CEPR - Center for Economic and Policy Research


En Español

Em Português

Other Languages

Home Publications Blogs Beat the Press Bad News on Social Security: We're Going to Live Longer

Bad News on Social Security: We're Going to Live Longer

Sunday, 06 January 2013 06:28

That's what Gary King and Samir Soneji tell us in a NYT column this morning. The gist of the piece is that the authors have assessed trends in mortality rates from a variety of factors and concluded that the Social Security Administration is underestimating life expectancy. Therefore the program will cost more than is projected, meaning that the long-term funding gap is larger than projected.

Before dealing with the scary prospect of living longer let's first address some trivia. The piece tells readers:

"For the first time in more than a quarter-century, Social Security ran a deficit in 2010: It spent $49 billion dollars more in benefits than it received in revenues, and drew from its trust funds to cover the shortfall."

That's not exactly right. The program spent more than it received in payroll taxes, but Social Security also earned more than $117 billion in interest on the government bonds in the trust fund. This means that the program actually had an annual surplus and the trust fund grew in 2010.

But let's get to the crisis of living longer. Based on their projections of life expectancy, King and Soneji calculate that in 2031 Social Security will cost about 0.65 percentage points more than the trustees currently project measured as a share of taxable payroll. This comes to 0.25 percentage points measured as a share of GDP.

Should we be scared by this? Well, the amount is certainly not trivial, but the increase in defense spending associated with the wars in Iraq and Afghanistan came to 1.7 percent of GDP. So, we have dealt with much bigger expenses without too much disruption to the economy. So if King and Soneji's projections prove accurate, Social Security will not exactly be breaking the bank.

However there is a bit more to the story. They only dealt with the impact of improving health on life expectancy. There are other ways in which better health can be expected to affect the finances of the program. For example, the disability portion of the program currently accounts for almost 18 percent of the program's cost. If better health reduced disability rates then this could go a substantial portion of the way toward offsetting the higher costs associated with a longer period of retirement.

The second way in which better health could affect Social Security projections is by allowing people to work later into their life. A substantial portion of retirees are forced to retire due to poor health. If these people were in better health, many workers might put in more years of work before retirement, thereby improving the finances of the program by increasing tax collections.

Better health might also mean slower growth in health care costs. One drain on the Social Security system is the money paid to workers in the form of employer provided health insurance. This money, which has been a rapidly growing share of compensation, is not subject to the Social Security tax. If better health reduces the rate of growth of health care costs, a larger portion of compensation may be subject to the payroll tax, which would also improve the program's finances.

Finally, improved health would likely reduce the cost of other government programs like Medicare. This could means that we will be paying out more money in Social Security to retirees but paying less for their Medicare and Medicaid expenses.

All these effects may not be entirely a wash, meaning that our longer lives will mean more net expenditures from the government, but we would want to look at all these factors before we hit the panic button.  

Comments (17)Add Comment
written by Chris Engel, January 06, 2013 6:28
That's the bad news.

The good news is it's entirely affordable to support this demographic change with a sensible tax/regulatory framework in the economy that reduces inequality.
Extending Retirement Age
written by Bart, January 06, 2013 6:45

Once again some desk-bound researchers propose raising the retirement age as high as 69 or 70; failing to explain how roofers, carpenters, etc., will be able to work that long.
Dean, a question about your budget numbers
written by pjm, January 06, 2013 8:59
D., generally you make the case about the "non dire-ness" of the budget situation including the point that the budget was well under control prior to Great Recession, which is no doubt true. My question (and I don't mean to imply that the answer is, as the lawyers say, dispositive) is: to what extent was the rosiness of the pre-recession fiscal situation due to the boom of the housing bubble itself? I.e., how much worse would the budget have looked (given Bush tax cuts, wars etc) would the deficit look without the bubble? Does it matter?
disability, working old
written by traktorina, January 06, 2013 10:11
According to Glaeser, current (and future) disability rates would appear to be uncorrelated with life expectancy:

The whole reason we have Social Security to start with is the national tragedy of unemployed elderly that emerged during the great depression (see Kennedy's Freedom from Fear).

Declining rates of population replacement will also put a squeeze on entitlement funding. This argues pretty strongly for fiscal economic stimulus to stimulate growth, but deficit hawks are prevailing for now. This is just a re-run of Japan in the 90s.
What about Productivity?
written by A Populist, January 06, 2013 10:57
As Dean has noted many times, productivity growth (if it is shared) will tend to offset the problem of low end workers paying for their retirement - whether through savings, or increased SS money. And, BTW: Since SS is funded by a payroll tax, if the gains from productivity are shared through higher minimum wages (and other wages earned by large numbers of workers), this could have a large and significant impact (positively) on SS funding through the payroll tax.

And then there is the potential problem that (all other factors equal) our increased productivity will (unless demand is increased) lead to more unemployment. Unless demand is increased, or workers have some financial support for leaving the workforce or working fewer hours: How can we be sure that there will be jobs for all these people who are asked to "work 'till they drop"?

OK, all else isn't equal - but productivity is a very strong countervailing factor, so any serious and balanced discussion must include it's effects.

As a regular reader of NYTimes, I have been seeing a pattern.

A few weeks back, they ran some articles talking about the decreases in productivity "growth" - but they were talking all gloom and doom - as if productivity *itself* were actually decreasing - and not just productivity *growth* decreasing.

And now, this latest piece, which seems to be doing the work of those who want to cut SS.

Apparently, these guys don't understand exponentials and compounding, and how even a very small (but positive) growth in productivity continues to provide benefits (and problems - if it is not shared).

For a newspaper which gets smeared with the "liberal" label as much as the NYTimes does, they are doing a really poor job of proposing ways to improve SS and maintain benefits - or even presenting a balanced view. There does not seem to be a mainstream news outlet in existence (liberal or otherwise) that does not seem bent on cutting SS, even though it is not necessary.
written by Eideard, January 06, 2013 12:40
Remove the cap on SSA contributions. For those above the point of change, it simply feels like every pay period before. And takes us into the 22nd Century.

Only politicians and demagogues need to make it more complex than that.
Harvard and Dartmouth
written by aldole, January 06, 2013 5:19
One factor that might be considered is new research suggesting that retirement itself, although popular, may reduce life expectancy by breaking lifelong routines and disrupting deep social connections. One might question how much government policy should actively encourage retirement, as opposed to merely making it an option.

I can't believe that these "researchers" would actually waste their valuable column inches on this boneheaded notion. How exactly does the government "encourage" retirement. By cutting jobs and accepting 8% unemployment as 'structural' perhaps? By all means, lets have some more options, please.
are we at least as "humane" as Neanderthals?
written by watermelonpunch, January 06, 2013 5:23
When things like this are discussed, I just wonder... what exactly is anyone thinking should be done instead of making sure seniors & disabled will be able to live?
If you consider cutting social security & medicare, or doing away with them, what are you saying you want to happen to people when they are no longer able to work?
There's evidence that even Neanderthals cared for their sick & disabled!!

Puzzling Commentary Worthy of Mitt Romney
written by Ron Alley, January 06, 2013 5:34
The text of the op-ed piece by King and Soneji repeatedly asserts that the Social Security Administration's actuarial forecasts are inadequate but fails to state precisely where and how the actuaries have gone wrong.

The accompanying graphic attributed to Bill Marsh and citing King and Soneji contains the following criticism in its point 4:

AGES 55-59 The chart predicts that everyone who happens to be 55-59 in 2028 would die. (Yet half of those who are age 95 that year would live on.)

Yet in item 5, entitled "So: People Will Live Longer", suggests that if the King-Soneji method had been used people aged 53 to 57 and 58 to 62 in 2030 would live on the average only 14.2 months longer than forecast.

Does this mean that the actuaries who concluded that all the 55 to 59 year old recipients in 2028 would die within one year were only 14 month off?
You don't quite see.
written by OJCsr, January 06, 2013 7:43
Dean says "..allowing people to work later into their life.."; he's not recognizing that "work later in life" is most often not a voluntary choice. Layoffs happen, physical or mental deterioration does occur, not to mention plain bad luck. Denying SocSec benefits to those so caught does save money -- more people die of privation before SocSec cuts in.
An insolvent Trust Fund?
written by Jack, January 06, 2013 7:50
I found this particular sentence in the article deserving of special examination. "If the amount of money coming in through payroll taxes does not increase and if the amount of money going out as benefits remains the same, the trust funds will become insolvent less than 20 years from now."

Insolvent? Maybe King and Soneji were just a little careless with their terminology, but they've written for publication in a major news journal. Carelessness is unacceptable. Or maybe they just don't understand the structure of the SS system and the Trust Fund component as well as they comprehend statistical measurement. The SS system will continue to have significant revenues from worker contributions, the so-called payroll tax. Insolvency is defined as "not solvent; unable to satisfy creditors or discharge liabilities, either because liabilities exceed assets or because of inability to pay debts as they mature." The Trust Fund may become depleted in 2033, but the system will continue to have revenue. That's not insolvency.

And what of their suggested solutions? They are the same old canards being tossed around by the Peterson crowd and their cohorts such as Bowles and Simpson team of destroyers.
written by JDM, January 06, 2013 9:19
I for one welcome the contributions of these pundits: how else would I have learned that Mr. Spock's "live long and prosper" was a curse?
Do unto others...
written by Jack, January 06, 2013 10:38
One last thought regarding the suggestions from King and Soneji regarding delaying retirement benefits to a later age. Two university professors whose jobs it is to read, think and pontificate should try a bit of labor before making suggestions as to how long one should have to work. It's a great life for those who have the stamina to work into older age, or have positions which don't require too much physical exertion. It's a great relief for those others who have spent their energies over a full work life and need a bit of rest before their end point. Next King and Soneji will be telling us how much longer one lives by avoiding the drudgery of retirement.
they only differ by 2 years from SS trust fund!
written by pete, January 07, 2013 9:59

Trust fund says 2033, these guys say 2031. Big deal. Anyway, of course, SS payments are just transfers. They are not a drag on the economy. They really should not affect traditional macro models, since they are not spending on goods and services. Nothing to see here....

The major problem is using inefficient taxes on labor rather than efficient taxes say on carbon as the Irish are doing quite successfully. Government payments to retirees is a fine idea for those in need, certainly not Buffetts and Gateses. But funding those payments should be done efficiently. Now admittedly, the 12.5% flat tax (payroll) is better than a progressive rate, incentive-wise. And of course the new health care head tax is even better! Whowouldathunk?
written by Alan Barnes, January 07, 2013 10:22
So, King & Soneji say that we should raise the retirement age because longer lifetimes threaten S.S. solvency, but "...[there]is new research suggesting that retirement itself ... may reduce life expectancy ...." A bit of a contradiction, no?
Trust fund interest is an asset, but not cash income
written by Don Levit, January 07, 2013 4:21
Trust fund interest is an asset to the SS trust fund, and a liability to the Treasury, so it is a "wash."
The only cash flow to the fund is non-interestr income.
This is why even though the trust fund balance grows, its cash dedficit grows as well, for the trust income of $117 billion is backed out of the cash flow equation.
"The net cash flow during any period is total non-interest income less total outgo."
Don Levit
yes indeedy don...
written by pete, January 08, 2013 10:48
Interestingly, the interest rate is not a market rate, but a fictitious rate, somewhat like the rate that federal employees get on their "G" fund. So, if Dean really believes that the interest on the fund is real, then congress could easily solve the SS problem by raising this interest rate!

Write comment

(Only one link allowed per comment)

This content has been locked. You can no longer post any comments.


Support this blog, donate
Combined Federal Campaign #79613

About Beat the Press

Dean Baker is co-director of the Center for Economic and Policy Research in Washington, D.C. He is the author of several books, his latest being The End of Loser Liberalism: Making Markets Progressive. Read more about Dean.