Wait for Social Security - breakeven returns

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
CurlyDave
Posts: 1328
Joined: Thu Jul 28, 2016 11:37 am

Re: Wait for Social Security - breakeven returns

Post by CurlyDave » Sun Nov 10, 2019 11:05 pm

ObliviousInvestor wrote:
Sun Nov 10, 2019 10:32 pm


Imagine a person who:
1) Has a portfolio that is 60% stocks, 40% TIPS, and
2) Is currently planning on taking Social Security at their FRA.

And let's also grant without argument, for the sake of making this point, that stocks have an expected real return of 7%...
This will not get you the real combined average return of 7%.

To be very clear, what I meant by a real combined average return of 7% is:

(% stocks/100) x stock return + (% bonds/100) x bond return = 7% . If the only assets in the portfolio are stocks and bonds then
% bonds = 100 - % stocks.

While this might seen like a very high return, the 10 year trailing return for VOO is 13.65%.

User avatar
ObliviousInvestor
Posts: 3587
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: Wait for Social Security - breakeven returns

Post by ObliviousInvestor » Sun Nov 10, 2019 11:15 pm

CurlyDave wrote:
Sun Nov 10, 2019 11:05 pm
ObliviousInvestor wrote:
Sun Nov 10, 2019 10:32 pm


Imagine a person who:
1) Has a portfolio that is 60% stocks, 40% TIPS, and
2) Is currently planning on taking Social Security at their FRA.

And let's also grant without argument, for the sake of making this point, that stocks have an expected real return of 7%...
This will not get you the real combined average return of 7%.

To be very clear, what I meant by a real combined average return of 7% is:

(% stocks/100) x stock return + (% bonds/100) x bond return = 7% . If the only assets in the portfolio are stocks and bonds then
% bonds = 100 - % stocks.

While this might seen like a very high return, the 10 year trailing return for VOO is 13.65%.
OK, you pick the hypothetical allocation then.

If the bond allocation is greater than 0%, the general premise of what I wrote is still applicable (though they may run out of bonds prior to FRA, and then find it advantageous to file at that point). If additional risk is desired, and there are bonds in the portfolio, using those bonds to fund the additional risk will usually be preferable to filing for Social Security early in order to fund the additional risk.
Mike Piper, author/blogger

CurlyDave
Posts: 1328
Joined: Thu Jul 28, 2016 11:37 am

Re: Wait for Social Security - breakeven returns

Post by CurlyDave » Mon Nov 11, 2019 12:19 am

ObliviousInvestor wrote:
Sun Nov 10, 2019 11:15 pm

OK, you pick the hypothetical allocation then.
1. I was not looking for a global optimum, I was only pointing out the condition necessary for an early claim to be more beneficial than a delayed claim.

2. If you want a hypothetical, how about this one. The dates are not cherry picked -- they are chosen because I have data close at hand for them.

Consider a retiree who can take age 62 SS starting with a first payment on 11-1-2009.

The 10 year trailing return for VOO is 13.65%, the 10 year trailing return for BND is 3.65%. Consider a portfolio consisting on only those two components.

If I let fs=decimal fraction of stocks, and i = inflation rate in %, I can write:

13.65 x fs + 3.65 x (1-fs) = 7 + i

for the minimum condition for an early claim to be more beneficial. If we take i = 2% (I think the actual is closer to 1.84%, but 2% biases the result to higher stock allocations).

13.65 x fs + 3.65 x (1-fs) = 9, or

10 fs = 5.35 and fs = .535 or an AA of 53.5 % VOO and 46.5% BND.

So a portfolio with 53.5% VOO would make an early claim more beneficial if we can project the last decade's returns over the remaining life of the retiree. And if the math behind the third figure in the OP's linked article is correct. Obviously, higher stock allocations produce greater benefits for an early claim.

Notes:

1. This is just a mathematical exercise. I don't think we will see the same VOO returns over the next two decades as we saw over the past one.

2. In this case, and I think many others, the true maximum comes when fs = 1 , i.e. 100% stocks.

3. I do think that an assumption of Stock Return > 7 + i is reasonable going forward, but this is controversial.

User avatar
ObliviousInvestor
Posts: 3587
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: Wait for Social Security - breakeven returns

Post by ObliviousInvestor » Mon Nov 11, 2019 1:31 am

CurlyDave wrote:
Mon Nov 11, 2019 12:19 am
So a portfolio with 53.5% VOO would make an early claim more beneficial if we can project the last decade's returns over the remaining life of the retiree.
No, it would not, even if we project those returns over their remaining life.

We are not concerned with the weighted-average return of the portfolio. We are concerned with the portion of the portfolio that would be used as a replacement for early-claimed Social Security benefits. If the portfolio includes bonds, we can use those.

The return from delaying Social Security from 62 to 70, if the person's mortality turns out to be "average" in the sense that it matches the SSA's 2016 Period Life Table is about 1.8% real for an unmarried man or about 3% real for an unmarried woman. And that somewhat understates the reality, because period life tables somewhat understate life expectancy.

And this is without accounting for the possibility of segregating the bond allocation, using Social Security (delayed benefits) as a replacement for the government portion of BND and using a corporate bond fund for the corporate portion of BND. (If we do that, then the return from the bond portion that we are giving up -- the government portion -- is even lower.)
Mike Piper, author/blogger

smitcat
Posts: 4326
Joined: Mon Nov 07, 2016 10:51 am

Re: Wait for Social Security - breakeven returns

Post by smitcat » Thu Nov 14, 2019 8:34 am

CurlyDave wrote:
Sun Nov 10, 2019 9:19 pm
smitcat wrote:
Sun Nov 10, 2019 7:02 pm

Additionally - I would never model the difference by comparing the early SS with stock returns unless maybe my target AA was 100% stocks.
That would certainly be comparing apples to oranges - you could also just move more money to stocks and still delay SS with a superior result..
My actual AA was 100% stocks, so for me that is the correct comparison.

you could also just move more money to stocks and still delay SS with a superior result..

Where would that money have come from? I don't think that "superior result" is available.

"Where would that money have come from? I don't think that "superior result" is available."
Certainly there is no where it can come from if you are 100% in stocks already.

smitcat
Posts: 4326
Joined: Mon Nov 07, 2016 10:51 am

Re: Wait for Social Security - breakeven returns

Post by smitcat » Thu Nov 14, 2019 9:12 am

CurlyDave wrote:
Sun Nov 10, 2019 10:05 pm
grabiner wrote:
Fri Nov 08, 2019 10:44 pm
ObliviousInvestor wrote:
Thu Nov 07, 2019 3:55 pm
Sure. That's somewhat uncommon though, because it implies that you're already 100% stocks in the portfolio. (This is what Grabiner was getting at above.)
One in 10 participants have taken an extreme position, holding either 100% in equities (6% of participants) or no equities (3% of participants).
https://institutional.vanguard.com/iam/pdf/HAS2019.pdf
And even that overstates the number of investors who are 100% stock, because people have multiple accounts; some investors have stock in their 401(k)s and bonds in their IRAs or taxable accounts. (I am the other way around; my employer plan holds all my bonds, while my HSA, Roth IRA, and taxable account are 100% stock.)
I did not realize that my 100% stock allocation put my in such an elite crowd. :D

I do want to point out that for the purposes of this discussion, a 100% stock allocation is not necessary to make early claiming beneficial. All that is necessary is that the SS payments are accumulated in a portfolio (or sub-portfolio) where the real combined average return is greater than 7%.
"All that is necessary is that the SS payments are accumulated in a portfolio (or sub-portfolio) where the real combined average return is greater than 7%"
- and that early SS does not block Roth conversions that would tip the analysis
- and that it takes into account a surviving spouse
- and that all the above are measured in 'spendable' after tax dollars.

3-20Characters
Posts: 683
Joined: Tue Jun 19, 2018 2:20 pm

Re: Wait for Social Security - breakeven returns

Post by 3-20Characters » Thu Nov 14, 2019 11:15 am

smitcat wrote:
Thu Nov 14, 2019 9:12 am
CurlyDave wrote:
Sun Nov 10, 2019 10:05 pm
grabiner wrote:
Fri Nov 08, 2019 10:44 pm
ObliviousInvestor wrote:
Thu Nov 07, 2019 3:55 pm
Sure. That's somewhat uncommon though, because it implies that you're already 100% stocks in the portfolio. (This is what Grabiner was getting at above.)
One in 10 participants have taken an extreme position, holding either 100% in equities (6% of participants) or no equities (3% of participants).
https://institutional.vanguard.com/iam/pdf/HAS2019.pdf
And even that overstates the number of investors who are 100% stock, because people have multiple accounts; some investors have stock in their 401(k)s and bonds in their IRAs or taxable accounts. (I am the other way around; my employer plan holds all my bonds, while my HSA, Roth IRA, and taxable account are 100% stock.)
I did not realize that my 100% stock allocation put my in such an elite crowd. :D

I do want to point out that for the purposes of this discussion, a 100% stock allocation is not necessary to make early claiming beneficial. All that is necessary is that the SS payments are accumulated in a portfolio (or sub-portfolio) where the real combined average return is greater than 7%.
"All that is necessary is that the SS payments are accumulated in a portfolio (or sub-portfolio) where the real combined average return is greater than 7%"
- and that early SS does not block Roth conversions that would tip the analysis
- and that it takes into account a surviving spouse
- and that all the above are measured in 'spendable' after tax dollars.
⬆️ Succinct!

User avatar
#Cruncher
Posts: 2858
Joined: Fri May 14, 2010 2:33 am
Location: New York City
Contact:

Re: Wait for Social Security - breakeven returns

Post by #Cruncher » Fri Nov 15, 2019 11:24 am

ObliviousInvestor wrote:
Sun Nov 10, 2019 10:32 pm
Imagine a person who:
1) Has a portfolio that is 60% stocks, 40% TIPS, and
2) Is currently planning on taking Social Security at their FRA.
...
Let's say the person's Social Security at FRA of 67 would be $1,000/month. So at 62 it would be $700/month.
So they can file at 62 and put that $700/month into stocks.
Or they can wait until FRA, and starting at 62 start moving $700/month from TIPS into stocks.

When you compare those two, you will find that the second strategy comes out ahead, for a single person with an average life expectancy. (underlines added)
Nice post, OI. I've prepared an illustration. It assumes a $100,000 TIPS investment at age 62 and shows how this grows or shrinks under each strategy. Strategy number 1 is shown in column B where the balance simply increases by the assumed 0.5% real yield. Strategy number 2 is shown in column E where $8,400 is taken each year to buy stocks and -- beginning at age 67 -- $12,000 of SS benefits is added. As shown in column F, by age 80 one will come out ahead with the 2nd strategy. Takeaways:
  • The breakeven is independent of the assumed return on stocks. Whether it is 0%, 7%, or 70% the 2nd strategy will come out ahead by age 80. This is because the stock investment is the same under both strategies.
  • The breakeven is also independent of the assumed initial size of the TIPS portfolio -- as long as it is large enough to fund the stock purchases until delayed SS benefits begin.
  • The breakeven is the same as if one simply compares the discounted value of the two SS cash flows ($1,000 / mo starting at age 67 versus $700 / mo starting at age 62), using the assumed TIPS yield as the discount rate. The formula in cell C5 computes this to be age 79.2 which corresponds to the point where column F turns positive.

Code: Select all

Row  ColA     Col B      Col C      Col D      Col E      Col F
  1  TIPS yield          0.50%
  2  SS start 62 $         700
  3  Delay to age           67
  4  SS start 67 $       1,000
  5  Breakeven age        79.2   =IF(C1=0,C2*(C3-A8)/(C4-C2),LN((C2-C4)/(C2*(1+C1)^(C3-A8)-C4))/LN(1+C1))+C3
           Start 62   ---------- Start 67 ----------   Start 67
  7         Balance   To Stock    From SS    Balance     vs 62

Code: Select all

  8   62    100,000                          100,000            
  9   63    100,500     (8,400)               92,100     (8,400)
 10   64    101,003     (8,400)               84,161    (16,842)
 11   65    101,508     (8,400)               76,181    (25,326)
 12   66    102,015     (8,400)               68,162    (33,853)
 13   67    102,525     (8,400)               60,103    (42,422)
 14   68    103,038     (8,400)    12,000     64,004    (39,034)
 15   69    103,553     (8,400)    12,000     67,924    (35,629)
 16   70    104,071     (8,400)    12,000     71,863    (32,208)
 17   71    104,591     (8,400)    12,000     75,822    (28,769)
 18   72    105,114     (8,400)    12,000     79,802    (25,312)
 19   73    105,640     (8,400)    12,000     83,801    (21,839)
 20   74    106,168     (8,400)    12,000     87,820    (18,348)
 21   75    106,699     (8,400)    12,000     91,859    (14,840)
 22   76    107,232     (8,400)    12,000     95,918    (11,314)
 23   77    107,768     (8,400)    12,000     99,998     (7,771)
 24   78    108,307     (8,400)    12,000    104,098     (4,210)
 25   79    108,849     (8,400)    12,000    108,218       (631)
 26   80    109,393     (8,400)    12,000    112,359      2,966  <===
 27   81    109,940     (8,400)    12,000    116,521      6,581 
 28   82    110,490     (8,400)    12,000    120,704     10,214 
 29   83    111,042     (8,400)    12,000    124,907     13,865 
 30   84    111,597     (8,400)    12,000    129,132     17,534 
 31   85    112,155     (8,400)    12,000    133,377     21,222 
 32   86    112,716     (8,400)    12,000    137,644     24,928 
 33   87    113,280     (8,400)    12,000    141,932     28,653 
 34   88    113,846     (8,400)    12,000    146,242     32,396 
 35   89    114,415     (8,400)    12,000    150,573     36,158 
 36   90    114,987     (8,400)    12,000    154,926     39,939
ObliviousInvestor in same post wrote:Point being, if a single person:
a) wants more risk ...
If you simply consider stocks and TIPS, strategy 2 is indeed more risky since it has a higher proportion of stocks. But this need not be the case if the capitalized value of future SS benefits is included in one's fixed income allocation. For more see my post, Re: Effect of Social Security when Retiring before Benefits Begin.

Post Reply