HOW MUCH INTEREST DOES SOCIAL SECURITY PAY?
By Dale Coberly
HOW MUCH INTEREST
DOES SOCIAL SECURITY PAY?
A reader asked what interest does Social Security pay on your “investment”? I think it is important to realize that this is not the important question about Social Security. Social Security is an insurance program and it doesn’t make any more sense to talk about the “interest” you earn on your “investment” than it would to ask what interest you earn on your car or home insurance. Or grocery budget.
However a person should know what he is paying for his insurance and whether the insurance is worth the premium. In this sense it might be useful to make an estimate of Social Security’s “return on investment.” But such an estimate can be dangerously misleading if you do not keep in mind that you are paying for insurance.
By using the 2013 and 2009 Trustees Reports, I have calculated some “returns” reasonably associated with the “premium”… the “payroll tax”… “invested.” My results compare fairly closely with some standard estimates of SS “return on investment” in the “average” case, and diverge widely in the case of lower earners and what I regard as a more likely pattern of earnings. Because SS is intended as insurance, the “earnings” of the “average worker” are not as germane to the task of Social Security as are the “earnings” of those who earn less than average.
By taking the “Average Wage Index” as an “average wage” over the 35 years that SS counts wages (and therefore taxes paid) for purposes of computing your contribution and benefit…
and assuming the one tenth of one percent increase in the tax, per year, that would be required to pay for the scheduled benefits (discussed at some length in other papers by this author)…
and taking the tax as the premium, and calculating the interest (at a steady rate over the 35 years) on that premium that would be required to generate the amount of money that would be needed, if invested in an annuity at the same rate of interest, to generate an income equal to that of the benefits that Social Security would provide, based on that record of earnings and taxes paid on those earnings, for the life expectancy that the Trustees project for the eventual retiree… the person paying the tax…
The necessary rate of interest would be…
For a low lifetime earner… 45% of Average Wage, about 18,000 dollars per year in today’s terms:
7.23% if you count the tax paid by both the worker and the employer, to pay benefits for for the worker, or
8.74% to pay benefits for the worker plus wife.
If you consider… as i do… that the low income worker would NOT get the employer’s share if the government did not require the employer to pay that share, the worker would have to earn from his share of the tax…
10.9% to get the same benefit for himself (without wife).
These are “nominal” interest, and assume an inflation rate of 3%, so the “real” interest translates to about
4.3% if paid by both the worker and the employer, with no benefit for wife
5.7% if paid by both worker and employer, with benefit for wife
7.9% if paid by worker alone, without wife.
For an average worker… best 35 years at the Average Wage
6% (nominal, 3% real), if paid by worker and employer, no wife
7.5% if paid by worker and employer, with wife
9.9% if paid by worker alone, without wife
For a high lifetime earner, 160% of average earnings, about 64 thousand dollars a year today
5.2% nominal (2.2% real) if paid by worker and employer, no wife
6.8% if paid by worker and employer, with wife
9.2% if paid by worker alone, without wife
And for someone who earns at the cap for the best 35 years (equivalent to over 110 thousand dollars today)
4.5% nominal, if paid by worker and employer, no wife
6%.. with wife
8.6% if paid by worker alone…. in this case it would probably be fair to consider that the employee has enough leverage to get paid the “boss’s share”, if he is not the boss himself, so the “combined” rate above would be more realistic.
There is some room to complain about these results. For example I only count the taxes from the “best 35 years.” On the other hand, I think that assuming someone earns at the cap.. or even “high” wage… or for that matter “medium wage”… every year from the day they turn 22 until they retire at 67… is unrealistic. Even the low earner is likely to have many years of unemployment, for which he pays no tax, so it would not be reasonable to assume “forty five years of taxes” to get a benefit based on thirty five years of taxes.
What is worse for the calculators of “rate of return” is that they ignore the insurance value of Social Security. I have personally known people who were high earners until their business failed, or their skills became obsolete, and by the time they retired their lifetime average earnings were not much better, if at all, than medium or even low earners…. in which case they get the “interest rate” of those lower earners; that is they get a higher interest rate than they would if they had stayed lucky. That’s why it’s insurance.
So it may come down in the end to an “ideological” dispute between those who are sure they can drive a hundred miles an hour “safely,” and enjoy the thrill, and those people who are reasonably sure they can’t… or that not everyone can… and therefore prefer to have speed limits on the public roads. As well as require insurance to drive on them.
At the end of the day, the high earner is not out any money… he gets benefits fully equal to what he paid in, adjusted for inflation, and with a small but not unreasonable effective interest on top of that….. and he gets to where he is going in just about the same time as it would have taken him if he drove the speed limit. Maybe faster, considering the time taken by traffic jams cause by wrecks. And without Social Security there will be wrecks. (sorry if this metaphor is incomprehensible to you. it makes perfect sense to me.)
I calculated the “Present Value” of those taxes and those returns. I found that the present value is critically dependent on the “discount rate” that you choose. A lower rate will get present values of benefits higher than present values of taxes. A higher rate will get the opposite.
This leaves it up to the present value people to explain just where in the real world the Social Security “taxpayer” can get an interest rate that will without fail yield the return he gets from Social Security. Plus the insurance he gets from Social Security, against inflation, against market losses, against his own failure to make enough money over a lifetime to save enough … even at the Magic Present Value Bank… to be able to afford to retire. That way they don’t have to explain why present value is a useful measure for an insurance program, or a grocery budget… which is what Social Security is. Or how the Magic Present Value Bank is going to avoid attendant unpleasant social costs… that is, personal costs to many or most people…. that will follow from large numbers of people reaching old age without enough money to retire. And remember, it’s their own money. It’s not welfare. They paid for it themselves.
For those who haven’t figured it out… the “interest” comes from growth in the economy so that, in principle, each generation can pay the same payroll tax rate and yet provide more money to the retired generation than that generation paid in. This is not theft. The younger generation will get the same deal. And it works exactly the same as ANY investment. Except that the Social Security “stock” is invested in “The United States of America,” the prosperity of which you have something to do with.
I would not take my interest rate calculations too seriously. But I hope you take the lesson not to take “their” interest rate calculations too seriously either. There are too many variables for any calculation to be very useful in the end. What matters is what other “investment” plan will guarantee that you will have “enough” to live on if all else fails? At a rate that most sane people don’t even notice.
Coberly and I have disagreed philosophically about the benefits of SS. But his numeracy with regard to the program is beyond reproach and this is no exception, as his numbers are reasonably aligned with the SS actuaries:
http://www.ssa.gov/OACT/NOTES/ran5/an2006-5.html#wp152648
“so it would not be reasonable to assume “forty five years of taxes” to get a benefit based on thirty five years of taxes”
My impression is that SS detractors are mostly people who have not seen too many bumps in the road and therefore don’t see the value of insurance. He may really have worked for 45 years. It does make him harder to convince.
Arne
I may have worked 45 years too, but in some of those years i did not make enough to have “saved” anything that would matter in the end.
certainly would not have saved anything if i had been making the decision myself. even the SS “tax” couldn’t save enough out of what i made to make a difference… and at least I didn’t have to think about it.
not having to think about it is one of the great beauties of Social Security, but I doubt I could convince any of the great economic thinkers here of that.
A recent title of a post from ‘those who shall remain nameless’ blasted SS recipients getting more than they payed in, as if the goal was to always make sure they would get less.
Lord
it is not for nothing that i call them Big Liars. during the big Bush push for privatization they were complaining because SS did not pay enough.
any lie they think you will believe.
MJed
thanks
reasonably close is as close as we can reasonably expect to get.
as for philosophy… when i was seventeen i could go a hundred as well as the next guy.
I agree with much of what you say, but seems to me that real rate of return for each cohort is the most objective way to evaluate the system.
Of course one needs to keep in mind that SS is in some sense insurance (an annuity has both investment and insurance aspects), and that one’s individual rate of return is not truly known until they die, but as citizens, SS participants, and voters we should know what the rate of return is for each cohort in the past and project rate of return for each cohort going forward.
From M.jed’s earlier post: http://www.ssa.gov/OACT/NOTES/ran5/an2006-5.html#wp152648
Going forward it seems best to me to just bite the bullet and follow the “payable benefits scenario”. Yes overall rates of return will be somewhat lower for some period of time, but continually raising the SS tax results in a dog chasing its tail scenario where we continue raising the SS tax rate lest some generation (series of cohorts) gets a worse deal than the previous.
What I have not yet seen is an evaluation of the likely drag on the economy by increasing the SS tax. If the economy grows more slowly than projected in the tables from link from M.jed’s post, the amount of money the system can payout will suffer, and politicians might again call for higher SS tax rates in order to hide the problem (which I why I call SS a Ponzi scheme when I am in a cynical mood).
Michael your comment seems reasonable enough. Until you start applying some arithmetic tests to your assertions (to the extent that they allow such tests). At which time they break down.
For example this:
“but continually raising the SS tax results in a dog chasing its tail scenario where we continue raising the SS tax rate lest some generation (series of cohorts) gets a worse deal than the previous.”
Assumes that a constant rate scenario (essentially your ‘payable benefits scenario’) actually results in a “worse deal than the previous” and so that only increasing rates allows for future cohorts to keep up. But the actual arithmetic shows that under standard projections a policy of ‘Nothing’ which keeps FICA rates constant and results in a 25% benefit cut STILL results in a ‘better deal’.
That is by something I call “Rosser’s Equation” you can show that the benefit in real basket of goods terms faced by those who would take a 25% cut from ‘scheduled benefits’ to ‘payable benefits’ come Trust Fund Depletion in 2033 would still be around 12-15% better than current retirees get today.
Now there are very good reasons to insist that this 12% improvement in standard of living in 2034 over 2013 is not in fact equitable when compared to that retirees standard of living pre-cut in 2033 or then currrent workers’ standard of living in both years. And those reasons I suggest are what keep Coberly leery of “Rosser’s Equation” and its possible suggestions for policy. But the point is that is makes your assertion that rate increases are needed just to keep the dog chasing its tail void, invalid, null. Because that is NOT what rate increases would do at all.
The reason you, when “in a cynical mood”, are tempted to call Social Security a Ponzi scheme is I suggest because your understanding of the relation between scheduled and payable benefits and the level of FICA needed to support either is fatally flawed. Even though it makes superficial sense the arithmetic isn’t there. Which to me suggests maybe you (following MJed) paying some attention to the arithmetic.
You might still come out disagreeing with Coberly on his policy proposals and maybe make his life miserable by appealing to “Rosser’s Equation” to draw the policy conclusion which neither I (who named it) or Professor Rosser (who inspired it) would assent. But at least we could go at this ‘mano a mano’ or ‘numero a numero’ rather than from your ‘back of the envelope calculations’ that in fact seemed to have required neither an envelope or actual calculations.
hansberry
this is all essentially what you said last time. i tried to answer you then. what can i say now that has any better chance of explaining it to you?
try to think of it as paying for your groceries…. which is what it is. it is you paying for your groceries now so you will be able to eat after you are too old to work.
if the cost of groceries goes up, or your income goes down, you don’t just stop eating. the cost of future groceries will go up because you are going to be living longer… and so eating more meals over more years…. and your income may go down, because the “interest” you get on your savings may go down. the “interest” in social security comes from growth in the economy… or the workers wages.
if the interest… growth in the economy… goes down, it is most likely that other forms of savings will also generate lower returns. lower stock prices, lower bond interest.
in any case the important point here is that you can’t just “fix” the cost of social security and then tell everyone they have to stop eating if the cost of groceries goes up, or the income from a no longer growing economy is less than it was in the last generation.
there is no ponzi scheme here. when you say that you might as well wear a big sign that says “I have been lied to, and I believe the lie, and that settles it.”
raising the SS tax will NOT result in any dog-chasing-its-tail scenario. how could it? you need to think that one through. it makes no sense at all.
we are not raising the tax rate “lest some generation get a worse deal than the previous.” on the contrary i am try to explain to people that that “generational theft” argument is a Big Lie. Just as interest rates go up and down, and the price of bread goes up and down, we may get a better or worse deal on our SS “investment.” But not much worse, and in any case, if we are going to eat when we get old, we will need our SS.
I’d ask you if you have a better plan, but I don’t want to hear some “if only” fantasy about every rising returns on stocks and bonds.
Bruce,
It is not my arithmetic, it is the SS trustee’s. See link provided by M.Jed. Most cohorts will get a slightly lower rate of return under the “payable” scenario (as compared to present), that’s not the end of the world, but not something to be ignored either.
Coberly,
If you think the trustee’s are lying, then I must ask why then do you support the system?
My plan is as stated in my last post, stick with the “payable” scenario and let things play out.
Your unwillingness to consider rate of return is why you cannot see the tail chasing that results from raising the tax rate. If you want to kill the program, make is so a large percentage receive a negative rate of return -then try to tell them it doesn’t matter.
hansberry
try to imagine a mechanism whereby Social Security could “hurt the economy.”
i don’t think you can do it.
How can my buying bread to give to someone’s aged grandmother “hurt the economy”?
What the hell is an economy for?
hanesberry
i wish the hell you could read. “rate of return” what the hell do you think this post was about”?
the trustees are political appointees. they lie as much as they can get away with. the Big Liars are the Petersons.
i support the system because it works and will work forever unless idiot
unless idiots let the liars destroy it.
i tried to be nice, but what can you do with somone who reads an essay about rate of return and then complains about “not looking at rates of return”?
Coberly,
No need to imagine anything. In recent memory the current administration cut SS tax rates in order to stimulate the economy. This only makes sense if the administration considered SS taxes as a drag on the economy. And when the rates were re-instated many claimed that the higher rates would hurt the recovery.
If the trustees are lying, in which direction could we expect them to lie? Seems they ought to be lying (if they are lying) in direction of supporting the party in power, which is the Dem party. So, if anything, the Trustees are likely understating the problems with SS, see the more recent trustee reports here: http://www.ssa.gov/OACT/NOTES/ran5/index.html
Compare tables 4,5,and 6 in the Mar 2013 report. As in the 2005 report (created under Repub admin) rates of return are falling in most cohorts.
The system will only work forever if it is well managed. Politicians aren’t likely to see that is done.
Coberly said:
there is no ponzi scheme here…raising the SS tax will NOT result in any dog-chasing-its-tail scenario. how could it?… we are not raising the tax rate “lest some generation get a worse deal than the previous.” -(end quote)
But it will result in dog chasing tail, because while raising the tax rate will essentially maintain the rates of return (compare tables 4 and 5), we will then have to raise them again at some point in order to maintain rates of return when those cohorts who will have worked under the increased rates begin to retire. And so on, and so on, and so on.
Thank you Michael
you restore my faith in humanity.
Hansberry you are an arithmetic moron.
The Administrations (foolish) cut in FICA was fully compensated by a offset in revenues from the General Fund making its impacts on e economy if anything an argument for progressive tax rates rather than an argument that FICA was a drag on the economy as such.
Dale and I opposed it at the time but if anything it’s ‘success’ undermines your argument rather than supporting it. The argument for a FICA cut is an argument for raising tax on capital and lowering it on labor compensation.
Thanks for playing our Home Game.
As to rate of return you totally ignore increase in real basket of goods terms (Rossers Equation) whil discounting insurance values to zero.
Put those together and no alternative actually suggested measures up. So what is your suggestion? Soylent Green futures?
Bruce,
Once again these are not my figures, they are the SS trustees figures, most people would be able to follow the link and realize that,
http://www.ssa.gov/OACT/NOTES/ran5/index.html#sthash.exNVzBaX.dpuf
Due to the increases in SS tax rate and real growth in wages over time, later cohorts pay more into the system in real terms than earlier cohorts do over their working lives. That later cohorts will get back more in real terms at the same, or even a slightly reduced, rates of return should not be surprising.
The link provides much info on the Trustee’s methods, if you feel they are not properly accounting for the cost of the disability insurance provided by SS when calculating rates of return, you should take it up with them without delay.
Those numbers – particularly the ROR on higher earners – are much higher than estimates I’ve seen from other places. I recall thinking it would be trivially easy to beat the ROR for a higher earner.
re “insurance”: most insurance covers risks that may or may not occur, and for those products it probably doesn’t make sense to think about it in terms of ROR. for things like life insurance, OTOH, one would be foolish *not* to consider the likely ROR. the people who resist that thinking? insurance and annuities salespeople.
Jpe,
Agreed. The cost of the disability portion of SS ought to be deducted from what each cohort pays into the system (reduce “payment”) before calculating the rate of return on the lifetime annuity-like portion of SS.
This will produce higher rate of return as compared to an overly simplified calculation that ignores the disability benefit, but one more reflective of the benefits provided by the SS system.
I have also seen ROR projections for younger high earners that are already negative. Could be those estimates are not taking into account the disability benefit , or maybe using lower growth estimates for the economy. For this thread I thought it better to use the SS trustee ROR projections in order to put aside discussion about methodology, and focus on best next steps. In any case, even the SS Trustee projected ROR are barely positive for med-high earners born 1970 and later.
It seems to me we ought to take a “first, do no harm” approach in determining best next steps. That is why I favor the “payable benefits scenario” and simply letting things play out.
I’m a little unclear about how you calculate those taxes I paid prior to 1986 and the Reagan tax shift from income taxes to Social Security. I know my direct and indirect social security taxes went up, but how does that affect the rate of return?
Logically the rate of return for someone who retired and stopped paying taxes in 1986 must be higher than the rate of return of someone who started paying taxes in 1986.
“these are not my figures, they are the SS trustees figures”
As often the case Bruce and coberly have allowed emotion into their argument. All it takes is numbers.
Go back to the numbers and do the correct comparison, Table 2 to Table 3 (or 5 to 6) and you will see the ROI is higher for the increased tax case than for the payable benefits case. Opposite what it sounds like Michael is concerned about.
“the current administration cut SS tax rates in order to stimulate the economy. This only makes sense if the administration considered SS taxes as a drag on the economy.”
Bruce already responded, but the “this only makes sense” part is just so hard to read. Cutting the payroll tax was one of the only politically palatable ways to increase deficit spending (generating stimulus), but politics is not much about what makes economic sense.
Willis
when someone says “logically” it usually means “there is something i haven’t thought of”
but yes, if you narrow your focus to only actual taxes paid to Social Security and benefits collected from Social Security… the earlier cohorts got a better deal than later cohorts.
this doesn’t mean that Social Security will be a bad deal for later cohorts. it just means they didn’t buy IBM when it was “new.”
meanwhile… did you lose your savings in the depression? did you fight in world war 2 or korea or vietnam? did you lose two years to the draft?
are you going to increase the GDP as fast as they did? are you going to keep as big a share of GDP going to workers as they did?
and where did you see the guarantee from god that every generation would get at least as good a deal as the last?
meanwhile you don’t know what your own personal deal on Social Security will be because it is insurance. if you fall off a ladder you will get a much higher j”rate of return” than if you don’t.
so logically the only thing for you to do is fall off a ladder. maximize your rate of return, especially as compared to your evil parents. that’s what life is all about, right?
Arne,
I am not sure why you say ” Opposite what it sounds like Michael is concerned about..”
My concern is precisely that the rate of return is propped up by continually raising the SS tax rate.
I agree with Bruce and Copberly when they argue that ROR will fluctuate over time (due to changes in demographics and the economy) and so we shouldn’t get too hung up on that. Where I disagree is in calling for continuous ramping up of the tax rate which actually does make SS a Ponzi scheme though it avoids the near term pain of Politicians telling the people that there is no free lunch, and yes some cohorts are going to get near zero rates of return.
Moreover continually raising the tax rate will likely deduct from economic growth (which is the goose that lays the golden eggs in terms of providing the means for giving the next generation a real return on their SS taxes paid) and thus harms the system.
“My concern is precisely that the rate of return is propped up by continually raising the SS tax rate.”
Are you making a valid comparison with the numbers that coberly has posted? The cash value of an insurance company annuity deducts the cost of insurance, so to retain its ROR would also require increasing the premium as your life expectancy increases.
If you want to maintain your lifestyle in retirement for longer than your parents, you are going to need to save more while you are working. It’s true for the part you save on your own to make retirement more interesting and more comfortable. It’s true for the part that means having a roof and food, which SS can insure.
I think it is interesting that the ROI is higher if you increase the rate than if you don’t, but only somewhat intellectually, since raising rates is not about propping up ROIs anyway.
“Moreover continually raising the tax rate will likely deduct from economic growth”
I have heard this before, but I think it is sloppy analysis.
SS does not consume payroll taxes. The money gets spent, just by someone else. Net impact on economic growth – zero.
I have also heard it gets transferred from people with less wealth to people with more wealth, but such people tend to compare wealth of 35 year-olds to wealth of 65 year-olds. But the average saver is over 35 and the average SS beneficiary is older than 65. Given how few SS recipients have other sources of income left it seems their propensity to spend is higher than worker, so there is a valid argument that SS increases economic growth.
(If SS allows people to stop working before they would have, it clearly reduces GDP as measured, but it also indicates that people value the unmeasured more than the measured. I understand your argument, but I don’t find it compelling.)
Arne said:
I think it is interesting that the ROI is higher if you increase the rate than if you don’t, but only somewhat intellectually, since raising rates is not about propping up ROIs anyway. (end quote)
It would be a neat trick if we could divorce benefits/payments from ROI but we cannot.
and
Arne said:
SS does not consume payroll taxes. The money gets spent, just by someone else. Net impact on economic growth – zero. (end quote)
One person works, but another spends. That sounds swell, just as long as you put me on the list of those spending the money. If you put me on the list of those doing the work, then I might have a different opinion of your plan. and might be somewhat less willing to cooperate.
“It would be a neat trick if we could divorce benefits/payments from ROI but we cannot”
SS is Pay As You GO. ROI does not determine benefits. ROI is only an intellectual exercise.
“put me on the list of those spending the money”
You are on the list if you are paying payroll taxes. You just wait to spend it the same way you wait to spend your pension or your 401k if you are lucky enough to have those as well.
And I note that the rejoinder did not comment on the validity of the economic growth argument.
Arne
thanks. brave try.
i used to believe, myself, that if only i could explain it clearly enough they would understand it. but they don’t want to understand it.
Arne said: SS is Pay As You GO. ROI does not determine benefits. ROI is only an intellectual exercise. (end quote)
You may as well have said “pay no attention to the man behind the curtain.” There is an ROI whether you calculate it or not, it is a function of each cohort’s taxes paid and benefits received (demographics are a given). You may not want to know what the ROI is, but it still exists. You cannot wish it away.
As to the economic growth argument, I gather that you believe that people will work just as hard when a large portion of the fruits of their labor is being taken from them and handed to another person. I do not share that belief. I do not expect to be able to convince you with examples from the past (for example: enclosing of the commons leading to gains in agricultural output). But why is it that in 2013 a mandate is needed to prop up the PPACA? Answer: because people are not the selfless producers that would be needed to make PPACA w/o mandate or your version of SS work. Its not that people are evil, it is that they are not saints. I believe People work mostly for themselves and their own families. People will seek to get the best deal for themselves even in the face of government coercion. That is why people hide their assets before going into a nursing home. That is why people use every single device available to lower their taxes, etc.