Your Social Security "Money's Worth"

[dude]

Great article by the always excellent Scott Burns:

http://assetbuilder.com/scott_burns/your_social_security_moneys_worth

[EricP]

Workers at medium earnings levels born in 1920 and retired at 65 in 1985 did well. A single male receives $1.37 for every $1 committed. The next generation, those born in 1943 and retired in 2008, gets only 74 cents on the dollar. Their children, born in 1964 and retiring in 2029, will do somewhat better, at 94 cents on the dollar.

But that 94-cent figure depends on Social Security paying full benefits after the trust fund is empty. This will happen, the actuaries say, around 2033. If benefits are paid from available employment taxes, their money’s worth will decline to 81 cents.

This is interesting.  A few weeks back I was part of a discussion about how millenials and Gen Xers are getting fleeced by Boomers in Social Security and it seems that is exactly the opposite.  Even if the Trust Fund runs out and payments are reduced, the younger generations are making out better than Baby Boomers.

[RangerOne]

That quote doesn't even discuss the millennial generation, who will either see reduced benefits across the board or increased taxes to pay for the full benefits. So I don't see how there is any evidence that millennials will make out in this deal.

As for the disparity due to taxes. Would it not make sense for social security benefits to be un-taxed and not add to your income?

[EricP]

That quote doesn't even discuss the millennial generation, who will either see reduced benefits across the board or increased taxes to pay for the full benefits. So I don't see how there is any evidence that millennials will make out in this deal.

As for the disparity due to taxes. Would it not make sense for social security benefits to be un-taxed and not add to your income?

Then look at the tables and you'll see that millenials across the board are getting more "Money's Worth" than Baby Boomers.

[lackofstache]

The youngest workers may do worst of all. Medium income male workers born in 1985 and retiring in 2050 will receive 97 cents for every $1 paid in. This will happen according to current law. If benefits are cut when the Social Security Trust fund is empty, the same worker will receive only 75 cents on the dollar.

This according to the linked article doesn't seem to say that Millenials win...

[beltim]

Is anyone else annoyed that most of the tables are labeled with "percent" even though they clearly mean "ratios?"

[MrMoogle]

Wish they used inflation adjusted numbers somehow, at least for their contributions.

[beltim]

Wish they used inflation adjusted numbers somehow, at least for their contributions.

That they do.  They use inflation-adjusted numbers for everything.

[EricP]

The youngest workers may do worst of all. Medium income male workers born in 1985 and retiring in 2050 will receive 97 cents for every $1 paid in. This will happen according to current law. If benefits are cut when the Social Security Trust fund is empty, the same worker will receive only 75 cents on the dollar.

This according to the linked article doesn't seem to say that Millenials win...

Women earners of medium income retiring in 2014 will receive 88 cents for every $1 paid into Social Security. Men will receive only 78 cents.

So a Baby Boomer will get 78 cents and a Millenial will get 75 cents IF benefits are cut.  IF benefits are cut, they are within a few percentage points of a Baby Boomer, and if they aren't cut they get much more than the Baby Boomer.  It's anyone's guess what happens in the future, but Millenials are not getting completely hosed and will potentially be doing much better than Baby Boomers.

[RangerOne]

That quote doesn't even discuss the millennial generation, who will either see reduced benefits across the board or increased taxes to pay for the full benefits. So I don't see how there is any evidence that millennials will make out in this deal.

As for the disparity due to taxes. Would it not make sense for social security benefits to be un-taxed and not add to your income?

Then look at the tables and you'll see that millenials across the board are getting more "Money's Worth" than Baby Boomers.

I see it now...

Yes millennials show ~6% higher ratio of return under present law...

Increased tax scenario brings us closer to parity at about ~2% diff.

A tweak to payable benefits shows us at a 2% disadvantage. Clearly the first scenario is not going to happen.

Under the assumption that there with be a payable benefits change to cover shortage post 2033 millennials will get a worse return than baby-boomers. In the event of a tax increase we will punish our youth to pay for a retirement that is negligibly higher than what boomers will get. Either way it looks like maxed income average earners don't get much out of social security for what we pay in...

[RangerOne]

I would agree based on these numbers there is no reason to assume baby boomers are getting a better deal than we potentially will. It could go either-way and in both cases numbers are nearly at parity.

[beltim]

So how does this jive with real rates of return being positive for almost every income level and marital combination?

http://ssa.gov/OACT/NOTES/ran5/index.html

[Paul der Krake]

Of course the benefits curve is skewed to help out the low income folks. This is why it's called social security.

Not mentionned in the article, but of real significance for early retirees: you need 40 quarters to be eligble. For some of us on track to be FI within 10 years of entering the workforce, there are easy ways to grab the remaining credits.

Making as little as $1,220 per quarter, is enough to get one of the 40 necessary credits.

[brooklynguy]

Not mentionned in the article, but of real significance for early retirees: you need 40 quarters to be eligble. For some of us on track to be FI within 10 years of entering the workforce, there are easy ways to grab the remaining credits.

Making as little as $1,220 per quarter, is enough to get one of the 40 necessary credits.

It's even better than that, because, unlike a few decades ago, "quarters" are no longer relevant to how credits are earned.  The number of credits you can earn in a single year is capped at four, but there is no requirement for the credits to be spread over any amount of quarters per year.  So if you earn $4,880 in a single quarter (or in a single day), you've already earned the maximum possible amount of credits for that year.

[forummm]

Not mentionned in the article, but of real significance for early retirees: you need 40 quarters to be eligble. For some of us on track to be FI within 10 years of entering the workforce, there are easy ways to grab the remaining credits.

Making as little as $1,220 per quarter, is enough to get one of the 40 necessary credits.

It's even better than that, because, unlike a few decades ago, "quarters" are no longer relevant to how credits are earned.  The number of credits you can earn in a single year is capped at four, but there is no requirement for the credits to be spread over any amount of quarters per year.  So if you earn $4,880 in a single quarter (or in a single day), you've already earned the maximum possible amount of credits for that year.

Also, you could work just in January, make $5k, and get your 4 credits for the year. But if you only made $5k per year for 10 years and that was your only income and you never married, your SS benefits would be very small--around $115/month at FRA. So you still have to have some significant earnings (whether high earnings for 10 years or slowly over 35 years) to get anything appreciable back.

[Paul der Krake]

The calculators online don't exactly give the option of sitting on your bum for 30 years, but if I input $20,000 over 40 years (starting to work at 22), which should be the same as averaging $80,000 over 10 year, then I get $712/month at 62. Not bad at all.

[brooklynguy]

So you still have to have some significant earnings (whether high earnings for 10 years or slowly over 35 years) to get anything appreciable back.

Yes, but that's a prerequisite for being able to retire in the first place :)  The point was that you may need to top off your credits to unlock your benefits if you retire in less than 10 years.

[Cathy]

Not mentionned in the article, but of real significance for early retirees: you need 40 quarters to be eligble. For some of us on track to be FI within 10 years of entering the workforce, there are easy ways to grab the remaining credits.

Making as little as $1,220 per quarter, is enough to get one of the 40 necessary credits.

It's even better than that, because, unlike a few decades ago, "quarters" are no longer relevant to how credits are earned.  The number of credits you can earn in a single year is capped at four, but there is no requirement for the credits to be spread over any amount of quarters per year.  So if you earn $4,880 in a single quarter (or in a single day), you've already earned the maximum possible amount of credits for that year.

This is generally true, although there are exceptions (both favourable and unfavourable to beneficiaries). The general rule and some exceptions are codified in 42 USC § 413.

For people who have worked outside the USA, there are exotic favourable exceptions to be found in various international "agreements" to which the US is a party. Prospective early retirees should familiarise themselves with these international agreements. For example, under Article VII, §§ 1-2 of the US-Canada Social Security Agreement, each year during which a person contributed to the Canada Pension Plan at all counts as 4 quarters for US Social Security, subject to the fact that a person cannot receive credit for the same quarter twice (which could otherwise happen between the application of this rule and the normal US domestic rules).

[forummm]

The calculators online don't exactly give the option of sitting on your bum for 30 years, but if I input $20,000 over 40 years (starting to work at 22), which should be the same as averaging $80,000 over 10 year, then I get $712/month at 62. Not bad at all.

This one does
http://www.ssa.gov/planners/retire/AnypiaApplet.html

[Purple Economist]

So how does this jive with real rates of return being positive for almost every income level and marital combination?

http://ssa.gov/OACT/NOTES/ran5/index.html

In all of the Social Security threads on this site, the difference between the money's worth ratio (which indicates many do not get their money's worth) and the internal rate of return (which is positive for essentially everyone) has always perplexed me.  Your question finally inspired me to try and square the differences.

This is a link to a SSA working paper from April 1995:

http://www.ssa.gov/policy/docs/workingpapers/wp67.pdf

The relevant quote I found is:

"Thus, if the internal rate of return is larger than the interest rate available to workers for their own investments; then they receive more than their money's worth from Social Security; that is they receive a higher "interest rate," or internal rate of return, from the Social Security program than from their private savings.  Conversely, if the internal rate of return is smaller than the interest rate that workers can earn privately, then they do not get their money's worth from Social Security."

This indicates that the money's worth ratio referenced in the OP's link takes into account the opportunity cost from investing the taxes you pay to Social Security.  The internal rate of return does not appear to.

This is a link to an NBER working paper:

http://www.nber.org/papers/w6722.pdf

In this paper, the assumed interest rate one could achieve was 2.3%  With this assumption, any internal rate of return below 2.3% would represent a money's worth ratio of less than 1 indicating that worker is not getting their money's worth.

Furthermore, if for yourself, you assume a real interest rate on investments (i.e. stocks) of 4%, 5% or 6%, then your personal money's worth ratio is going to be very much below 1 and you may view Social Security as a terrible deal.  This is assuming that you value Social Security only for the retirement income.

[Zamboni]

Interesting. I view my social security payments as going directly to my parents. When my Mom was paying into it, she said it was for her Granny. It makes me feel better about the whole system. Anything I get from it will just be gravy, since obviously my plan is to well . . . plan.

[Purple Economist]

Interesting. I view my social security payments as going directly to my parents. When my Mom was paying into it, she said it was for her Granny. It makes me feel better about the whole system. Anything I get from it will just be gravy, since obviously my plan is to well . . . plan.

Whose payments are going to people that have no kids?

[Paul der Krake]

Interesting. I view my social security payments as going directly to my parents. When my Mom was paying into it, she said it was for her Granny. It makes me feel better about the whole system. Anything I get from it will just be gravy, since obviously my plan is to well . . . plan.

Whose payments are going to people that have no kids?

Immigrants who have no parents.

[coppertop]

There are also people who have died and no one has collected on their accounts.  My dad got whatever the so-called 'Lump Sum' was back in 1980 for my mother who died at age 50.  My sister, who was age 17, received no benefit for the month of February of that year, because my mother was alive for six days in February.  She turned 18 on May 29.  She received no benefit that month because she turned 18 at the end of that month. So for all of my mother's working years, my family received the joke of a "lump sum" and my sister received a payment for March and a payment for April.  The rest of my mother's contributions wound up being forfeit to the system. 

[forummm]

It's insurance. Really valuable insurance. Don't expect it to pay out exactly what you put in. It could be much more or much less. It protects you from your investments failing, or for living much longer than you expected. And it does it all without charging you profits that private insurer annuities require.

[brooklynguy]

It's insurance. Really valuable insurance. 

People also tend to forget about the non-retirement related benefits, like disability and survivors' benefits, which provide insurance against other risks as well.

[EricP]

It's insurance. Really valuable insurance. Don't expect it to pay out exactly what you put in. It could be much more or much less. It protects you from your investments failing, or for living much longer than you expected. And it does it all without charging you profits that private insurer annuities require.

Are you arguing that Social Security is cheaper than a comparable private annuity?  Because I find that hard to believe since private annuities can invest their trust funds which I'm guessing more than compensates for the profit they make, without even discussing how government is inefficient.

[mm1970]

Workers at medium earnings levels born in 1920 and retired at 65 in 1985 did well. A single male receives $1.37 for every $1 committed. The next generation, those born in 1943 and retired in 2008, gets only 74 cents on the dollar. Their children, born in 1964 and retiring in 2029, will do somewhat better, at 94 cents on the dollar.

But that 94-cent figure depends on Social Security paying full benefits after the trust fund is empty. This will happen, the actuaries say, around 2033. If benefits are paid from available employment taxes, their money’s worth will decline to 81 cents.

This is interesting.  A few weeks back I was part of a discussion about how millenials and Gen Xers are getting fleeced by Boomers in Social Security and it seems that is exactly the opposite.  Even if the Trust Fund runs out and payments are reduced, the younger generations are making out better than Baby Boomers.

Hm that's pretty interesting.
My dad was born in 1926, and I think he was on SS for almost 20 years (from age 62 to almost 82, when he died).  He didn't get much though.
My mom was born in 1944.  I don't remember when she started collecting, but she died at almost 68, so it wasn't very long).
My sister was born in 1964, so I guess we'll see how she does!

I have two older sisters already on SS (they are 64 and 59), and one who is eligible but still working.  They probably fall in the middle.

[beltim]

So how does this jive with real rates of return being positive for almost every income level and marital combination?

http://ssa.gov/OACT/NOTES/ran5/index.html

In all of the Social Security threads on this site, the difference between the money's worth ratio (which indicates many do not get their money's worth) and the internal rate of return (which is positive for essentially everyone) has always perplexed me.  Your question finally inspired me to try and square the differences.

This is a link to a SSA working paper from April 1995:

http://www.ssa.gov/policy/docs/workingpapers/wp67.pdf

The relevant quote I found is:

"Thus, if the internal rate of return is larger than the interest rate available to workers for their own investments; then they receive more than their money's worth from Social Security; that is they receive a higher "interest rate," or internal rate of return, from the Social Security program than from their private savings.  Conversely, if the internal rate of return is smaller than the interest rate that workers can earn privately, then they do not get their money's worth from Social Security."

This indicates that the money's worth ratio referenced in the OP's link takes into account the opportunity cost from investing the taxes you pay to Social Security.  The internal rate of return does not appear to.

This is a link to an NBER working paper:

http://www.nber.org/papers/w6722.pdf

In this paper, the assumed interest rate one could achieve was 2.3%  With this assumption, any internal rate of return below 2.3% would represent a money's worth ratio of less than 1 indicating that worker is not getting their money's worth.

Furthermore, if for yourself, you assume a real interest rate on investments (i.e. stocks) of 4%, 5% or 6%, then your personal money's worth ratio is going to be very much below 1 and you may view Social Security as a terrible deal.  This is assuming that you value Social Security only for the retirement income.

Fantastic post!  This explanation makes perfect sense.  I sure wish the Moneys Worth Ratios reports mentioned this - because this is a huge difference.

To check whether this explanation, I looked at a few cases where the moneys worth ratio was about 1.0.  From Table 1, a low earning single male retiring in 2008 has a moneys worth ratio (MWR) of 1.00, while the equivalent entry on the real rate of return was 3.21%.  A medium earning single female retiring in 1995 has a MWR of 1.00 and real rate of return of 2.84%.  A max income one-earner married couple retiring in 2014 has a MWR of 1.01 and real rate of return of 3.05%.

From Table 4, a two-earner couple with both earners at the medium level retiring in 1995 had a MWR of 1.02 and rate of return of 2.95%.  From Table 5, a two-earner couple with high/med earners retiring in 2038 had MWR of 1.01 and rate of return of 2.52%.

So it looks like the historical opportunity cost they used was 2.8-3.0%, and the future rate is more like 2.5%. 

Oh, lord, I just realized there's a whole table with the interest rates they used: Table B on Page 4 of the MWR report.  They're using about 1.0% now, ramping up to 2.9% after 2027.

[forummm]

It's insurance. Really valuable insurance. Don't expect it to pay out exactly what you put in. It could be much more or much less. It protects you from your investments failing, or for living much longer than you expected. And it does it all without charging you profits that private insurer annuities require.

Are you arguing that Social Security is cheaper than a comparable private annuity?  Because I find that hard to believe since private annuities can invest their trust funds which I'm guessing more than compensates for the profit they make, without even discussing how government is inefficient.

I haven't compared the prices. Have you? You'd have to include all the disability and spousal benefits, etc, to have a full comparison. Also it would have to be from a company that could never fail and would invest in assets that were guaranteed like Treasuries.

All I was saying is that the comparable cost is not going to include profit. And I would bet that SS's administrative costs are significantly lower than a private insurers. Similar to health insurance. There are some things government does really well.

[MoonShadow]

This article is meaningless, because it doesn't disclose how much it's representative workers actually earn during their tenure, before claiming that this one gets that much and that one gets this much.  The payout in SS is progressive, in the sense that the more you make, on average across a 35 year period, the lower your payout ratio goes.  This article doesn't really disclose it's other assumptions either, such as how it is expecting benefits to be 'cut' in the event that taxes are not raised.  It must also have a basic assumption about how much Millenials will actually earn over their careers, which is hugely debatable.

[beltim]

This article is meaningless, because it doesn't disclose how much it's representative workers actually earn during their tenure, before claiming that this one gets that much and that one gets this much.  The payout in SS is progressive, in the sense that the more you make, on average across a 35 year period, the lower your payout ratio goes.  This article doesn't really disclose it's other assumptions either, such as how it is expecting benefits to be 'cut' in the event that taxes are not raised.  It must also have a basic assumption about how much Millenials will actually earn over their careers, which is hugely debatable.

All of those are addressed in the actual social security document linked.  I agree that the original article does a terrible job of addressing those issues though.

[MoonShadow]

It's insurance. Really valuable insurance. Don't expect it to pay out exactly what you put in. It could be much more or much less. It protects you from your investments failing, or for living much longer than you expected. And it does it all without charging you profits that private insurer annuities require.

Are you arguing that Social Security is cheaper than a comparable private annuity?  Because I find that hard to believe since private annuities can invest their trust funds which I'm guessing more than compensates for the profit they make, without even discussing how government is inefficient.

I haven't compared the prices. Have you?

I have. I've found that most annuities have an underlying growth assumption of between 3.5 and 4%.  So with regard to a simple fixed income, they kick SS's ass; but as an investment vehicle, they actually suck.

You'd have to include all the disability and spousal benefits, etc, to have a full comparison.

Yes, but that is only complex because SS benefits are so complex, not because it's difficult to get quotes for disability coverage or life insurance (for spousal benefits).  In practice, most everyone in the US has access to both these insurance products, either independently or subsidized by their employer as a benefit.  However, such insurance products are not paid for as a percentage of wages; so the premiums become complex.  I just looked it up, and I paid $1,681.39 last year for disability insurance.  I can't get life insurance, so that number isn't relevant to me; but I paid $7,254.00 in SS and $1,750.19 for Medicare taxes.  I also have a SS statement that says that I can expect $1450 per month (in today's dollars) at 67 if I continue to work till then and my current rate.  I have another 27 years of working (not going to happen, but) to make that $1450 per month in constant spending power happen.  Does anyone want to run those numbers and make sense of it?  I'm not motivated enough.

Also it would have to be from a company that could never fail and would invest in assets that were guaranteed like Treasuries.

Well, that is a fallacy for several reasons, but even if it were completely correct; which do you think has a longer average survival rate; life insurance companies, government programs or actual governments?  Take your mind away from the US for a minute, and consider the number of governments on Earth today.  We can limit ourselves to those we consider modern or "Western" governments, if you like; but you must keep in mind that just because a particular nation still has the same name and borders as in the past, doesn't conclude that the government that manages the nation's affairs.  For example, the governments of Germany & Austria aren't really older than 1945, and Spain's has only been around since 1962.  Granted, there are much older examples of continuous governments around Europe, but we are also talking about Europe; other regions of the world have much worse records, such as central and south America, and Africa; despite there being wonderful examples of modern & 'Western" democracies on both continents.  And size of country doesn't seem to really be a factor either, since Russia's government has only been around since 1991.  For me, it's hard to say which is more likely to exist past my own lifetime, any particular government or any particular life insurance company.

All I was saying is that the comparable cost is not going to include profit. And I would bet that SS's administrative costs are significantly lower than a private insurers.

You would lose that bet easily.

Similar to health insurance. There are some things government does really well.

Yes, there are.  Insurance is not one of them.