Theory Behind taking Social Security Early?

[pb4uski]

Based on most likely mortality, they will break even. The government hasn’t designed SS to give you a lot of extra money ....... they know what they’re doing.

I see you keep claiming that in numerous posts... but you are wrong.

The discounts and premiums for taking SS early or late have not changed since the 1980s but mortality has improved dramatically since the 1980s with medical advances... IOW, today's 62 year old is likely to live longer than a 62 year old back when the 25-30% discount was developed. So while SS benefit payments may have been actuarially neutral when the factors were developed, they probably are not today.

For example, according to the 2014 Actuarial Life Table, a 62 year old male or female have remaining life expectancies of 20.00 and 22.81 years, respectively. The life expectancies according to the 2004 table were 18.85 and 21.89 years, respectively. And those changes were only for 10 years but it has been 30 years or more since the factors were changed.

Another thing that you are not considering it that any actuarial neutrality is for the population... people who frequent this board tend to be more wealthy and longevity is better for wealthy people so people who frequent this forum will likely collect longer than the average Joe or Jill.

And then there is joint mortality... even if the discounts and premium were perfect for a single person, for married couples joint mortality.... the likelihood that one or the other will be alive to collect benefits... extends the collection period by several years but the discount and premium are not different between single and joint lives.

 

[pb4uski]

Am I correct in thinking that the proper discount rate to enter on the opensocialsecurity.com (I think that's the URL site mentioned up thread) advanced options is my expected rate of return on the investments that I would sell or not sell depending on whether I took SS earlier or later?

IMO, yes, as long as you enter a real rate of return (after inflation) rather than a nominal rate of return.

For other planning that I do I use a 6.0% nominal earnings rate and a 2.7% rate of inflation... so for that tool I used 3.3%.

I undertand the TIPs rate thing... I just don't agree with it particularly in my case since I plan to keep a 60/40 AA from 62 to 70 and beyond.

 

[daylatedollarshort]

You hit your head too Hard! ........ You don't need to know the Year of Death, just like you don't when managing your portfolio. A 'Real Mathematician' would know what to solve for.

You do if you are trying to maximize total lifetime benefits, per the chart imoldernu posted earlier in this thread - http://www.early-retirement.org/for...cial-security-early-93511-24.html#post2109375.

 

[Fedup]

You do if you are trying to maximize total lifetime benefits, per the chart imoldernu posted earlier in this thread - http://www.early-retirement.org/for...cial-security-early-93511-24.html#post2109375.

You are not reading at that chart do you. I just looked, it shows claiming at 70 have a highest value, something like $290k. Of course, if you croak at 80, then it’s a different story, but it’s not 62, it’s highest at 67.

I don’t care if people take it whatever they want, but let’s not spread misinformation here.

 

[pb4uski]

You do if you are trying to maximize total lifetime benefits, per the chart imoldernu posted earlier in this thread - http://www.early-retirement.org/for...cial-security-early-93511-24.html#post2109375.

Glad that you posted that:

According to SSA's 2014 mortality table, a 70 yo male will live on average another 14.3 years (to 84.3) and female another 16.5 years (to 86.5)... and those expectations are on the low side due to 3 years of additional mortality improvement plus the impact of wealth.... so it is more likely than not that the average 70 year old will live long enough to collect more.

https://www.ssa.gov/oact/STATS/table4c6_2014.html

All of that said, my playing around with opensocialsecurity.com using 2017 mortality and adjusting for the time value of money suggest that the expected present values are all within 2-3% of each other so the decision doesn't matter all that much.

 

[pb4uski]

You are not reading at that chart do you. I just looked, it shows claiming at 70 have a highest value, something like $290k. Of course, if you croak at 80, then it’s a different story, but it’s not 62, it’s highest at 67.

I don’t care if people take it whatever they want, but let’s not spread misinformation here.

I don't get it.... what is the misinformation?

 

[daylatedollarshort]

Glad that you posted that:

According to SSA's 2014 mortality table, a 70 yo male will live on average another 14.3 years (to 84.3) and female another 16.5 years (to 86.5)... and those expectations are on the low side due to 3 years of additional mortality improvement plus the impact of wealth.... so it is more likely than not that the average 70 year old will live long enough to collect more.

https://www.ssa.gov/oact/STATS/table4c6_2014.html

All of that said, my playing around with opensocialsecurity.com using 2017 mortality and adjusting for the time value of money suggest that the expected present values are all within 2-3% of each other so the decision doesn't matter all that much.

More likely than not is not the same as saying you don't need to know X (death date) to maximize lifetime benefits, which is the post I was replying to. Still the factor of SS benefits past 70 to not have an iron clad guarantee of being the same or more to consider. If you feel future SS benefits won't be changed and will live longer than the break even point then delaying is the right decision for you.

 

[Fedup]

I don't get it.... what is the misinformation?

Maximize total lifetime benefits as per her post.

 

[pb4uski]

Maximize total lifetime benefits as per her post.

I thought that all that she was saying was that if you know when you will die that you can use the table to maximize your lifetime benefits... I don't see much wrong with that.

 

[Fedup]

I thought that all that she was saying was that if you know when you will die that you can use the table to maximize your lifetime benefits... I don't see much wrong with that.

Ok, I may have misread it. But since she keeps saying she is taking SS at 62, implying that taking it at 62 she is maximizing her lifetime benefits.

 

[SecondCor521]

That depends.

If I figured that I didn't need SS for my own expenses, but I were trying to maximize the estate I leave for children or charity, I would use the expected return on the type of assets I'd buy with my SS benefit.

If I'm using SS to meet my own spending, and considering SS as longevity insurance, I'd use a pretty conservative number. The default 20 year TIPS number seems okay.

(This is like asking what investment return you use in determining a safe withdrawal rate - the most likely return on your assets, or something more conservative.)

Note that either way, I want a "real" return, not a "nominal" return.

Also, since I'm using the advanced options, I'd specify an "assumed age at death" rather than using any of the mortality tables. The math with the mortality tables assumes smaller and smaller payouts each year as I age. That's great for an insurance company pricing an SPIA, not so good for one person.

(And, if I use a mortality table, I definitely would not use the default 2015 SS Period Life table. It is not representative for anyone here who would ask this question.)

IMO, yes, as long as you enter a real rate of return (after inflation) rather than a nominal rate of return.

For other planning that I do I use a 6.0% nominal earnings rate and a 2.7% rate of inflation... so for that tool I used 3.3%.

I undertand the TIPs rate thing... I just don't agree with it particularly in my case since I plan to keep a 60/40 AA from 62 to 70 and beyond.

Thanks, both of you.

I mentioned up thread that if I put in anything above a very low number (1.28% IIRC?) it has me claiming at 62 as the most lucrative strategy. That was using the 2017 preferred mortality table.

My AA at 62 is very likely to be between 90/10 and 100/0, so I'd be comfortable with a real rate of 3-4% (which is obviously >> 1.28%).

If I did take it at 62, I would look at it as taking those SS dollars and spending them and leaving an equivalent number of dollars invested which I otherwise would have had to take out to spend...I guess plus the income taxes due on those withdrawals. So it makes sense to me to use my portfolio real rate of return.

Being single and having a very low WR, I don't really value SS for the longevity or really the extra spending that it might give me.

I still need to balance the RMD tax torpedo issue though, which may sway me back to taking it at 70. My gut thinks that the tax savings of RMD's from 62 to 70 might outweigh the somewhat higher SS benefit NPV. The last I looked at that spreadsheet I was dancing with trying to stay under the bottom of the 32% bracket. I should probably build a spreadsheet for that at some point in the next 12 years.

 

[CardsFan]

You hit your head too Hard! ........ You don't need to know the Year of Death, just like you don't when managing your portfolio. A 'Real Mathematician' would know what to solve for.

So, you are you really talking about theoretical mathematics? Or statistics? What are you trying to maximize? Total payout? Legacy to heirs? Protection against longevity?

I agree, you can play the odds, and that might be the right thing to do, depending on your circumstances. If you are trying to maximize the CALCULATED SWR based on a formula, I agree that for most people, 70 is the answer. BUT, you have already set a parameter based on your gut/emotion.

But if you already have a WR less than 3%, and that will drop to 1-2% when SS kicks in (whenever that might be), then the priorities might change.

 

[daylatedollarshort]

You are not reading at that chart do you. I just looked, it shows claiming at 70 have a highest value, something like $290k. Of course, if you croak at 80, then it’s a different story, but it’s not 62, it’s highest at 67.

I don’t care if people take it whatever they want, but let’s not spread misinformation here.

If you claim at 62 and die at 63 (X per Cardsfan post), you have maximized your life time benefits by claiming at 62. No way to know X in advance for most of us, which was Cardsfan point.

 

[Fedup]

If you claim at 62 and die at 63 (X per Cardsfan post), you have maximized your life time benefits by claiming at 62. No way to know X in advance for most of us, which was Cardsfan point.

It’s true you don’t know X in advance, so why can’t X be 103. The X at 103 scares me more than the X at 63.

My sister has several friends who died before taking SS, so I thought naturally she would take at 62. But no she is waiting until FRA to take it. I’m not swaying her one way or the other.

 

[daylatedollarshort]

Ok, I may have misread it. But since she keeps saying she is taking SS at 62, implying that taking it at 62 she is maximizing her lifetime benefits.

Maximizing my lifetime benefits is not my goal. If I gave you that impression in this thread please let me know the post and I will try to correct it.


We have kids we would like to leave money to if we would die when they are younger and don't have their own savings built up yet, we want to diversify our income streams, are concerned about future asset testing and we don't need longevity insurance, so our priorities might be very different than other posters.

 

[daylatedollarshort]

It’s true you don’t know X in advance, so why can’t X be 103.

It can be. But you can't say with certainty you maximized your SS benefits unless you know in advance how long you will live and what changes, if any, may occur to SS benefits over your collecting lifetime.

 

[Fedup]

Maximizing my lifetime benefits is not my goal. If I gave you that impression in this thread please let me know the post and I will try to correct it.

Post 628, since you post maximizing lifetime benefits in that post.

 

[daylatedollarshort]

Post 628, since you post maximizing lifetime benefits in that post.

I did not say that was my goal in that post. I think most people here get Cardsfan's point about not being able to truly maximize benefits without knowing X, date of death, as this has been discussed before, so I'll leave it at that.

 

[Fedup]

We have kids we would like to leave money to if we would die when they are younger and don't have their own savings built up yet,

Honestly, this is the first time I’ve read this from you. In the past, you went around and around with lots of reasons that never made sense to me. You don’t want to use your nest egg because your kids are young adults, in case you might die young. At least it’s a reason, I can understand.

But wouldn’t life insurance is a better alternative.

 

[rayvt]

But the "break even" question is not a useful one, anyway.

No, the "break even" question is the primary useful one.

You have to compare like to like, and dragging in personal factors like assumed longevity, etc. obscures that. You will die at the same age whether you take SS at 62 or 70.

Here's the break-even question reasoning:
Assuming you actually have the option of forgoing the SS income at 62, you can get along just fine without it.
1) Take SS at 62, put it all in a dedicated account (savings account, investments, T-bills, bonds, whatever.)
2) At 70, figure out what you would be getting if you had waited until 70 to start SS.
3) Compute the difference between this and what you are actually getting (from having filed at 62). This is the shortfall.
4) Now you stop saving up the SS you've been getting since 62 and start spending it.
5) Each month, Withdraw enough from this dedicated account cover the shortfall.
6) How long is it before the dedicated account is exhausted? This totally depends on the return that your dedicated account makes.

Up until this B/E point, your financial income related to SS is the same. Before 70 you spend $0 SS money and after 70 you spend the age 70 benefit. It is only AFTER the dedicated account is depleted that you see any difference in spendable income. The question is how long does that take? How long does that account last?

That's easy to figure out. Plug the numbers into a financial savings&spending calculator. Deposit $X/month into an account that earns Y% on the balance, for ((70-62) *12) months, then withdraw $Z/mo.

If Y% is 0%, it lasts 10.6 years. Age 81
If Y% is 5%, it lasts 20.8 years. Age 91
If Y% is 6%, it lasts 27.6 years. Age 97
If Y% is 7%, it lasts more than 38 years. Still going strong at age 108.

Accounting for inflation adds a little complexity. So call the above "Y% above inflation."
Interesting......at 9% rate and 2% inflation, the side account is still growing at year 38.

FWIW, since 1950 the simple (non-compounded) average inflation was 3.5% and the average 10-YR T-bill rate was 5.7%.

 

[rayvt]

The discounts and premiums for taking SS early or late have not changed since the 1980s but mortality has improved dramatically since the 1980s with medical advances... IOW, today's 62 year old is likely to live longer than a 62 year old back when the 25-30% discount was developed.

It worse than that. Or better, depending on which way you look at it.

What matters in not the overall change in mortality rates, what matters is the change in mortality rates for our particular cohort. That being: people who had above average lifetime incomes and who retired (early!) with far above average liquid net worth. None of us worked in steel mills or dug ditches or installed asbestos.

I'm sure this information is in some table somewhere, but I think that OUR longevity hasn't increased very much. We always got a good deal on the SS discounts/premiums scale.

 

[ERD50]

No, the "break even" question is the primary useful one. ...

I'm only responding because you actually made it easy for me to be concise, which is something that isn't always easy for me.

You still just don't get, or are not listening to the "longevity insurance' viewpoint. We don't buy car ins, or home ins, or an umbrella policy with "break even" as the primary reason or primary evaluation question.

We buy these forms of insurance to protect against things we can't predict. Like, oh, I dunno - how long we are going to live?

-ERD50

 

[Independent]

If Y% is 0%, it lasts 10.6 years. Age 81
If Y% is 5%, it lasts 20.8 years. Age 91
If Y% is 6%, it lasts 27.6 years. Age 97
If Y% is 7%, it lasts more than 38 years. Still going strong at age 108.

Accounting for inflation adds a little complexity. So call the above "Y% above inflation."
Interesting......at 9% rate and 2% inflation, the side account is still growing at year 38.

FWIW, since 1950 the simple (non-compounded) average inflation was 3.5% and the average 10-YR T-bill rate was 5.7%.

Note that today, 20 year TIPS are yielding 0.9%.

Yes, I can make a table of break even years for different investment returns. Mine is more complicated than yours because I'm married an need to think about survivor benefits. But, I'm sure I can do the math.

But, what do I do then? Which investment yield and which age a death should drive my decision?

If I plan to never spend my SS benefit, but just accumulate it as a windfall for children or charity, I probably want to focus on a "most likely" age at death. I also want to invest it 100% in equities, and think about the "most likely" equity return. That probably says I should take it earlier. The start @62 will most likely be ahead on my most likely date of death.

But, that's not how I plan for the income I intend to spend. When I look at withdrawal rates, I don't pick the most likely years to death and a most likely investment return. I pick 30 years, and my withdrawal rate is determined by the worst 5% historic returns.

If I focus on most likely, I'd ask FireCalc for a withdrawal rate that survives for 20 years in 50% of the investment scenarios. It will tell me I can spend 8%, not 4%.

When I'm picking a withdrawal rate, I want to be "conservative" regarding life span and investment returns. Shouldn't I be equally conservative when I'm deciding when to start SS?

 

[Independent]

It worse than that. Or better, depending on which way you look at it.

What matters in not the overall change in mortality rates, what matters is the change in mortality rates for our particular cohort. That being: people who had above average lifetime incomes and who retired (early!) with far above average liquid net worth. None of us worked in steel mills or dug ditches or installed asbestos.

I'm sure this information is in some table somewhere, but I think that OUR longevity hasn't increased very much. We always got a good deal on the SS discounts/premiums scale.

I can't help providing a factoid.

AFAIK, there is no table for people who retired early with substantial assets as compared to other retirees.

Actuaries who value pension plans have long used different mortality rates for "white collar" vs. "blue collar". White collar has lower mortality.

The SSA actuaries have done some work on mortality by lifetime earnings. https://www.ssa.gov/policy/docs/ssb/v73n1/v73n1p1.html
Higher lifetime earnings are associated with lower mortality rates.

If we focus on change in mortality rates over time, I can find this paper https://www.ssa.gov/policy/docs/workingpapers/wp108.html
It says that most of the mortality improvement for the two birth cohorts they studied came from the higher income group.

So, yes, higher income people have probably always had better mortality, but the rate of improvement is also faster for the higher income group. This has political implications when people propose raising the SS normal retirement age.

I don't know that this is really relevant to your discussion with ERD.
Certainly, the discounts for early benefits are very crude and can't be expected to be accurate over variations in personal circumstances, or changes over time in investment returns and mortality.

 

[samclem]

You have to compare like to like, and dragging in personal factors like assumed longevity, etc. obscures that. You will die at the same age whether you take SS at 62 or 70.

If somebody wants to do this break-even forecasting for their own case, they would be well advised to use information that best applies to their own situation. Averages for population mortality will be most applicable if they don't know their gender, their own medical history or condition, risks they have been exposed to.

If Y% is 0%, it lasts 10.6 years. Age 81
If Y% is 5%, it lasts 20.8 years. Age 91
If Y% is 6%, it lasts 27.6 years. Age 97
If Y% is 7%, it lasts more than 38 years. Still going strong at age 108.
Accounting for inflation adds a little complexity. So call the above "Y% above inflation."

Including inflation assumptions is important to this exercise. When we hit "the age" above, when the amount in the "set aside" account goes to zero, what then? If we'd taken SS at age 70, there would be no "what then", because we'd have the exact same >inflation adjusted< check coming in for the rest of our lives. But if we started the SS checks at age 62, then at age 85 inflation has had 23 years to do its work. If the monthly difference between the age 62 SS check and the age 70 SS check was $800, then 22 years of 4% inflation means we'll have to dip into our >other< investments for $1,700 each month just to make up for the missing dough. And that will go up with inflation every year past the "break even" year.

No, the "break even" question is the primary useful one.

We don't buy car ins, or home ins, or an umbrella policy with "break even" as the primary reason or primary evaluation question.

We buy these forms of insurance to protect against things we can't predict.

Agreed.

When I'm picking a withdrawal rate, I want to be "conservative" regarding life span and investment returns. Shouldn't I be equally conservative when I'm deciding when to start SS?

I also think it is useful for people to try hard to imagine the decisions they may have at advanced ages. When a person is 85+, to strictly use universal longevity tables as a guide for practical withdrawal rates can significantly increase the chances of running out of money, or of leaving a lot of money on the table after death, unless you die "right on schedule." This is a place where annuitization makes sense, it lets us offload the risk to someone else. They have the advantage of the Law of Large Numbers," we don't. Taking SS later is the cheapest way to effectively annuitize.