Misconceptions About Delaying Social Security

[veremchuka]

Quick read. YMMV.

Misconceptions About Delaying Social Security

 

[sengsational]

Bottom line is that comparing delaying social security to a bond is not correct, but comparing the delay to a lifetime annuity is. Yep, one ends at a known point and the other not.

 

[braumeister]

I give him a good bit of credit. He's only 29 years old and has quite a number of books he's selling, besides this one.

 

[tmm99]

I am not sure when he wrote what he did, but just this morning, I watched (replayd) a part of Vangard webcast from 12/18/2012 called Retirement income in a low-yield environment and Steve Utkus said the same, exact thing in the webcast, so it may be this young guy (Mike Piper) is digesting what he heard and regurgitating it, but it is definitely something to ponder. I ran Firecalc with different SS age after this to see how my spending power might change, after the webcast.

1. Delaying Social Security is like buying a bond that pays 8% interest, and
2. Delaying Social Security gets you an 8% return.

 

[mickeyd]

I have subscribed to the Oblivious Investor weekly newsletter for the last few months and Mike Piper has presented some very interesting ideas there.

There are many ways to think of SS. I think of it as a very flexible program that I was forced to join 50 years ago and am currently doing my best to game. Can't argue that its is not like an annuity.

 

[haha]

I don't see how what he wrote could possibly be wrong. It's correct, punto. And almost anyone who has thought about it realizes this.

The only arguments against delaying SS are 1)I am dying; 2) I don't trust the government: 3)I am very good at investing, and I can do better without increasing my risk; 4)I have no interest in any annuity type cash flow; 5) I do not like my wife, or my lower earning partner of whatever sex; 6)I cannot physically survive without getting my hands on this money asap; 7) I have government pension and for complex reasons will not get much SS anyway. There may be other highly particular reasons for preferring early draw.

Though I suppose that pretending there is really an issue here is good for generating discussion, the ultimate point of this forum after all.

Ha

 

[Mike Piper]

Thanks for sharing the article, veremchuka.

I am not sure when he wrote what he did, but just this morning, I watched (replayd) a part of Vangard webcast from 12/18/2012 called Retirement income in a low-yield environment and Steve Utkus said the same, exact thing in the webcast, so it may be this young guy (Mike Piper) is digesting what he heard and regurgitating it, but it is definitely something to ponder.

For those curious, the article was inspired by another article that I encountered recently via the Bogleheads forum, in which the writer made the case that you get an 8% return from delaying.

As I tend to do when I'm criticizing somebody else's writing, I intentionally refrained from linking to the article or mentioning the writer by name because my goal is simply to focus on the ideas rather than to offend. In addition, both of these misconceptions pop up pretty frequently, so there was no need to pick on any one person/article in particular.

I have subscribed to the Oblivious Investor weekly newsletter for the last few months and Mike Piper has presented some very interesting ideas there.

Thanks for the words of support.

 

[CH]

I am not sure when he wrote what he did, but just this morning, I watched (replayd) a part of Vangard webcast from 12/18/2012 called Retirement income in a low-yield environment and Steve Utkus said the same, exact thing in the webcast, so it may be this young guy (Mike Piper) is digesting what he heard and regurgitating it, but it is definitely something to ponder. I ran Firecalc with different SS age after this to see how my spending power might change, after the webcast.

I don't buy the concept of a "return" for delaying SS. Actuarialy it all works out the same. It you take it early, you get smaller payments over more years. If you take it later, you get larger payments over fewer years. To me, the real benefit of delaying is for married couples where one spouse earns significantly more than the other. Delaying makes it possible to create a larger, guaranteed monthly payment that adjusts for inflation for the surviving spouse that she cannot outlive.

 

[Throwdownmyaceinthehole]

I need someone smarter than me to figure this out, as non of these articles looks at SS after tax, along with the RMDs at 70. I'm thinking taking it earlier before RMDs kick in would give me a higher return. Can anyone point me to research on that?

 

[Mike Piper]

I need someone smarter than me to figure this out, as non of these articles looks at SS after tax, along with the RMDs at 70. I'm thinking taking it earlier before RMDs kick in would give me a higher return. Can anyone point me to research on that?

Dana Anspach's recent book discusses the topic a fair amount:
Control Your Retirement Destiny: Achieving Financial Security Before the Big Transition: Dana Anspach: 9781430250227: Amazon.com: Books

I've never been able to come up with anything like a broadly applicable rule of thumb, other than the general statement that, once you start receiving Social Security, there's a range over which your marginal tax rate is significantly higher than just your income tax bracket (because each additional dollar of income not only causes the normal income tax, it also causes 50 or 85 cents of Social Security to be taxable).

In many cases, it is advantageous to delay Social Security and use the pre-Social Security period of retirement to do Roth conversions, such that once Social Security kicks in, the retiree can stay in the range where none of the benefits are taxable. But, of course, this is a very case-by-case thing.

 

[CH]

I need someone smarter than me to figure this out, as non of these articles looks at SS after tax, along with the RMDs at 70. I'm thinking taking it earlier before RMDs kick in would give me a higher return. Can anyone point me to research on that?

This is the best I have read on the subject. Let me know if the link does not work:

http://research.prudential.com/docu...vativestrategies1112&bu=GC&ref=Newsroom&cid=1

 

[tmm99]

I actually didn't read the OP's whole article when I posted earlier - I just read the bullet points (I was at a restaurant...) And the vangard webcast conversation went something like this.. in case anybody is interested (It's a loooong webcast, but the subscript is available also.) I was listening to the webcast while cooking breakfast, so I was just getting the jest of it, but Steve Utkus is comparing SS with infation adjusted income stream and he advises that we delay it (at least that's what I got from it.) I don't think he ever got into "but what if we die prematurely" scenarios.

From what I gather from what Mike said, this must be a hot topic many people talk about...

Steve Utkus: Right, yeah. It's an interesting question. So one of the important things to recognize about Social Security, people don't think of this as one of the principal retirement income questions that you have to make in your transition from work to retirement, but it is. And the key thing is every year you defer Social Security is a year you buy more guaranteed inflation-adjusted income in the form of a higher payment, roughly 8% a year. So it's an incredibly efficient way to buy more guaranteed income. In fact, it's a better way to buy guaranteed income than buying a traditional annuity from an insurance company.
Rebecca Katz: So what you're saying is that when you defer, you're essentially locking in a sort of 8% return because it steps up?
Steve Utkus: Well, it's actually that, not just an 8% return, but the actual value for the rest of your life is 8% higher.
Rebecca Katz: I see.
Steve Utkus: Yeah, so it's actually substantially higher from a rate-of-return perspective. So the way to think about this is there is some advantage. If you need more guaranteed income as part of your portfolio, there is some advantage to spending from your portfolio and deferring receipt of Social Security. And that's really the trade-off.
Maria Bruno: I guess the only other thing to add there is, the flipside of that—full normal retirement age. If you take benefits prior to that, you get a reduction. So you really want to be cautious as far as what your normal retirement age is because that's increasing. Social Security website has estimates and calculators that can also help with that decision as well.
Steve Utkus: You know the advice—so I have my famous cousin example—I have a cousin who was thinking of retiring because her husband was retiring later in his 60s, and so she said, "Well, I'm going to retire at 62 because I can take Social Security." I can tell you that's the wrong answer, and I encouraged her not to think of it that way.
The way to think about it is, "How long can you wait before you take Social Security" because the benefits of that longer term in the rest of your life will far exceed the fact of whether you retire at 62 versus 63 versus 64.

 

[packrat44]

IMHO Ha has it pegged correctly.

 

[rodi]

I don't see how what he wrote could possibly be wrong. It's correct, punto. And almost anyone who has thought about it realizes this.

The only arguments against delaying SS are 1)I am dying; 2) I don't trust the government: 3)I am very good at investing, and I can do better without increasing my risk; 4)I have no interest in any annuity type cash flow; 5) I do not like my wife, or my lower earning partner of whatever sex; 6)I cannot physically survive without getting my hands on this money asap; 7) I have government pension and for complex reasons will not get much SS anyway. There may be other highly particular reasons for preferring early draw.

Though I suppose that pretending there is really an issue here is good for generating discussion, the ultimate point of this forum after all.

Ha

I'd add some more hypothetical reasons to delay SS:
- I have a family history that suggests I'll die sooner than the actuaries present... or a slow moving terminal condition... or... some other reason why I'm sure I'll die younger than older. Might as well get the cash up front.
- The lower earning spouse is significantly older than the higher earning spouse. All the file/suspend-claim on the spouses record go out the window in this scenario. And being able to upgrade the lower earning spouses SS goes out the window since they're older, and likely to die sooner. So love/hate of the spouse doesn't play at all.
- Other cash flow sources might be coming on line later and you need the cash flow sooner. (partnership payouts, real estate sales, etc.)

 

[samclem]

I think you meant :

I'd add some more hypothetical reasons not to delay SS:

 

[haha]

I think you meant :

In any case, these are marginal reasons. Older lower earning spouse- not important, since it is better to delay even for a single man; poor family history- family history is one small part of longevity. My dad expected to die almost every day past age 40. He lived to 88.

I already mentioned "I need the money". I agree that is a pretty good reason.

These tricky tax scenarios, I admit I have not really gone into them deeply. They will not affect me, and I do not do paid financial advising, so it just wasn't that interesting and it is very complex and I am not sure it can be predicted with sufficient accuracy to make it worth the effort to try. In my case, I had enough taxable income that SS was going to be 85% taxable no matter what I did. I assumed it was that way for almost every retiree, but I have since learned otherwise. Many retirees have almost all their wealth in deferred accounts, which changes this calculus.

In any case, people make most decisions emotionally, often not even knowing what emotions have been called up. So these discussions are great for a forum pastime but they have little practical significance.

The often quoted statement-"all ages are actuarially equal" is correct, for a narrow range of conditions. But those who state this are usually not aware what those conditions are. For example, how could it be possible, when the government is clearly not varying its discount rate used in the SS actuarial calculations, for different age cases to be identical in very different interest rate environments?

Ha

 

[timo2]

To me, the real benefit of delaying is for married couples where one spouse earns significantly more than the other. Delaying makes it possible to create a larger, guaranteed monthly payment that adjusts for inflation for the surviving spouse that she cannot outlive.

+1

 

[Mike Piper]

The often quoted statement-"all ages are actuarially equal" is correct, for a narrow range of conditions. But those who state this are usually not aware what those conditions are. For example, how could it be possible, when the government is clearly not varying its discount rate used in the SS actuarial calculations, for different age cases to be identical in very different interest rate environments?

This is a critical point. For an individual person, the when-to-claim decision is actuarially neutral only in a fairly rare set of circumstances. (Specifically, the person is unmarried, and current real interest rates just happen to be such that the breakeven point is equal to the person's life expectancy.)

 

[tmm99]

My dad expected to die almost every day past age 40. He lived to 88.

Nice to hear your dad could be with you much longer than he anticipated. I think we all have the tendency to think we will die young (my mom has been talking like that for many years now also, but she is still with us, and I am very happy for it.)

I am not finance savvy, or SS savvy for that matter, but I do get that it matters when you die (you will get much less total SS if you delay your SS and you die right away). But if I plug in 92 years old as my end date using Firecalc and Fido RIP both, I get a much better spending power if I claim SS at Age 76 than at Age 62. So aren't these tools telling me yes, it does matter if I delay my SS (because by delaying it, these tools are telling me I would have more spending money from the get-go when I retire?) Or is it just telling me delaying SS matters if you compare taking it at full retirement age or you start early withdrawal with reduduction in pay (=penalty)?

 

[retiredunder50]

As I understand it, taking out social security at age 62 through 70 is all the same as long as we die around 80 years old. Of course that ignores many variables - investment performance if saving the payout, tax rates, changes to program, spouse considerations... I keep in mind that if I spend down other assets to take social security later, I am most likely using resources I was planning to pass on to family - aside from my wife, nobody collects my social security when I die. One topic that is never discussed in most articles promoting us to take it later is the yearly incremental advantage that follows the break-even year. Will we really miss the several hundred to maybe a few thousand dollars per year when we are over 80 versus the use of it earlier? I like the philosophy I read about of breaking down retirement into 3 phases - go-go, slow-go, and no-go. I think I will be spending less, except for some medical, in the no-go years.

 

[Katsmeow]

But if I plug in 92 years old as my end date using Firecalc and Fido RIP both, I get a much better spending power if I claim SS at Age 76 than at Age 62.

Was that a typo? Don't you mean age 70, not 76?

 

[Brett_Cameron]

I need someone smarter than me to figure this out, as non of these articles looks at SS after tax, along with the RMDs at 70. I'm thinking taking it earlier before RMDs kick in would give me a higher return. Can anyone point me to research on that?

The only research I know of this are the complex calculated projections I have done for DW and my situation. edit/Most research is based on maximizing SS income and does not consider the whole picture. end edit/

Situation: non-COLA pension, 401k, will have one minor child at age 62. DW disabled and drawing SSDI.

Calculation assumptions:
I will have the same future expenses regardless of when I take SS.
All projections are in future dollars using BLS and SS parameters for near future projections.
Future tax estimates are based on current tax rates, deductions, credits and brackets inflated per the rules governing such.
Default inflation assumed to be 2.5% per year.
401k return assumed to be 3.5% after inflation (total 6%)
Expenses are inflated using best estimate of the inflation rate of the particular expense as multiple of base rate of 2.5%. That is to say medical and college expenses escalate at a higher rate than base inflation.

Method:
Four scenarios are projected. 1) take SS at age 62, 2) take SS at age 62 and suspend at FRA of 66 y 2 mo and restart at age 70 (this scenario involves Roth conversions between ages 62 and 66) 3) take SS at FRA of age 66 y 2 mo, 4) take SS at age 70. All income, taxes and expenses are projected for each of the next 30 years.
Comparisons are made at ages 70, 78 and 86. No one in my family has ever lived past age 86. I stopped my projections at that age.

Calculate estimated SS.
Calculate 401k withdrawals to cover projected expenses. This includes 401k balance at the beginning of every year.
Calculate estimated expenses for use in tax calculations.
Calculate taxes. This includes estimated deductions and tax credits.
Calculate net after tax income.
Calculate non-taxable portion of SS income.
Calculate DW survivor income at ages 74 and 86. This includes pension, SS, and RMDs for her.

Results Summary:
1) Sum of net after tax income is always highest for taking SS at 70 and lowest for taking SS at 62.
2) Sum of taxes paid is always highest for taking SS at 70 and lowest for taking SS at 62.
3) Annual SS income at age 70 is highest for taking SS at 70, next highest is the 62-70 scenario. Lowest is taking at age 62. The 66 and 62-70 are nearly the same.
4) 401k balance is always highest for taking at age 62. Next highest balance is for the 62-70 scenario. Taking at age 70 results in smallest 401k balance.
5) the differences among scenarios in the sum of net income plus 401k balance is always highest for taking at age 62 until sometime between age 78 and 86 when the age 62-70 scenario becomes the highest.
6) DW survivor income is always highest for taking at age 62, Next highest is the age 62-70 scenario which is about 97%. Age 66 is 87% Age 70 is 79%
7) at age 70, SS+pension+RMDs will cover all expenses with a surplus of about 5% for the age 62-70 scenario. The other scenarios do this starting at age 71. Surplus at age 71 is highest for age 70 followed by 62-70 then 66 and then age 62 scenarios. This is a function of my specific expense profile.
8) the sum of total SS income at age 86 is highest for the age 70 scenario, 11% lower for the age 62 scenario and 5% lower for the 66 and 62-70 scenario.

Criteria for selecting which scenario to implement:
1) must cover projected expenses with surplus (all meet this)
2) must allow for sizable 401k balance to fund long term care if/when needed (only 62 and 62-70 meet this) and/or other unaccounted for expenses.
3) must allow for high DW survivor benefits at my age 74 and 86. (Edited out error)
4) must strike a balance between reliance on 401k performance and SS income ( 62-70 and 66 meet this)
5) must be able to survive 25% cut in future SS benefits (62 and 62-70 meet this)
6) Firecalc success rate must stay above 95% to age 86 ( 62 and 62-70 meet this)


So based on this fantasy calculation, I have decided to start SS at 62, suspend at FRA and restart at age 70. No real surprises here from the calculations.

I understand that this is a fantasy calculation based on many variables and assumptions that will change. I did this as an exercise to attempt to quantify many of aspects of consequences of taking SS at various ages along with tax consequences and ability to maintain some flexibility and to compare DW survivor benefits under various circumstances.


This is all generated by spreadsheet except for the Firecalc calculations. I can model different inflation rates, 401k return rates. I can update periodically with actual returns, inflation, changes in projected spending, etc.

The selected scenario will likely change in the future as I approach 62. If the 401k performs well I may delay taking SS. That is a job for Futureman.


I hope this is intelligible and can provide some insight into your evaluation of your plan.

 

[tmm99]

Was that a typo? Don't you mean age 70, not 76?

Hi Katsmeow,

Thank you for catching it. Yes that was a typo. I to the numbers transposed. I meant to say age 67, not age 76. Correction in italic below.

But if I plug in 92 years old as my end date using Firecalc and Fido RIP both, I get a much better spending power if I claim SS at Age 67 than if I cliam it at Age 62.

So basically, what I was trying to say was that despite that people were saying delaying SS doesn't matter, by doing so, I am getting a better spending power by delaying it even from Age 60 when I plan to retire. It means I am not sacrificing anything. Am I misunderstanding something here? (Like I said, I am not finance or SS savvy, so maybe I am missing something huge...)

One topic that is never discussed in most articles promoting us to take it later is the yearly incremental advantage that follows the break-even year. Will we really miss the several hundred to maybe a few thousand dollars per year when we are over 80 versus the use of it earlier? I like the philosophy I read about of breaking down retirement into 3 phases - go-go, slow-go, and no-go. I think I will be spending less, except for some medical, in the no-go years.

Like I said in the previous paragraph, it's not that I get more money when I am older. I get more money in the go-go years also, at least by Firecalc and Fido RIP. The only thing I changed in both tools was my SS starting year and the estimated amount of SS payout.

Am I missing something here? Of course, the total withdrawal of SS may be lower if I start late and die early, but if my spending power is higher by doing it from Age 60, why would I want to take less early? If I spend the same amount regardless of which options I choose (start SS or delay until 67), I will have extra 400K left in the bank if I die at Age 92 (according to Fido RIP).

Am I missing something HUGE here? Everything points me to want to delay SS...?

 

[veremchuka]

Who knew Mike was here!

For me I started at 62 for lots of reasons. Breakeven for a male is what 78 years? So if I die sooner I made the right call and if I die later then maybe I should have waited. So complicated but for me I think I made the right call. Time will tell.

 

[gerntz]

Since one can start collecting 1/2 spouse's FRA benefit when you reach your FRA, the increase in delaying beyond your FRA is greater than 8% annually till 70 by: (your FRA benefit/(your FRA benefit less 1/2 spouse's FRA benefit)) since you're account is growing fully while you collect something in the interim. Sweet.