There is no shortage of reading material if you wish to determine “The best time to collect social security.”
The vast majority conclude that it is best to wait as long as possible because the monthly income will be higher. Delayed social security income is larger than early social security income, and More is MORE!
I plan to do the opposite.
Why I Plan to Collect Social Security as Early as Possible
The arguments for delay are actually compelling. The increase in monthly payments by delaying from age 62 to 70 is a whopping 76%.
The Social Security Administration offers a great benefit calculator (directly after you login.) I will turn 50 this year, and even though I haven’t really worked in a decade (and have no plans to work) I am looking at a nice $24k+ in annual income at age 62. If I delay to age 70, that increases to $44k+ (inflation adjusted.)
An extra $20k per year seems nice. Delay is better, case closed, right?
Well… what is the rate of return for delaying? If I collect earlier and invest that income, can I do better?
The following chart shows the investment return required to outperform the government (but excludes spousal benefit)
(Related: Social Security Return on Investment.)
For those living beyond age 88 or so, the real rate of return is 5%+. It is pretty hard to beat a 5% rate of return with zero volatility….
I can understand why delay is often recommended.
(For more analysis of the return of delay, Kitces has the best analysis I have seen.)
But…
I am generally an optimistic guy. People are living longer, medical advances continue, etc…
But accidents happen. Sudden illness happens. The ROI of delay if dying before age 70 is -100%, a total loss. The ROI if dying before age 80 is zero.
And… 7 out of 10 people die before age 88, the age at which delaying social security starts to have a really compelling ROI.
Even then… at death all benefits stop. I can’t bequeath a Social Security income stream to our children or charity.
By delaying… at best I get an investment return lower than the SP500 has returned over the past 40 years or so. At worst I get nothing.
Thus… I plan to collect SS as early as possible.
Exceptions
I can think of a few situation where it might make some sense to delay, at least a few years if not until age 70.
Once SS income starts it will impact the tax rate on Roth conversions. For people with (very?) large Traditional IRA/401k, it might make sense to continue aggressive Roth conversions for a few more years. (Although I haven’t done the math.)
For those with few assets, it can make sense to delay in order to reduce total assets to zero. This increases likelihood of eligibility for Medicaid support for nursing homes, etc.. if needed.
For people with longevity genes, a guaranteed 5%+ real rate of return is nice and definitely convenient.
Some people have ACA based health insurance with income based subsidies. It can make sense to delay SS until Medicare age at 65.
(hat tip to Katydid in the comments)
Summary
I plan to collect Social Security as early as possible because, even though I plan to live a very long time, I can’t bequeath a Social Security income stream and the long-term ROI of delay is not an obvious win over just dumping early SS income into an index fund.
When do you plan to collect Social Security?
Love this post. I’ve always been of the mindset of delaying until age 70, but your math and arguments are very compelling. The oldest family member I had in my family was age 90. My current retirement plan takes me through age 94….
I’m a couple years behind you so I will look forward to your future blog posts when you take your social security!
I plan to collect as early as possible too. Unless I don’t need the money. Then I might delay…
I won’t need the money, it will all be extra. By taking early, that extra will be EXTRA.
Relevant to this post: Spending Future Social Security Income Now
I love this post, too. Great job per usual and thanks for sharing your knowledge. Could use some more sarcasm, though… Ha! ;)
You know something, you make an excellent point. Noted for future writing efforts.
Exactly! Every expert says to wait. The way I see it taking it early I am betting that I will die young, and this is a bet I will be happy to lose.
This may be the most compelling reason to take it early! Thanks.
I’ve always sort of thought of it in investment terms and plan to diversify the risks by taking it somewhere in the middle. 64-67. If I live forever, it’ll help and if I die early~ish I’ll (hopefully) still have pulled some money out of it.
While I usually agree with your general strategy (trust math), I like to think of SS as “old person” insurance. There is nothing like guaranteed income, especially at the stage in your life when you are least able to adapt (physically and mentally) to financially stressors. My strategy is to delay SS benefits as long as your finances allow you to.
I wouldn’t argue with you about your strategy. It seems a fine choice.
If you/spouse live a long life then it is a respectable return with zero volatility.
I do think that we will be in a stronger financial position and thus less susceptible to financial stressors by taking early.
If Social Security runs out of money to pay full benefits, or if benefits are cut, at least we get some of our money. I am planning to take Social Security early, too, and invest and save it.
SS is inflation adjusted annuity – which cannot be commercially purchased. I plan to delay taking until age 70 in order to maximize its value of mitigating longevity AND inflation risk.
Agreed, SS is an inflation adjusted annuity. And in fact it is superior to any other annuity that can be purchased.
But I have no plans to buy any annuities (due to similar logic) and will continue to use the stock market as an inflation risk mitigator.
Craig, you are correct about annuities, but annuities depend on the stability of the issuer. There is a serious risk of a reduction in benefits across the board in about 10 years. A bird in the hand…..
I’m collecting Social Security at 70 because of the survivor benefit I’ll leave to my wife when I die.
That makes sense and will affect my decision as well. However, the maximum spousal benefit is 50% of the higher earning spouse’s benefit at full retirement age. So although you would increase your spousal benefit by waiting between ages 62 and 67, there would be no further increase in spousal benefit between ages 67 and 70.
Loach: you are conflating the spousal benefit and the survivor’s benefit. The surviving spouse’s benefit gets topped off to bring it to the full level of the deceased spouse, if that benefit is higher.
Mike Piper’s book “Social Security Made Simple” can clarify this for you.
In my and my late husband’s case, the advice that the higher earner delay to 70 was very good and I am very grateful that mu husband did that.
Here are the numbers from our experience: Mike Piper’s Open Social Security suggested that I (seven years younger and the lower earner) file at 62, suspend benefits, and that my husband collect a spousal benefit on my record until 70, and then file for his own benefit. We should have done that! I filed at 65; he filed at 70. He got sick at 71, died at 72, and as his survivor, my benefit went from my $1500/month to his $4000/month, now up to $4200+/month with the inflation adjustment.I’m getting that now at age 68, and for the rest of my life. The effect of the COLA compounding on a higher benefit is significant, too.
Jeremy–I love your analyses, but I think this one could be enhanced with a closer look at spousal survivor benefits.
My ex husband died before starting to collect and did not remarry. I was told i could start collecting his full benefit at age 60 so since i would get 2 “extra”
Years, i am thinking of taking it early….?
I think that file and suspend loophole is no longer possible since 2015. Is anyone with better knowledge able to speak to that?
Trudy, I don’t believe I am conflating anything. The file and suspend strategy that you benefited from is unfortunately no longer available.
I think you have the correct answer if the goal is to get maximum dollars.
However SS is insurance that has benefits not measured in dollars. As others mentioned, longevity insurance.
I also think it is a risk to assume that my own facility with managing my assets will not decline and/or that I won’t make errors.
Or that my wife will of I preceed her.
As well we are all at risk of fraud or theft, whether strangers or family.
These are all fine reasons to take a different approach than me.
There is no investing prowess required though. Just put the $ in an index fund and ask Vanguard to transfer $x every month from the investment account to a checking account.
If it was just me, I’d take it earlier. But the spousal benefit is important also. Genetically, bet on my wife to die first, but it is only a bet. She had minimal working income and only gets about 35-40% of my pension income if I kick off first. So SS survivor benefit would close that gap. I’m thinking 62 her, 67 to 70 for me….
Just one thing on the survivor benefit… the question isn’t “Do you want a smaller or larger income after my death?”
It is, “Do you want a larger income or a smaller income and a GIANT PILE OF CASH?!”
More details in the post: Why Early Social Security Provides the Greatest Spousal Benefit
Jeremy, what are your thoughts regarding spousal benefits? I assume it’s the same given your analysis and that Winnie will also claim her benefit at 62? I’ve read a number of articles that suggest a strategy of the lower income withdrawing immediately and the higher income delaying.
We can think of the spousal benefit as just the higher income earner living longer. The ROI of delay asymptotically approaches some maximum (~6.5% in the chart in this post.) Even another few decades doesn’t increase the maximum.
My wife started SS immediately at 62 for the simple reason that longevity is not a forte of her family. I planned to delay at least until FRA but started at age 63 when I decided I could do better with the money. Plus I have a healthy distrust of government at all levels, so I was betting that if I was already collecting SS I might not be affected by any of the machinations BOTH parties might do to steal even more from SS.
>I decided I could do better with the money
Yup. Doesn’t seem so hard to get 5%+ real over the long term.
I agree with the early start on Social Security. A couple of thoughts:
Your chart does not account for any cost of living increases
Maintaining current Social Security rates are questionable with the state of the fund?
All numbers are real / inflation adjusted. If the chart says 6% and inflation over that time period was 3%, you would need a 9% actual cagr. (SP500 return over past 40 years = ~11%.)
If SS payouts are just a straight cut then the ROI of delay stays the same. If they are not a straight cut… I can’t calculate the unknown.
To each their own, you make a good argument. But SS was never meant to bequeath to heirs, it’s for making sure the elderly don’t starve when everything else runs out. Live long, better to wait, going to die younger better take it. Who knows!
Why it exists is not germane to what is mathematically superior.
The math says live long take early and invest it in an index fund.
My plan has long been to take social security at 62…but if I have even a small W2 income in 10 years when it’s my turn then they will claw it back or whatever that wierd formula is. Why spend whatever % of your portfolio for 5-8 more years when you earned your social security. Should help to manage marginal tax brackets as well instead of possible RMDs. I’m always curious do you target a bracket and convert up to it ? My target is 12%. If I were ever lower I would covert or simply take tax deferred income up to the top of 12%
I target an income level where I am eligible for a bunch of credits and health insurance premiums, which is just under 250% FPL for our household of four.
Once we are on Medicare at 65 I will target a tax bracket… either fill the 12% or 22%, depending on how much the IRAs grow between now and then.
Another reason to take it ASAP–Taxes. Between 62 and 70 my wife and I paid zero in income taxes,even though we were taking SS and had other income. At 70 (now 73) I have my RMD which raises my income a lot so that all my SS payments will be taxed at a high rate. Effectively making my break even point even higher, assuming I lived that long and didn’t care I lost the time value of tax free SS payments for 8 years.
This is interesting because it is the opposite of one of the exceptions I listed, where you might choose to delay SS in order to continue doing large Roth conversions in order to minimize RMD taxes.
You should have done Roth Conversions and paid the taxes for only 8 years 62-70. Now you will pay them the rest of your life. Whenever I hear retired people complain that they are struggling in retirement it is because they took SS early and often did not save much. Then they blame the government for their financial problems.
Interesting. And I think I may be in the the same situation eventually. I’m currently 61 as is my wife. But how did you pay zero tax even on your SS income during that time? Real Estate? We have a lot of rental properties and also a sizable traditional IRA. But I can’t really do any Roth conversions now because of our high rental income.
Will you receive benefits for your children when you start collecting? The strategies are so interesting. Thanks!!
I think the only benefit for children is if I die while they are still minors.
I will be claiming at 62 also as your allowed an additional 50% for your minor child up to your family maximum.
From Social Security Administration:
When a parent receives Social Security retirement or disability benefits, or dies, their child may also receive benefits. Under certain circumstances, a stepchild, adopted child, or dependent grandchild or step-grandchild also may qualify. To receive benefits, the child must be unmarried and: Younger than age 18.
I plan to collect at 67, but may start earlier. I think the main reason to delay taking social security is to work longer. If you need or want to work longer, then it’s better to delay.
RMD is a good point in the earlier comment.
Completely agree! Also by collecting early, that $24k the first year from SS is $24k you don’t need to take from investments…
There’s a nice calculator at https://opensocialsecurity.com/ to help decide, from a numerical perspective, when to take SS. Like another commenter said, my “plan” is to wait until 70 simply for my wife’s benefit. She’s had very little income so the calculator suggests her starting at 62 (but literally looking at $300/month in today’s dollars) then switching to the spousal benefit at 68.
I used this calc just now and agree, if one is looking to “maximize the total dollars you can be expected to receive over your lifetime” then delay is the answer. SS pays more if you wait, so this is the only possible conclusion.
It doesn’t show an alternative – What if you collected SS as early as possible and invested 100% of that income? What rate of return would you need to earn before your wife would end up better off financially after your death than if you delayed SS? That is what the IRR shows.
We actually have a similar situation – my wife has never worked in the US so her expected benefit is zero.
Jeremy, you raise a very valid point, namely that the calculators regarding when to collect SS cannot take into account how much $ could possibly be made by investing early SS proceeds. After awhile all one can do is use real world examples. In our case my wife and I are both 70. She retired at 56 and starting collecting SS at 62. I retired at 60 and started collecting at 63. After all these years of retirement, with only SS and the returns on investments as our income, our investment net worth is still a goodly amount higher than at retirement.
Granted, there is no way of calculating how much of that is due to the returns from taking SS early and offsetting expenses by taking less from investments. But there is no denying it is some part of it, so in that respect the breakeven figures are never as correct as they could be in the SS calculators.
I am currently 55 years old but retired at age 40. I have spent the last 15 years actively managing our family portfolio. My wife just retired at the age of 52.
While I have never done the math I have assumed it would make more sense to take the money early and invest it. I won’t need the money so I can be aggressive with my investments. I think I can easily double if not triple that 5% return.
The only reason I can think for not taking it early are the income limits for Roth conversions.
Income limit for Roth conversions?
I could be wrong, but I thought I read somewhere that a couple filing jointly could convert up to $25k a year tax free as long as their taxable income wasn’t more than $75k a year including the $25k conversion.
There are no limits to being able to do a Roth conversion. You could Roth convert 100% of your IRAs in one year if you wanted, even if you had high income.
The issue is how much you will pay in taxes as a result. 100% of the Roth conversion is taxable income.
What you are most likely reading is because the standard deduction for MFJ is roughly $25k, you could do that much per year tax free (assuming the other $75k income you mention is from qualified dividends and long-term capital gains, which are taxed at 0% as long as total income is less than ~$100k.)
Yes, you did a much better job of explaining it.
I shouldn’t have said “income limits”
All of my future income will be long term capital gains until I start tapping into IRA/401ks
It’s all good, we are on the same page now.
I’m doing IRA rollovers to age 65, then I plan to wait to age 70 to take SS. I figured out a long time ago, that in theory, taking it early and investing it is potentially a superior return. That depends on the past repeating in the future. However, I don’t need the money until well after age 70, and my wife had little income so the higher survivor benefit is attractive for her. Family lifespans are into the 90s so it seem a good option to me.
My concern is dementia messing up the ability to manage our money/investments. So our plan is to delay and consider SS as longevity insurance unless there is an obvious health issue in our 60s. We are mid 50’s and about to both retire on our investment income only and we will be rolling over to our Roths into our 60s. Our working income history is almost identical so starting one before the other isn’t a factor.
65 for me, as I have a bridge income from my pension, and can take RRSP and TFSA income to bring it up, its a no brainer.
We have the Canada Pension Plan, and Old Age Security. But the same thing occurs they try to convince you to leave it in until 70. so break even from 65, or 70 is 82 years old… and that didn’t even account for interest. I will do a comparison on my own site. Can I post it here?
We used the “Restricted Application” strategy which is no longer available to those born after January 1, 1954. I used the calculator and analysis offered by Lawrence Kotlikoff (Maximize my social security), Mike Piper (opensocialsecurity.com) and a couple other calculators to make our decision. They all uniformly recommended the strategy of having the lower wage earner apply for SS as soon as the spouse reached FRA, have the spouse apply for spousal benefits and then wait until the spouse reaches 70. At 70 SS benefits are maximized and the lower earner applies for spousal benefits. The numbers were a no brainer. It was like a 8%, tax free, risk free, inflation adjusted return – something I did not have any confidence of achieving.
We did not need the cash flow as soon as we were eligible for SS and are totally pleased with the way it worked out. Its a pity that the strategy is no longer available.
Hello, I understand, and agree with your reasoning except for one thing. I am still working, so I have not activated my SS because I don’t want to pay half of my income to taxes. I got divorced at and didn’t want anything from my ex, so I had to play catch up on my retirement income, as I didn’t have any,LOL! I caught up in 7 years and have built up my retirement income so I will be taking my SS at 66.8 years. What are your thoughts?
The one thing that stops me from planning to claim early is the penalty for excess earnings. Even if I don’t have to work, I can’t imagine I would want to hang up my self-employment hat at age 62, and you can’t earn much before they start docking your benefits. They say they recalculate when you hit full retirement age so you somehow make that lost money back, but I don’t quite understand how that works.
I, too, would rather spend that money early than late. I can get more enjoyment out of that $2K in my early 60’s than $4k in my 80’s.
Excellent point that many people overlook in the SS discussions, Jason, namely one’s health. The decision-making in the case of taking early or late distributions is usually in one’s 50s or early 60s, and many people are still physically fine and active during those years. Start to get towards 70 and beyond and things can change dramatically for many, which will effect how much enjoyment you can get out of those extra $.
This is the *KEY* reason to take it at 62. I know many 70+s, and their spending goes down annually.
The ‘utility’ value of money is much higher when 62 than 70.
All great points. EarlyRetirementNow did a historical analysis and found that there wasn’t much difference between the two options.
Some caveats that I have is that I wouldn’t look at average life expectancy if you are in above average health, education, environment etc. For example the average life expectancy in California is 7.1 years older than Mississippi. The average american is overweight so if you’re normal weight then you can expect to live a little longer. Being FI you have less stress and more time to exercise, eat better and socialize which often leads to longevity.
Also for those with one spouse having a much larger SS benefit than the other should look at joint life expectancy rather than individual since the lower earning spouse can take the higher ones benefit in their demise. The median joint life expectancy at 65 is early 90s which is much higher than the individual expectancy that you mentioned of 7/10 not making it to 88.
So for those caveats… I’m not looking at any life expectancy, pre-existing health conditions, exercise habits, or marriage status.
Look at the IRR chart… no matter how long you live, you don’t outperform the SP500 by delaying.
“no matter how long you live, you don’t outperform the SP500 by delaying”
You don’t outperform the historical average SP500 return. Your statement is based on a deterministic scenario comparison however one side has no volatility and the other has 100% equity volatility. It is not apples to apples and it should include the probability of various SP500 rerurns and the joint life expectancy conditional of being healthy at 62.
It is not apples/apples nor should it be.
There is no 20 year period during which the SP500 had a negative return. The 20 year return on SS delay is guaranteed to be negative
Here is an example where delaying would come out on top once it exceeds 1.5% ROI in your graph (at about age 82) which is above 1.49% real return between 1965 and 1984.
From SP500 return calculator:
If you invested $100 in the S&P 500 at the beginning of 1965, you would have about $443.61 at the end of 1984, assuming you reinvested all dividends. This is a return on investment of 343.61%, or 7.73% per year.
This lump-sum investment beats inflation during this period for an inflation-adjusted return of about 34.49% cumulatively, or 1.49% per year.
For almost all years there is a point where delaying comes out on top, typically 25+/- years later (62+25 = 87.) For periods with high inflation, that is a little sooner. For periods with low inflation, a little later.
What about those of us using ACA insurance and receiving subsidies due to managing for a low AGI? One can start Social Security payments at 62 but not qualify for Medicare until 65. Seems like delaying SS payments until 65 could make sense in that scenario. Thoughts?
Yes, I will add this as an exception
If you are in favor of taking SS, which is inflation adjusted, at 62, I imagine you would be doubly in favor of taking a non-inflation adjusted pension as early as possible too?
To wit, my wife and I are federal government employees. We would be eligible to take our full pensions at age 62, but could take them as early as 57 with a 5% reduction in payout per year taken, so taking them at 57 would be a 25% reduction. They are not adjusted for inflation until age 62, afterwhich they get an adjustment of the CPI each year (or the CPI minus 1% if the CPI is over 2%). Thoughts on taking it at 57 instead of 62?
Thanks!
I plan on delaying (within 62-70) until there is a bear market. If that happens when I’m 62, I’ll take it. If that doesn’t happen until I’m 70, I’ll wait.
Probably doesn’t make much of a difference but it helps me sleep at night knowing I wont have to draw down as many assets when things aren’t going well.
This is smart.
The income for 70+ is 44076-25032
“The income for 70+ is 44076-25032”:
= $19,044 / y
Which explains how your IRR numbers were calculated.
Second graph shows 70+ SS received as $44,076 / y
$44,076 / y is the correct number to use (not $19,044 / y).
We are calculating the IRR of delaying SS, in which we exchange one income stream for another. At age 70+ we receive 44,076 and also don’t receive 25,032.
We are calculating the IRR of an investment.
From 62 to 69 SS not received (forgone) is an investment of $25,032 / y required to receive $44,076 / y from 70 on.
‘What is the IRR of an investment of $25,032 / y paid in for the first 9 y where after the investment will pay out $44,076 / y for N years’
Could these posts be moved to my post of January 20, 2024 at 4:16 pm?
We are entitled to a lifetime income stream starting at age 62. In your math, you collect 8 years of it and then ignore the rest.
We are entitled to a lifetime income stream starting at age 62. In your math, you collect 8 years of it and then ignore the rest”:
Opting not to collect SS until 70 means not collecting $25,032 / y for 8 y, $200,256.
Equivalent to paying cash into an investment, the capital cost of the investment; a negative cash flow.
On becoming 70 the entitlement is $44,076 / y; a positive cash flow.
As per table in my post of January 20, 2024 at 4:16 pm.
In your arithmetic, on becoming 70 you show only ($44,076 – $25,032) = $19,044 / y being collected. Whereas you are entitled to $44,076 / y.
*sigh*
It is your position that delaying SS yields a long term rate of return in excess of 13%? Real. That makes sense?
If so, we can probably just agree to disagree.
*sigh*:
If invest $25,032 / y for 8 y, $200,256 real, then withdraw $44,076 / y for (93 – 70 + 1) = 24 y, $1,057,824, then IRR is 13%.
Real using your real inputs.
“It is your position that delaying SS yields a long term rate of return in excess of 13%? Real.”
I have no reason to doubt accuracy of Excel calculation of IRR.
Life expectancy: male 62 is 19 y:
https://www.ssa.gov/oact/STATS/table4c6.html
80.423% live to 62.
9.364% live to 93.
Mortality cliff between.
Only 1 in 7 get 13% return.
Excel is great, but GIGO.
There are numerous articles that highlight the error you are making. Here is one, but you can google more.
“There are numerous articles that highlight the error you are making. Here is one”:
https://www.financialplanningassociation.org/article/journal/APR15-return-investment-delaying-social-security-beyond-age-62
Calculates the MARGINAL IRR.
My table post January 20, 2024 at 4:16 pm calculates the TOTAL IRR.
Both are correct.
The person of average life expectancy from 62, ~86, to benefit from taking SS at 62 would have to invest in an alternative investment with a guaranteed, real IRR greater than 4%, with no volatility, over 24 y. Lucky & scarce.
The TOTAL IRR of SS taken from 70 to 86 would be 12.2%.
“person of average life expectancy from 62, ~86”:
USA:
Male 19.03 y = 81, MARGINAL 1.32% TOTAL 10.6%
Female 22.04 y = 84, MARGINAL 3.19%, TOTAL 11.7%
Makes sense to me! 5% isn’t a high hurdle at all, especially with the risk-free rate at 5% right now.
I’m glad my mom took SS ASAP. I’m probably going to take SS by 66 myself. Just gotta hope the body holds up!
I got to the same conclusion by using the mortality tables published by the Society of Actuaries. The actuarial present value assigns a probability (that I’ll remain alive) to each year of cash flow and then discounts it back to the present day. Using a 5% discount rate, a white collar male can expect ~10% better value to start early, while a white collar female can expect ~5% better value. Breakeven discount rate is in the 3-4% range.
As you pointed out, someone who expects to outlive the average white collar American might be better off deferring the start date, as would someone very bearish on investment returns. But however you look at it, those are bold assumptions that don’t even result in a significantly better deal to defer the start date (far far from the 76% difference that many are led to believe).
This is beautifully stated
Even if you or spouse live a long time, deferring isn’t a significantly better deal
Exactly. Everyone feels they will be that outlier that lives longer than the vast majority of the population, and approach taking SS from that perspective. But the payment tables by age are done using actuarial/insurance experts, and how many people think they are truly smarter than those people when it comes to computing returns and life expectancy?
The optimal strategy depends on what aspect is being maximized. You are optimizing for maximum inheritance that can be passed down to your kids and your conclusion to claim at 62 is valid (I think not including the benefit of Roth conversions makes the analysis incomplete and it is even more impactful for those retiring in late 50s and 60s as they dont have many years before 62 to take advantage of Roth conversions, there could also be IRMAA implications for Medicare when RMDs start). I choose to optimize for reduction in severity of impact to my lifestyle in a low market return scenario because that is where the risk of running out of money in retirement lies. Hence, I plan to claim at 70 to maximize spouse survivor benefit and take advantage of Roth conversions before 70 to lower overall lifetime tax payments. If markets deliver average returns or higher, the decision on when to claim SS will not matter much and I am willing to give up the additional upside. I am also ok with -100% return if I die before 70. If health takes a turn for the worst, one can always claim SS right away to account for the change in circumstances. I also like the optionality mentioned in a previous comment on claiming before 70 in case of a market downturn.
Great perspective, HB.
I’m with you, after I read this mymoneyblog article that linked to this death calculator: https://engaging-data.com/will-money-last-retire-early/
I wrote about that calc: You Will Die Before You Run Out of Money
My arithmetic gives a different result for Internal Rate of Return (IRR) vs Age.
Input seems same.
Columns:
Age
SS -forgone +received
Cumulative Cash
IRR
61 0
62 -25,032 -25,032 -100.0%
63 -25,032 -50,064 -100.0%
64 -25,032 -75,096 -100.0%
65 -25,032 -100,128 -100.0%
66 -25,032 -125,160 -100.0%
67 -25,032 -150,192 -100.0%
68 -25,032 -175,224 -100.0%
69 -25,032 -200,256 -100.0%
70 44,076 -156,180 -100.0%
71 44,076 -112,104 -16.5%
72 44,076 -68,028 -7.5%
73 44,076 -23,952 -2.1%
74 44,076 20,124 1.5%
75 44,076 64,200 4.0%
76 44,076 108,276 5.9%
77 44,076 152,352 7.3%
78 44,076 196,428 8.4%
79 44,076 240,504 9.3%
80 44,076 284,580 10.0%
81 44,076 328,656 10.6%
82 44,076 372,732 11.0%
83 44,076 416,808 11.4%
84 44,076 460,884 11.7%
85 44,076 504,960 12.0%
86 44,076 549,036 12.2%
87 44,076 593,112 12.4%
88 44,076 637,188 12.5%
89 44,076 681,264 12.7%
90 44,076 725,340 12.8%
91 44,076 769,416 12.9%
92 44,076 813,492 13.0%
93 44,076 857,568 13.1%
94 44,076 901,644 13.1%
95 44,076 945,720 13.2%
96 44,076 989,796 13.2%
97 44,076 1,033,872 13.3%
98 44,076 1,077,948 13.3%
99 44,076 1,122,024 13.3%
100 44,076 1,166,100 13.3%
I plan to collect as early as possible. I found the Daniel Amerman article. “Making Optimal Social Security Claiming Decisions” interesting. The article is a bit dated but covers variables I had never considered.
Linking that article here because it is really good.
Making Optimal Social Security Claiming Decisions
Curry, I would have put the link but some sites remove links, so I wrote it out. Thank you for posting it. I have not found any article that looked at the issue the way he did. If a haircut does happen in about 10 years, I want to be one already taking and with a lower payout due to taking at 62 in case they do a flat % cut or a higher % for those with bigger monthly payout amounts. Either way my $/mo cut will be less.
I haven’t decided yet. (I’m only 56 right now.) I don’t have kids, so leaving an inheritance isn’t an issue for me. From now until beginning taking SS, I’m looking at years of 0 federal tax, and I can avoid some tax on traditional IRA and 401(k) accounts by keeping my annual distributions around the standard deduction amounts. (The rest comes from long term cap gains in the 0% zone.) And the whole question for me is somewhat academic, because no matter what strategy I take, the most likely outcome is that I’ll die with money in the bank. So I’ll probably let the market decide for me – if we’re in a bull market that’s doing great when I hit 62, I’ll probably continue to draw from my savings/investments; but if the market turns down and I want to minimize the shares I need to sell, that’s when I would file for SS to turn on that income stream.
Same here Dean, I’ll only take it when I need it and I certainly don’t want to pay taxes on it. There is also ACA subsidy I don’t want to loose so I will be waiting, at the least till I can get Medicare.
“no matter how long you live, you don’t outperform the SP500 by delaying”
You’re right about “on average”, but delaying does mitigate the worst case scenarios that retirees face e.g. turning 62 in 1929/1968 and living for another 30+ years. Most retirees (and their heirs) are already in a great spot if they save 25x expenses and get average returns or die quickly . All of their risk is based on the scenario of getting below average returns and living a long time, which delaying mitigates. Its a trade off for accepting lower overall average returns (but still pretty good) in order to make the worst case scenarios a little better.
Also most retirees aren’t 100% in stocks, so maybe something more comparable would be 70/30 portfolio which 6% returns might be a reasonable prediction going forward given the historically high stock market valuations and relatively low long term bond yields.
Others have said in a 1929 event they would take early because it means you don’t have to sell as many assets when the markets are down, which is the opposite of what you are saying here. It is interesting that people will look at the same data and decide that 2 opposite actions are best.
You can pick an interest rate on the chart… 6% puts you out at a point when 95% of people have died and your rate of return is no better from delay. It also isn’t necessary to be 100% in stocks… just where you would put 100% of the SS income you take early, which doesn’t change the rest of your portfolio.
Take a look at Early Retirement now’s SWR series #59, he covered the 1929 scenario with a hypothetical $2million portfolio wtih Taking $3100 at 70 passed $1750 at 62. Taking early did perform better initially but eventually differing outpaced it at age 88.
In the 1968 high inflation scenario differing passed early SS at age 81
So if you’re looking to mitigate longevity risk and are in above average health, differing is the better option
Great analysis! Two thoughts:
1) Taking SS early and investing in a superior returning asset (index funds) sounds great on paper. The problem is the lack of discipline some might have and thus sabotage their financial well being. “This month’s SS goes towards my new 4-wheeler but I’ll definitely start investing next month’s payment.”
2) SS is an excellent insurance policy against longevity risk and the possibility of a sustained downturn in the markets. Even if the probability that I live a very long time and the markets go into a historic downturn is very small, wouldn’t it be a great diversifier to wait to 70 and minimize that risk? That’s where I’m leaning at the moment, but always open to changing gears.
I do like the idea of buying a 4-wheeler in my 70s
Social Security does provide those benefits in the very long term, but the rate of return on delay is negative for 25 years +/- and beyond that isn’t that great (but totally respectable and with no volatility.)
See the next post for more: Why Early Social Security Provides the Greatest Spousal Benefit
Waiting as my wife is 12 years younger and will continue collecting the higher benefit long after I’m gone….
Your anticipated social security benefit may be off because the numbers you are referencing are based on current projections of average earnings over 30 years of work. If you’re 50 and haven’t worked in 10 years then you may only have 20 or fewer years recorded.
It is based on my actual earnings history with zero future earnings. See chart and link at top of this post.
For ROI calculations though all that matters is the ratio of early vs delayed which is the same for everyone.
If you are investing the proceeds from taking SSN early, then where are you getting income? i.e. I have to take those $ out of the market to pay bills. Isn’t the better calculation the compound effect of letting the portfolio grow more by taking SSN early and not having to take that amount off the portfolio?
Money is fungible so these two things are the same.
Hi,
Thanks for providing this. I retired early from my job in tech sales at 57 and I’m 61 and my wife is 62. She still works a low level job with health benefits. We were facing this decision. Thanks for your work, I have enjoyed your work over the years.
One comment, on your chart.
After age 67 shouldn’t the amount in the bar graph be the 24K, or only the amount over what you are already collecting?
1. Always enjoy your blessedly brief works.
2. An inflation adjusted pension from the guys who print the money and collect the taxes is “different” from anything else I have. Measure 6x, cut once.
3. I’ve held off a year, but always seem to return to this topic. At this point I am taking CGs (but I was a market underperformer–grrr), converting assets, doing back door Roth, all of which have tax friction and don’t wanna zonk my ACA subsidies.
4. Both Ps are ~90 so I’ll live to ~95 (given the run rate)? So be patient?
5. Too bad we can’t do do overs if we guess wrong!
Is there really a wrong answer here? The comments are really interesting because there are people looking at the same choice and same data, with different conclusions as to what is “best.”
I could see waiting until at least 65 when Medicare replaces ACA, or with some longevity genes pushing it all the way to 70. Nothing wrong with 4-5% real return with no volatility.
This seems a more ‘intuitive’ presentation then IRR; showing 62+ and 70+ cumulative cashflows and their difference:
https://freeimage.host/i/Jab2vbR
Notice the cross over at age 80.
Columns:
Age
SS received 62+
SS Cumulative 62+
SS received 70+
SS Cumulative 70+
70+ minus 62+
61 0 0
62 25,032 25,032 0 0 -25,032
63 25,032 50,064 0 0 -50,064
64 25,032 75,096 0 0 -75,096
65 25,032 100,128 0 0 -100,128
66 25,032 125,160 0 0 -125,160
67 25,032 150,192 0 0 -150,192
68 25,032 175,224 0 0 -175,224
69 25,032 200,256 0 0 -200,256
70 25,032 225,288 44,076 44,076 -181,212
71 25,032 250,320 44,076 88,152 -162,168
72 25,032 275,352 44,076 132,228 -143,124
73 25,032 300,384 44,076 176,304 -124,080
74 25,032 325,416 44,076 220,380 -105,036
75 25,032 350,448 44,076 264,456 -85,992
76 25,032 375,480 44,076 308,532 -66,948
77 25,032 400,512 44,076 352,608 -47,904
78 25,032 425,544 44,076 396,684 -28,860
79 25,032 450,576 44,076 440,760 -9,816
80 25,032 475,608 44,076 484,836 9,228
81 25,032 500,640 44,076 528,912 28,272
82 25,032 525,672 44,076 572,988 47,316
83 25,032 550,704 44,076 617,064 66,360
84 25,032 575,736 44,076 661,140 85,404
85 25,032 600,768 44,076 705,216 104,448
86 25,032 625,800 44,076 749,292 123,492
87 25,032 650,832 44,076 793,368 142,536
88 25,032 675,864 44,076 837,444 161,580
89 25,032 700,896 44,076 881,520 180,624
90 25,032 725,928 44,076 925,596 199,668
91 25,032 750,960 44,076 969,672 218,712
92 25,032 775,992 44,076 1,013,748 237,756
93 25,032 801,024 44,076 1,057,824 256,800
94 25,032 826,056 44,076 1,101,900 275,844
95 25,032 851,088 44,076 1,145,976 294,888
96 25,032 876,120 44,076 1,190,052 313,932
97 25,032 901,152 44,076 1,234,128 332,976
98 25,032 926,184 44,076 1,278,204 352,020
99 25,032 951,216 44,076 1,322,280 371,064
100 25,032 976,248 44,076 1,366,356 390,108
I’m for whatever metric helps people reach a useful conclusion.
Cumulative cash flows will show you which option results in more $, but more $ doesn’t help me compare to alternatives like IRR. A 1% rate of return is more than 0% but I would rather put the money somewhere than I can earn 2%.
The cumulative graph posted above
[ https://freeimage.host/i/usa-ss-70-minus-62.Jab2vbR ]
shows how much an individual surviving to 100 would receive from 62 or 70.
An unlikely prospect for the vast majority.
An individual may die earlier by accident than their health might indicate or die later by accident avoidance or by slower than expected decay or by serendipitous medical intervention. [“Who by Fire”]
This graph shows how much cumulatively the average USAn man is likely to receive by multiplying survivorship proportion for each year greater than 62:
https://freeimage.host/i/usa-likely-ss-70-minus-62.Jc2hy0P
[ Using https://www.ssa.gov/oact/STATS/table4c6.html ]
There is negligible difference between Likely SS Cumulatively 62+ and 70+ after 80.
Which leaves unabated the large Likely SS Cumulative loss where SS commencement is delayed from 62 to 70.
Seems clear: ‘Do not delay.’
(Table too wide.)
Now that is a cool chart. You found a great way to integrate the longevity data!
“Seems clear: ‘Do not delay.’”
What of those with other income in addition to SS?
Would SS payments increase marginal tax rate?
Would waiting to retire reduce tax taken?
“way to integrate the longevity data”:
Of 100,000 live births, 80,423 survive to 62 and 545 to 100.
On average a 62er has a 545 / 80,423 = 0.68% chance of claiming a year’s SS payments.
Add 0.68% of a year’s SS payments to Likely Cumulative SS payments and similar for years between?
Your examples and explanations talk about investing 100% of the early SS in an index fund. But it shouldn’t be 100% because (at least at a certain income) the benefits are taxed by the Feds (and in some cases the State). Is there an income limit where having to pay the taxes up front before investing the rest makes waiting a better choice for most people?
That was my question as well, if I am investing the SSN in and index fund and I am simultaneously taking out of an index fund in a brokerage account to pay bills, I will have capital gains which will also make 85% of SSN taxable. The IRR should reflect after tax on capital gains, still probably wins.
Money is fungible and a dollar is a dollar… investing SS is equivalent to keeping an equal amount of money invested.
Let’s say I have $1 million in an index fund and I plan to spend $10k on a small boat.
Completely by coincidence I happen to receive $10k from the Social Security Administration. What’s the difference between:
a) spending the $10k SS on a boat directly
b) selling $10k of stock (now have $990k), buying the boat, then buying $10k of stock with the SS (now have $1 million)
Not really, no.
I thought it’s good to math stuff up, since it depends on the personal situation.
I found a calculator, plugged our numbers in, it told me that husband should wait until almost 70 but I should take right away at 62 (since my number will be a lot less). We have almost 6 yrs between us which is another factor
Most of these calculators don’t account for the big pile of cash that your husband would have at age 70 had he started collecting at 62.
See this post for more: Why Early Social Security Provides the Greatest Spousal Benefit
Interesting..I always suspected and had this same feelings and math in mind.
I intend to retire outside the US like you did when you lived in China for a few years. How would that impact taxes and benefits? A post about that would be great!! Thanks, big fan. Dan
Kickstart Your (Early) Retirement