what percipitated the early version of this post, that was not pleasant news)
2007:::1.95%:::2041
2008:::1.70%:::2041

Now let the fun begin. My oh my plenty of numeric fun to be added. Below the fold.

Lets take this as if it were a betting game, me betting my current dollar against some lost dollars going forward. Now clearly in any year that shows an absolute improvement in either magnitude of the gap and/or Depletion moving out in time is a winning bet for Nothing. As are a series of years with that same result. For example we could take it by term:
1996:::2.19%
2000:::1.89%
2004:::1.89%
2008:::1.7%
Winner, winner, chicken dinner!! Each term left four years of taxes not taken in peoples wallets and ended up with a cost per year going forward that was equal or less that the starting point of the term. The people who insist that ‘There is no time to wait’ ‘Sooner is better than latter’ and ‘If we delay, if will only cost more in the end’ in fact have no numeric evidence to back any of that up. Now certainly it is possible for this trend to reverse, in fact the outlook got considerably bleaker in the mid-ninties, things looked a lot better in 1993 then they did subsequently in 1997, which in the event showed ‘crisis’ at its peak. But that is just where the game gets interesting.

Lets say the 2009 Report shows a jump back to 1.95%. Does this make Nothing a losing bet, should we act this year after all? Not necessarily, you have to calculate the Cost of Inactivity for each individual. And moreover decide over what period you want to measure that. Moreover the effects are not necessarily reciprocal, at least over the short run. If the payroll gap actually goes up due to worse than projected current year results, that simply suggests the previous ‘fix’ would not in fact have fixed the entire gap, whereas when the gap moves down that current year dollar is real as real. But for the sake of argument lets say you could show that the entire uptick was due to failure to collect that one current year payroll dollar. How many years would it take to have that become a financial negative for any given individual. Well if you look at a situation where you have a level real wage when measured in constant dollars the calculation is pretty simple:

Amount of increase multiplied by years to retirement minus previous year payroll gap. For me in 2005 that worked out to:
.03 x 17 = .51 – 1.89% = -1.38%. That is over my working life Nothing in 2005 still worked out as a tax cut, just a little less than I might have hoped for. How about 2006?:
.10 x 16 =1.6% – 1.92% = -.32% Well kind of nets out as a skinny tax cut but one that works to one degree or another for everyone within 20 years of retirement. And of course the improvement from 2.02% to 1.95% meant Nothing was an absolute win short term and equally medium term. Take any year you like and multiply its payroll gap against the number of years between then and now. That is your cummulative tax cut equivalent for that span. Then use a variation of the equation to calculate how long it would have to be before that became a losing bet. The figure rapidly hits the hundreds of years.

Lets turn the bet on its head and look forward. What level of increase in payroll gap would actually make Nothing a losing bet? The calculation is a little more simple: Current year payroll gap divided by years to retirement or in my case 1.7% divided by 15 or .11%. Could the gap increase by that much? Well maybe, the jump in 2006 was close to that but even then I break even. Now if I was 22 and starting my first job I might take a little different attitude. I didn’t see any benefit from the dollars not taxed. On the other hand if I was 28 and still struggling to pay off my student loans that roughly 2% a year not taken out of my paycheck probably looks pretty good.

From my perspective any year where the payroll gap is less than the 2.23% of 1997 is a call for Nothing as a plan. Certainly we want to take a hard look if the numbers start trending upwards year over year, but as long as the number is fluctuating around the 2% level there is really nothing to worry about. Once you calculate the Cost of Inactivity against the current utility of that payroll dollar. After all doing nothing in 2008 will leave the median household with about $1000 or close to the amount of the current stimulus check. ‘Fixing’ Social Security would then be the fiscal equivalent of taking away the punchbowl right as the party was scheduled to start. Nothing smart about that.