Author Topic: Getting the most out of social security benefits  (Read 6878 times)

NestEggChick (formerly PFgal)

  • Stubble
  • **
  • Posts: 238
  • Location: Boston-ish, MA
    • Nest Egg Chick
Getting the most out of social security benefits
« on: June 19, 2014, 10:02:10 AM »
The rules will probably change before I'm ready to claim SS, but I'm guessing quite a few people on this forum could benefit from this. So enjoy!

http://m.sfgate.com/business/networth/article/A-twist-could-help-singles-Social-Security-5562797.php

Poorman

  • Bristles
  • ***
  • Posts: 260
  • Age: 46
  • Location: Orange County, CA
Re: Getting the most out of social security benefits
« Reply #1 on: June 19, 2014, 01:39:45 PM »
It's best to start collecting at the youngest age possible, even if you don't need the money.  The reason is that even a small return of 2% per year from investing the SS payments beats the 8% yearly increases from waiting to collect it.  Somebody that waits until age 70 to collect has a huge mountain to climb before they break even with the person that started collecting at age 62.  That's 8 years of payments they have to catch up to.  If the person that started collecting at age 62 simply invested all those payments, there's no way the person that starts collecting at age 70 can ever catch up, unless the investment returns are less than 1.5% or negative.  I wrote about it in a little more detail here:

http://forum.mrmoneymustache.com/mustachianism-around-the-web/another-save-10retire-by-70-article-ny-times/msg299283/#msg299283

The plan that you linked to sounds like the worst of both worlds to me because it's retroactively paying benefits that you could have collected and invested all along.  Essentially, it guarantees a 0% return on those retroactive payments.  It would be better to "self-insure" by taking the payments at the youngest possible age and investing them conservatively.

MDM

  • Senior Mustachian
  • ********
  • Posts: 11477
Re: Getting the most out of social security benefits
« Reply #2 on: June 19, 2014, 05:50:57 PM »
Perhaps someone could you point out what is wrong in the Excel calculations shown below.  To test, copy the table and paste (as text) into cell A1 in a blank spreadsheet.  Calculations assume no taxes and that all money is invested with annual returns given in cell F1 (2% in the table below).

If there is nothing wrong, the "code" below shows the age (ignoring months) at which "deferring to 70" beats "starting at 62", assuming benefits grow at 8%/year.  In other words, if you think you will live longer than "age", defer.  Otherwise, start at 62.

Code: [Select]
Return  Age
0% 79
1% 80
2% 82
3% 82
4% 84
5% 86

Also see http://forum.mrmoneymustache.com/ask-a-mustachian/what-age-do-you-plan-on-taking-your-ssi/ for more details and nuances about this topic.  This only scratches the surface.


AgeAnnual62BoY1Int1EOY10.02Annual70BoY2Int2EOY2
6210000=B2=C2*$F$1=C2+D2
=A2+110000=B3+E2=C3*$F$1=C3+D3
=A3+110000=B4+E3=C4*$F$1=C4+D4
=A4+110000=B5+E4=C5*$F$1=C5+D5
=A5+110000=B6+E5=C6*$F$1=C6+D6
=A6+110000=B7+E6=C7*$F$1=C7+D7
=A7+110000=B8+E7=C8*$F$1=C8+D8
=A8+110000=B9+E8=C9*$F$1=C9+D9
=A9+110000=B10+E9=C10*$F$1=C10+D10=B10*(1.08)^(A10-A2)=G10+J9=H10*$F$1=H10+I10
=A10+110000=B11+E10=C11*$F$1=C11+D11=G10=G11+J10=H11*$F$1=H11+I11
=A11+110000=B12+E11=C12*$F$1=C12+D12=G11=G12+J11=H12*$F$1=H12+I12
=A12+110000=B13+E12=C13*$F$1=C13+D13=G12=G13+J12=H13*$F$1=H13+I13
=A13+110000=B14+E13=C14*$F$1=C14+D14=G13=G14+J13=H14*$F$1=H14+I14
=A14+110000=B15+E14=C15*$F$1=C15+D15=G14=G15+J14=H15*$F$1=H15+I15
=A15+110000=B16+E15=C16*$F$1=C16+D16=G15=G16+J15=H16*$F$1=H16+I16
=A16+110000=B17+E16=C17*$F$1=C17+D17=G16=G17+J16=H17*$F$1=H17+I17
=A17+110000=B18+E17=C18*$F$1=C18+D18=G17=G18+J17=H18*$F$1=H18+I18
=A18+110000=B19+E18=C19*$F$1=C19+D19=G18=G19+J18=H19*$F$1=H19+I19
=A19+110000=B20+E19=C20*$F$1=C20+D20=G19=G20+J19=H20*$F$1=H20+I20
=A20+110000=B21+E20=C21*$F$1=C21+D21=G20=G21+J20=H21*$F$1=H21+I21
=A21+110000=B22+E21=C22*$F$1=C22+D22=G21=G22+J21=H22*$F$1=H22+I22
=A22+110000=B23+E22=C23*$F$1=C23+D23=G22=G23+J22=H23*$F$1=H23+I23
=A23+110000=B24+E23=C24*$F$1=C24+D24=G23=G24+J23=H24*$F$1=H24+I24
=A24+110000=B25+E24=C25*$F$1=C25+D25=G24=G25+J24=H25*$F$1=H25+I25
=A25+110000=B26+E25=C26*$F$1=C26+D26=G25=G26+J25=H26*$F$1=H26+I26
=A26+110000=B27+E26=C27*$F$1=C27+D27=G26=G27+J26=H27*$F$1=H27+I27
=A27+110000=B28+E27=C28*$F$1=C28+D28=G27=G28+J27=H28*$F$1=H28+I28
=A28+110000=B29+E28=C29*$F$1=C29+D29=G28=G29+J28=H29*$F$1=H29+I29
=A29+110000=B30+E29=C30*$F$1=C30+D30=G29=G30+J29=H30*$F$1=H30+I30

Poorman

  • Bristles
  • ***
  • Posts: 260
  • Age: 46
  • Location: Orange County, CA
Re: Getting the most out of social security benefits
« Reply #3 on: June 20, 2014, 02:59:16 PM »
The main issue I see is the BOY and Int should be $0 for the first year of benefits (whether you are using 62 or 70).  Since it takes a full year to get to $10,000 that should be the BOY balance for age 63 and the interest accrued should be credited starting that year.  (It's not perfect to calculate the interest yearly, but in real life most of it would be accrued after the 63rd birthday for that first $10,000.)

So I agree with most of your calculations except that they should be pushed one year back:

Return - Age
0% - 80
1% - 81
2% - 82
3% - 83
4% - 85
5% - 87

Still, your spreadsheet is showing that it's not as much of a no-brainer as I previously thought.  I don't have the spreadsheet I used anymore, so I can't see where the differences might have been, but I must have made an error.

Realistically, most people have a hard time calculating their own mortality, so I wouldn't advise people take benefits based on a "guess".  The question is if you live until age 62, what are the odds that you will continue to live until a given age to break even?  Assuming I could earn 2% consistently, what are the odds that I would live past 82 to see the benefit of delaying my payments?

According to the SSA actuarial table from 2009, men have a 52% chance of living until age 82, if they've already made it to age 62.  So statistically, roughly only half of men would live long enough to break even on SS payments by delaying until 70 if they were invested conservatively at 2% per year.  For women, the odds are better with 64% living long enough to break even.  Still that's only 2/3 even for women.  I got to these percentages by dividing the number of people out of 100,000 still alive at age 82 into the number out of 100,000 still alive at age 62.  For men, it was 43,215 out of 100,000 still alive at age 82 and 83,724 at age 62.  (43,215 / 83,724 = 52% still alive).  For women, it was 57,256 out of 100,000 still alive at 82 and 90,099 at age 62.  (57,256 / 90,099 = 64%).

http://www.ssa.gov/oact/STATS/table4c6.html

So I guess my final piece of constructive criticism would be to adjust the spreadsheet to discount for the odds that somebody could live long enough to break even at each given interest rate.

Good conversation.


beltim

  • Magnum Stache
  • ******
  • Posts: 2957
Re: Getting the most out of social security benefits
« Reply #4 on: June 20, 2014, 03:05:36 PM »
You're using the wrong numbers for before "normal retirement age."

According to ssa.gov
Quote
In the case of early retirement, a benefit is reduced 5/9 of one percent for each month before normal retirement age, up to 36 months. If the number of months exceeds 36, then the benefit is further reduced 5/12 of one percent per month.

For example, if the number of reduction months is 60 (the maximum number for retirement at 62 when normal retirement age is 67), then the benefit is reduced by 30 percent. This maximum reduction is calculated as 36 months times 5/9 of 1 percent plus 24 months times 5/12 of 1 percent.

So instead of 8% for the 3 years prior to normal retirement age (65-67 depending on when you were born), it should be 6.67% per year, and it should be 5% per year for the two years before that.

Edit:  Also, you compound the 8%, which is incorrect.  It's a simple interest, not compound.

beltim

  • Magnum Stache
  • ******
  • Posts: 2957
Re: Getting the most out of social security benefits
« Reply #5 on: June 20, 2014, 03:08:03 PM »
And, with those changes, the crossover point for a 2% yield is 89-90.  For a 4% yield, the crossover point is 99-100.

mxt0133

  • Handlebar Stache
  • *****
  • Posts: 1547
  • Location: San Francisco
Re: Getting the most out of social security benefits
« Reply #6 on: June 20, 2014, 03:22:32 PM »
And, with those changes, the crossover point for a 2% yield is 89-90.  For a 4% yield, the crossover point is 99-100.

Could you please clarify.  So based on your adjustments to the calculations then if you get 2% yield on the money you start collecting early then it is best to start collecting at 62 if you stop collecting (die) by the age of 89-90.

And if you expect to get a 4% yield on the money you start collecting early then you should start collecting at 62 if you stop collecting (die) by the age of 99-100.

Is that correct?

beltim

  • Magnum Stache
  • ******
  • Posts: 2957
Re: Getting the most out of social security benefits
« Reply #7 on: June 20, 2014, 03:46:56 PM »
And, with those changes, the crossover point for a 2% yield is 89-90.  For a 4% yield, the crossover point is 99-100.

Could you please clarify.  So based on your adjustments to the calculations then if you get 2% yield on the money you start collecting early then it is best to start collecting at 62 if you stop collecting (die) by the age of 89-90.

And if you expect to get a 4% yield on the money you start collecting early then you should start collecting at 62 if you stop collecting (die) by the age of 99-100.

Is that correct?

Yes, that's what I'm saying.

Poorman

  • Bristles
  • ***
  • Posts: 260
  • Age: 46
  • Location: Orange County, CA
Re: Getting the most out of social security benefits
« Reply #8 on: June 20, 2014, 03:54:43 PM »
And, with those changes, the crossover point for a 2% yield is 89-90.  For a 4% yield, the crossover point is 99-100.

Ok, so I wasn't tripping.  Thanks.

MDM

  • Senior Mustachian
  • ********
  • Posts: 11477
Re: Getting the most out of social security benefits
« Reply #9 on: June 20, 2014, 05:50:03 PM »
The main issue I see is the BOY and Int should be $0 for the first year of benefits (whether you are using 62 or 70).  Since it takes a full year to get to $10,000 that should be the BOY balance for age 63 and the interest accrued should be credited starting that year.  (It's not perfect to calculate the interest yearly, but in real life most of it would be accrued after the 63rd birthday for that first $10,000.)
And, with those changes, the crossover point for a 2% yield is 89-90.  For a 4% yield, the crossover point is 99-100.
Poorman & beltim, thanks for the double checks.

Rather than program the ssa.gov calculations for age 62 vs. 70, the calculations below use actual results from the ssa.gov website for one set of earnings and birth date: $1425/mo at age 62 and $2530/mo at age 70.  Also, rather than do a quick 'n' dirty interest calculation the Excel FV function is used.  Results are in between what the two of you posted, but closer to Poorman's.  beltim, how do your calculations differ?

Code: [Select]
Return Age
0% 80
1% 81
2% 82
3% 84
4% 86
5% 90

AgeMonthly62BoY1EOY10.06Monthly70BoY2EOY2
6214250=FV($E$1/12,12,-B2,-C2,1)
=A2+1=B2=D2=FV($E$1/12,12,-B3,-C3,1)
=A3+1=B3=D3=FV($E$1/12,12,-B4,-C4,1)
=A4+1=B4=D4=FV($E$1/12,12,-B5,-C5,1)
=A5+1=B5=D5=FV($E$1/12,12,-B6,-C6,1)
=A6+1=B6=D6=FV($E$1/12,12,-B7,-C7,1)
=A7+1=B7=D7=FV($E$1/12,12,-B8,-C8,1)
=A8+1=B8=D8=FV($E$1/12,12,-B9,-C9,1)
=A9+1=B9=D9=FV($E$1/12,12,-B10,-C10,1)25300=FV($E$1/12,12,-F10,-G10,1)
=A10+1=B10=D10=FV($E$1/12,12,-B11,-C11,1)=F10=H10=FV($E$1/12,12,-F11,-G11,1)
=A11+1=B11=D11=FV($E$1/12,12,-B12,-C12,1)=F11=H11=FV($E$1/12,12,-F12,-G12,1)
=A12+1=B12=D12=FV($E$1/12,12,-B13,-C13,1)=F12=H12=FV($E$1/12,12,-F13,-G13,1)
=A13+1=B13=D13=FV($E$1/12,12,-B14,-C14,1)=F13=H13=FV($E$1/12,12,-F14,-G14,1)
=A14+1=B14=D14=FV($E$1/12,12,-B15,-C15,1)=F14=H14=FV($E$1/12,12,-F15,-G15,1)
=A15+1=B15=D15=FV($E$1/12,12,-B16,-C16,1)=F15=H15=FV($E$1/12,12,-F16,-G16,1)
=A16+1=B16=D16=FV($E$1/12,12,-B17,-C17,1)=F16=H16=FV($E$1/12,12,-F17,-G17,1)
=A17+1=B17=D17=FV($E$1/12,12,-B18,-C18,1)=F17=H17=FV($E$1/12,12,-F18,-G18,1)
=A18+1=B18=D18=FV($E$1/12,12,-B19,-C19,1)=F18=H18=FV($E$1/12,12,-F19,-G19,1)
=A19+1=B19=D19=FV($E$1/12,12,-B20,-C20,1)=F19=H19=FV($E$1/12,12,-F20,-G20,1)
=A20+1=B20=D20=FV($E$1/12,12,-B21,-C21,1)=F20=H20=FV($E$1/12,12,-F21,-G21,1)
=A21+1=B21=D21=FV($E$1/12,12,-B22,-C22,1)=F21=H21=FV($E$1/12,12,-F22,-G22,1)
=A22+1=B22=D22=FV($E$1/12,12,-B23,-C23,1)=F22=H22=FV($E$1/12,12,-F23,-G23,1)
=A23+1=B23=D23=FV($E$1/12,12,-B24,-C24,1)=F23=H23=FV($E$1/12,12,-F24,-G24,1)
=A24+1=B24=D24=FV($E$1/12,12,-B25,-C25,1)=F24=H24=FV($E$1/12,12,-F25,-G25,1)
=A25+1=B25=D25=FV($E$1/12,12,-B26,-C26,1)=F25=H25=FV($E$1/12,12,-F26,-G26,1)
=A26+1=B26=D26=FV($E$1/12,12,-B27,-C27,1)=F26=H26=FV($E$1/12,12,-F27,-G27,1)
=A27+1=B27=D27=FV($E$1/12,12,-B28,-C28,1)=F27=H27=FV($E$1/12,12,-F28,-G28,1)
=A28+1=B28=D28=FV($E$1/12,12,-B29,-C29,1)=F28=H28=FV($E$1/12,12,-F29,-G29,1)
=A29+1=B29=D29=FV($E$1/12,12,-B30,-C30,1)=F29=H29=FV($E$1/12,12,-F30,-G30,1)
=A30+1=B30=D30=FV($E$1/12,12,-B31,-C31,1)=F30=H30=FV($E$1/12,12,-F31,-G31,1)
=A31+1=B31=D31=FV($E$1/12,12,-B32,-C32,1)=F31=H31=FV($E$1/12,12,-F32,-G32,1)
=A32+1=B32=D32=FV($E$1/12,12,-B33,-C33,1)=F32=H32=FV($E$1/12,12,-F33,-G33,1)
=A33+1=B33=D33=FV($E$1/12,12,-B34,-C34,1)=F33=H33=FV($E$1/12,12,-F34,-G34,1)
=A34+1=B34=D34=FV($E$1/12,12,-B35,-C35,1)=F34=H34=FV($E$1/12,12,-F35,-G35,1)
=A35+1=B35=D35=FV($E$1/12,12,-B36,-C36,1)=F35=H35=FV($E$1/12,12,-F36,-G36,1)

beltim

  • Magnum Stache
  • ******
  • Posts: 2957
Re: Getting the most out of social security benefits
« Reply #10 on: June 20, 2014, 05:55:27 PM »
The main issue I see is the BOY and Int should be $0 for the first year of benefits (whether you are using 62 or 70).  Since it takes a full year to get to $10,000 that should be the BOY balance for age 63 and the interest accrued should be credited starting that year.  (It's not perfect to calculate the interest yearly, but in real life most of it would be accrued after the 63rd birthday for that first $10,000.)
And, with those changes, the crossover point for a 2% yield is 89-90.  For a 4% yield, the crossover point is 99-100.
Poorman & beltim, thanks for the double checks.

Rather than program the ssa.gov calculations for age 62 vs. 70, the calculations below use actual results from the ssa.gov website for one set of earnings and birth date: $1425/mo at age 62 and $2530/mo at age 70.  Also, rather than do a quick 'n' dirty interest calculation the Excel FV function is used.  Results are in between what the two of you posted, but closer to Poorman's.  beltim, how do your calculations differ?

You're still using bad data.  The "actual results" that you used assumed that the person was working between the ages of 62 and 70.

Edit: wait, that might not be right.  I need to recheck it
« Last Edit: June 20, 2014, 05:57:45 PM by beltim »

beltim

  • Magnum Stache
  • ******
  • Posts: 2957
Re: Getting the most out of social security benefits
« Reply #11 on: June 20, 2014, 07:15:13 PM »
I think your updated spreadsheet is correct.  My calculations were wrong because I used the age 62 benefit as the base, rather than the age 67 benefit. 

It is interesting to note that taking Social Security early is always beneficial if you can earn a 7.4% return or greater, a number that is close to the long-term return of the market after inflation.

MDM

  • Senior Mustachian
  • ********
  • Posts: 11477
Re: Getting the most out of social security benefits
« Reply #12 on: June 20, 2014, 07:38:58 PM »
...taking Social Security early is always beneficial if you can earn a 7.4% return or greater, a number that is close to the long-term return of the market after inflation.

Interesting point.  Would the results depend heavily on the returns in the 8 years between 62 and 70?  E.g., has anyone used cFIREsim to compare Soc Sec options?  Take 1) $1425/mo for 30 years, or 2) start 8 years later and get 22 years of $2530/mo (or any valid set of options) - then compare what fraction of start dates have better results for #1 vs #2?  Not familiar enough with cFIREsim to know if that is a feasible question.

TomTX

  • Walrus Stache
  • *******
  • Posts: 5345
  • Location: Texas
Re: Getting the most out of social security benefits
« Reply #13 on: June 22, 2014, 08:14:27 AM »
Couples can start to play optimization games if they wait until "full retirement" instead of taking early.

For example, I can take my spousal benefit from 67->70 and still get the fully inflated personal benefit @ 70. Her benefit won't be huge, so at 70 (after collecting on her own benefit from 67->70) she'll switch to getting a spousal benefit based on my record. Presuming I die first, my wife would have a fully inflated widow's benefit replace the spousal benefit when I die. *

Anyway, by following this approach, we get paid cash for 3 years, PLUS future benefits keep going up 8% a year.

If you try this before "full retirement" SSA will "deem" that you are also filing on your own benefit, and you won't see the growth - "excess benefit hell."

There are a LOT of complexities from optimizing social security. Even SSA staff don't know the right answer a lot of the time. Don't get burned.

*There are a couple more reasons to do it this way in my situation. My pension dies when I do, so as contingency planning this will leave my (presumed) widow with a bigger income. Also, my pension has no COLA, so it's shrinking all the time.

Workinghard

  • Pencil Stache
  • ****
  • Posts: 636
Re: Getting the most out of social security benefits
« Reply #14 on: June 22, 2014, 05:46:46 PM »
I don't think that will work with us because my husband is six years older than me. His benefit at 66 is $2145 and at 70 will be about $2800.  Mine at 66 1/2 is $1700 and at 70 $2386. Part of me would like to draw so we don't have to pull from our retirement accounts, but I know financially it would make better sense to wait. He will retire at 65 when is eligible for Medicare. I will work for another year or two to delay withdrawing money and for the transition.

Couples can start to play optimization games if they wait until "full retirement" instead of taking early.

For example, I can take my spousal benefit from 67->70 and still get the fully inflated personal benefit @ 70. Her benefit won't be huge, so at 70 (after collecting on her own benefit from 67->70) she'll switch to getting a spousal benefit based on my record. Presuming I die first, my wife would have a fully inflated widow's benefit replace the spousal benefit when I die. *

Anyway, by following this approach, we get paid cash for 3 years, PLUS future benefits keep going up 8% a year.

If you try this before "full retirement" SSA will "deem" that you are also filing on your own benefit, and you won't see the growth - "excess benefit hell."

There are a LOT of complexities from optimizing social security. Even SSA staff don't know the right answer a lot of the time. Don't get burned.

*There are a couple more reasons to do it this way in my situation. My pension dies when I do, so as contingency planning this will leave my (presumed) widow with a bigger income. Also, my pension has no COLA, so it's shrinking all the time.