Author Topic: This should have been obvious to me: firecalc/cfiresim and SS start date  (Read 4096 times)

Roothy

  • Stubble
  • **
  • Posts: 120
I have always plugged in age 70 as my start date for Social Security in cfiresim and firecalc, for a planned retirement age of around 50.  I always figured this was the prudent and conservative thing to do.

Then someone made a passing comment about how the real value of SS for early retirees is to protect a dwindling nest egg in a bad retirement start year (in other words, to mitigate for a poor sequence of returns).  This made me realize that I should be plugging age 62 in as a start date, even though it has lower predicted monthly payments.  This improved the success of my retirement plan by about 8%.

If my retirement is going poorly within the first 12 years, then taking SS early allows me to recover, or at least not deplete as fast.  If my retirement is going well, then it doesn't matter whether I take SS at 62 or 70--SS is just icing on the cake.  So for planning purposes of avoiding a bad retirement outcome, only a 62-starting-age is even relevant.

This should have been obvious, I know.  But if it wasn't obvious for you, either, rerun your cfiresim/firecalc and you might be pleasantly surprised.  I think this is going to shave about a year and a a half off my retirement start age.

SeattleCPA

  • Handlebar Stache
  • *****
  • Posts: 1488
  • Age: 59
  • Location: Redmond, WA
    • Evergreen Small Business
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #1 on: February 27, 2017, 11:49:52 AM »
William Bernstein, the physician, investment advisor and author, did a review of my little ebook, "Thirteen Word Retirement Plan." My little book BTW covers similar territory as his book "If You Can..." and he gave me a very friendly review...

But there was one constructive criticism he had... he thought the opposite of what you suggest... he thought we should all delay SS until age 70 so as to bump up our SSI for the worst case scenario. (He suggested I add this advice to the next version of my book ... which I promised i'd do.)

I have a lot of trust in cfiresim and firecalc... so I'd believe the numbers they tell me... but I'd trust Dr. Bernstein more. He's a pretty smart fellow.

Roothy

  • Stubble
  • **
  • Posts: 120
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #2 on: February 27, 2017, 11:57:52 AM »
I'll delay until 70, too, if I can.  But not if I'm in a bad sequence of returns.  At that point, the value of even a lower payout when taking SS early is immeasurable.

So for planning purposes (that is, deciding whether you have enough to pull the plug), putting 62 into the firesim/cfirecalc programs is better, because what you are trying to predict with those calculators is failure.

In other words, "plan" for 62 for purposes of the calculator, and deciding whether you have enough.  But if you are in one of the vast majority of retirement start dates which do just fine, actually take SS at 70.  It is just icing on the cake at that point.

AdrianC

  • Pencil Stache
  • ****
  • Posts: 993
  • Location: Cincinnati
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #3 on: February 27, 2017, 11:59:34 AM »
I have always plugged in age 70 as my start date for Social Security in cfiresim and firecalc, for a planned retirement age of around 50.  I always figured this was the prudent and conservative thing to do.
...
I think this is going to shave about a year and a half off my retirement start age.

Using age 70 is the prudent and conservative thing to do. Using 62 gives you a lower FI number, which is less prudent and less conservative. NTTAWWT.

Dr. Bernstein is a great guy, but he's a bit risk-averse for this crowd.

CCCA

  • Pencil Stache
  • ****
  • Posts: 569
  • Location: Bay Area, California
  • born before the 80's
    • FI programming
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #4 on: February 27, 2017, 12:11:17 PM »
I only run my cfiresim scenarios until age 65.  This is because a combination of SS, paying off mortgage, pension and rental income will make my expenses less than my annual income, so the size of my stache is irrelevant at that point.  Thus the uncertainty in our retirement is entirely in the first 20 years.  If we can make it to age 65, we should have enough income from these sources that we don't even need to touch our retirement accounts and other investments (i.e. they could be zero), for the rest of our lives (i.e. infinite retirement).


So as a result, I always put 62 as my SS date into cfiresim in order to gauge whether I can make it to 65 or not. 

Roothy

  • Stubble
  • **
  • Posts: 120
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #5 on: February 27, 2017, 02:11:14 PM »
I was very unclear in my original post, and so there is confusion by one or two people about what I am proposing.  I'm not making any claims about when it makes sense to actually take SS.  (Actuarily, SSA is designed so that for the average person, it doesn't matter.  For the types who frequent these groups, who are likely to live longer than the average American, it should be 70.  But again, my argument has nothing to do with that.)

Instead, I'm making a claim about what you should put into a sophisticated retirement calculator like cfiresim or firecalc.  I argue that those calculators are about telling you a worst case scenario, and whether your nest egg would survive it.  If that is what you are using it for (and I think most people are), then plugging 62 is more sensible than plugging in 70, even if in the vast majority of cases you would intend to take SS at 70.

This is, in other words, about what age you should RETIRE and/or what your nest egg should be before you do so--not about when you should actually take SS.
« Last Edit: February 27, 2017, 02:34:50 PM by Roothy »

AZryan

  • Stubble
  • **
  • Posts: 153
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #6 on: February 27, 2017, 03:24:59 PM »
Might want to also rethink the idea that 62 will actually be available. Could be 64, 65, 66? by the time someone here hits 62.

But I think the point stands up well enough that calculating for it at the earliest date best assures your retirement chances.

AdrianC

  • Pencil Stache
  • ****
  • Posts: 993
  • Location: Cincinnati
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #7 on: February 27, 2017, 04:07:54 PM »
Might want to also rethink the idea that 62 will actually be available. Could be 64, 65, 66? by the time someone here hits 62.

But I think the point stands up well enough that calculating for it at the earliest date best assures your retirement chances.

It'll give less failures for a given withdrawal rate. Is that what you mean?

SwordGuy

  • Walrus Stache
  • *******
  • Posts: 5913
  • Location: Fayetteville, NC
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #8 on: February 27, 2017, 07:37:02 PM »
Might want to also rethink the idea that 62 will actually be available. Could be 64, 65, 66? by the time someone here hits 62.

But I think the point stands up well enough that calculating for it at the earliest date best assures your retirement chances.

This. ^

Anyone who's decades away from getting it can absolutely count on the collection dates being pushed back.   If the early date isn't pushed back the payout % will drop.

TomTX

  • Magnum Stache
  • ******
  • Posts: 3660
  • Location: Texas
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #9 on: February 28, 2017, 05:34:49 AM »
William Bernstein, the physician, investment advisor and author, did a review of my little ebook, "Thirteen Word Retirement Plan." My little book BTW covers similar territory as his book "If You Can..." and he gave me a very friendly review...

But there was one constructive criticism he had... he thought the opposite of what you suggest... he thought we should all delay SS until age 70 so as to bump up our SSI for the worst case scenario. (He suggested I add this advice to the next version of my book ... which I promised i'd do.)

I have a lot of trust in cfiresim and firecalc... so I'd believe the numbers they tell me... but I'd trust Dr. Bernstein more. He's a pretty smart fellow.

Even very smart fellows have blind spots. Social Security has a lot of emotion and vast amounts of "conventional wisdom" tied up in it.

Show him the numbers and ask him again.

TomTX

  • Magnum Stache
  • ******
  • Posts: 3660
  • Location: Texas
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #10 on: February 28, 2017, 05:38:23 AM »
There's an almost perfect crossover point at Age 82 (actually it is perfect in the comparison between 62 and 67, and it's close for 70) at which point it will finally become worth it to have delayed taking SS (assuming the point at which you stop making income is the same either way). If you were to expire before age 82, it would have been better to take it at 62. (That's not to say it's better or worse for your overall nest egg odds, but just that if you want "the most" money possible from SS.) So if you don't expect to live into your mid-80s and beyond, it makes sense to take it as early as possible anyway.

I see that as a common mistake with Social Security - not thinking "how can I withdraw to benefit my life the most?" - but instead "How can I get the highest absolute dollar from the system?"

AdrianC

  • Pencil Stache
  • ****
  • Posts: 993
  • Location: Cincinnati
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #11 on: February 28, 2017, 08:31:29 AM »
I tried it. My SWR went from 4.19% taking SS at age 70 vs 4.27% taking SS at age 62 (95% success rate, 40 years, FIRE now).

An interesting exercise. Doesn't affect me at all.

Roothy

  • Stubble
  • **
  • Posts: 120
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #12 on: February 28, 2017, 10:01:16 AM »
Huh!  My results are bigger.  I'm 4 years from 50 (age at which I'd start into my nest egg) and 16 years from 62 (age at which, if things went badly, I'd start withdrawing from SS).  I'm guessing you are younger than me?

I get a low of a 3% difference up to a high of almost 8%, depending on what I assume about how big the SS payments will be, what my AA is, etc.

Another thing that moved the needle *a lot* was to set a floor of about 10% below my usual withdrawal for bad years.  That's a reduction in "pay" I could sustain for many years, and it increased success rate bigly.

Both the "new" SS technique plus the floor (no ceiling) moved me from an 83.65% to 96.15% success rate.  Wow!

(I also just realized we are looking at it from different directions.  I'm trying to figure out success rates, you are trying to figure out maximum spend rates.)
« Last Edit: February 28, 2017, 10:05:47 AM by Roothy »

AZryan

  • Stubble
  • **
  • Posts: 153
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #13 on: February 28, 2017, 10:28:58 AM »
It'll give less failures for a given withdrawal rate. Is that what you mean?

No. I was saying that in the future people probably won't be able to start collecting SS at age 62. It'll probably be moved up to 63, 64, etc... You should probably factor that in depending on your age.

Then again, in several decades we might have a universal basic income, and/or AI and robots doing most of the work, and the concept of 'retirement' won't really mean anything compared to what we're talking about now. And that seems pretty likely based on technology trends and the projections of people creating that technology.

What I always worry about is the 'transition'. That's were riots and wars and madness occur.

AdrianC

  • Pencil Stache
  • ****
  • Posts: 993
  • Location: Cincinnati
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #14 on: February 28, 2017, 11:54:38 AM »
Huh!  My results are bigger.  I'm 4 years from 50 (age at which I'd start into my nest egg) and 16 years from 62 (age at which, if things went badly, I'd start withdrawing from SS).  I'm guessing you are younger than me?

I'm 4 years older. It's stache size compared to expected SS amount, probably. I've spent some of my work life outside the US. Mrs C has a pension coming too, which is more than her SS at age 62.

AdrianC

  • Pencil Stache
  • ****
  • Posts: 993
  • Location: Cincinnati
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #15 on: February 28, 2017, 11:59:31 AM »
It'll give less failures for a given withdrawal rate. Is that what you mean?

No. I was saying that in the future people probably won't be able to start collecting SS at age 62. It'll probably be moved up to 63, 64, etc... You should probably factor that in depending on your age.

May be so.

What I was referring to was your comment:

"But I think the point stands up well enough that calculating for it at the earliest date best assures your retirement chances."

It gives less failures for a given withdrawal rate. Or a better withdrawal rate for a certain success rate.

It doesn't improve your chances of a successful retirement at all. Saving more money does that.

Eric

  • Magnum Stache
  • ******
  • Posts: 4061
  • Location: On my bike
Re: This should have been obvious to me: firecalc/cfiresim and SS start date
« Reply #16 on: February 28, 2017, 03:08:17 PM »
There's an almost perfect crossover point at Age 82 (actually it is perfect in the comparison between 62 and 67, and it's close for 70) at which point it will finally become worth it to have delayed taking SS (assuming the point at which you stop making income is the same either way). If you were to expire before age 82, it would have been better to take it at 62. (That's not to say it's better or worse for your overall nest egg odds, but just that if you want "the most" money possible from SS.) So if you don't expect to live into your mid-80s and beyond, it makes sense to take it as early as possible anyway.

I see that as a common mistake with Social Security - not thinking "how can I withdraw to benefit my life the most?" - but instead "How can I get the highest absolute dollar from the system?"

Totally agree.  Trying to "maximize your payout" is the wrong goal.

Monkey Uncle

  • Handlebar Stache
  • *****
  • Posts: 1500
  • Location: West-by-god-Virginia
I had a similar realization a couple of months ago.  For those of us of a certain age, it seems to make sense to claim SS at 62, if your metric of interest is maximizing your safe annual spend.  Not sure it would work for the younger folks, but I'd be interested to see someone run their numbers.

http://forum.mrmoneymustache.com/welcome-to-the-forum/what-happens-to-social-security-payments-before-and-after-retirement-date/msg1383694/#msg1383694

dude

  • Handlebar Stache
  • *****
  • Posts: 2368
Most claiming strategies also take into account a spouse's SS benefits. If there's a decent age gap and/or earning status between spouses, then the spouse's secure retirement (i.e., should you predecease her (or him)) might well become the more significant consideration.