I’m Planning to Claim SS @62… Well, Why Not? ►Updated w/Funded Ratio

Non-investing personal finance issues including insurance, credit, real estate, taxes, employment and legal issues such as trusts and wills.
randomguy
Posts: 11295
Joined: Wed Sep 17, 2014 9:00 am

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by randomguy »

bradpevans wrote: Tue Sep 21, 2021 6:06 pm
wrongfunds wrote: Tue Sep 21, 2021 4:43 pm Even though I might be called stupid, I do NOT understand that graph. What is on y-axis? It goes from 1% to 4.5% What does life span has to do with that number?
the y-axis = chance of dying at that age.

To die at 73 covers the 365 day window from your 73rd birthday to the day before your 74th

if you summed the dots (for each color) you would get 100% because no one lives forever
I think a lot of people are looking for the graph that shows your odds of dying in the next year IF you are still alive at the start of the year. So you get something that starts at like .5% at 60 and ramps up to to like 10% at 90 and 25% at 95. The other graph that is useful is the one that gives your odds of being alive at a given age. You start at 100% and it rapidly approaches zero when your 100...
User avatar
ObliviousInvestor
Posts: 4212
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by ObliviousInvestor »

iceport wrote: Wed Sep 22, 2021 10:55 am First, a calculator like Mike Piper's Open Social Security site figures out a person's life expectancy and computes the theoretically optimum time to claim SS — all based on some kind of optimization algorithm that weights the PV of total benefits possible by claiming at various ages, and weights those values by the probability of being alive at those ages. <Honestly, I don't know how that calculation is made, what goes into it. And that's probably part of the issue here.>
What the calculator does is pretty straightforward. (At least conceptually. It's a wild amount of math to try to replicate manually. And the rules get complicated sometimes as well of course.)

What follows is the process for a married couple, because if you understand that then you can certainly understand what it's doing for the simpler situation for a single person.

It looks at the earliest possible combination of filing ages. Then, for that combination, it does a year-by-year calculation to determine the dollar amount of Social Security benefits received by the household that year if:
a) PersonA and PersonB are both alive,
b) PersonA is alive and PersonB is deceased,
c) PersonA is deceased and PersonB is alive, and
d) PersonA and PersonB are both deceased.

Then it calculates, for that particular year, the probability of each of those 4 mortality scenarios, given the mortality tables selected. And it weights the benefit amounts calculated in each of those 4 scenarios by the corresponding probabilities, and it sums those probability-weighted amounts. Then that sum is discounted back to its present value using the discount rate selected.

Then it does the same calculation for the next year. And it keeps going until the year in which both spouses have reached age 115.

So what you get then is the total expected present value for that combination of filing ages.

Then it does the same calculation for all the possible combinations of filing ages and reports which one had the highest total expected present value.
Mike Piper | Roth is a name, not an acronym. If you type ROTH, you're just yelling about retirement accounts.
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

ObliviousInvestor wrote: Wed Sep 22, 2021 4:07 pm
iceport wrote: Wed Sep 22, 2021 10:55 am First, a calculator like Mike Piper's Open Social Security site figures out a person's life expectancy and computes the theoretically optimum time to claim SS — all based on some kind of optimization algorithm that weights the PV of total benefits possible by claiming at various ages, and weights those values by the probability of being alive at those ages. <Honestly, I don't know how that calculation is made, what goes into it. And that's probably part of the issue here.>
What the calculator does is pretty straightforward. (At least conceptually. It's a wild amount of math to try to replicate manually. And the rules get complicated sometimes as well of course.)

What follows is the process for a married couple, because if you understand that then you can certainly understand what it's doing for the simpler situation for a single person.

It looks at the earliest possible combination of filing ages. Then, for that combination, it does a year-by-year calculation to determine the dollar amount of Social Security benefits received by the household that year if:
a) PersonA and PersonB are both alive,
b) PersonA is alive and PersonB is deceased,
c) PersonA is deceased and PersonB is alive, and
d) PersonA and PersonB are both deceased.

Then it calculates, for that particular year, the probability of each of those 4 mortality scenarios, given the mortality tables selected. And it weights the benefit amounts calculated in each of those 4 scenarios by the corresponding probabilities, and it sums those probability-weighted amounts. Then that sum is discounted back to its present value using the discount rate selected.

Then it does the same calculation for the next year. And it keeps going until the year in which both spouses have reached age 115.

So what you get then is the total expected present value for that combination of filing ages.

Then it does the same calculation for all the possible combinations of filing ages and reports which one had the highest total expected present value.
Thanks so much for the explanation, Mike! I really appreciate your help with this. You might think this is all conceptually straightforward, but I'm struggling to understand it fully.

So the probabilities of those 4 mortality scenarios must add up to 100%, correct?

Also, since the PIA is in today's dollars, aren't all the benefit amounts already computed in today's dollars? Why the need to discount back to today's dollars? I seem to have a mental block with this detail. I realize the fact that benefits not received yet must be valued differently than benefits received today, but isn't that already accounted for by not inflating the benefits to future values to begin with? Somehow I've got it in my head that the future benefits are being doubly discounted.

But the explanation does help fill in some of the gaps in my understanding.

Even so, I still find the magnitude of the potential shortfalls from optimum expected present values that occur in the event of deaths at particularly sub-optimum ages to be disconcerting. And the probability of death coming within a range of particularly sub-optimum ages is not at all trivial; it's actually substantial. I'm referring to the risk of a person claiming at 62 living an especially long life, or of a person deciding to claim at 70 dying within a few years on either side of 70.

In your opinion, do you think it's worthwhile to consider such risks in conjunction with the present value analysis?

And if so, depending on one's particular circumstances, do you think it might be reasonable to take steps to hedge against some of the worst possible (relative) outcomes?
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
User avatar
ObliviousInvestor
Posts: 4212
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by ObliviousInvestor »

iceport wrote: Wed Sep 22, 2021 5:53 pm So the probabilities of those 4 mortality scenarios must add up to 100%, correct?
Yes.
iceport wrote: Wed Sep 22, 2021 5:53 pm Also, since the PIA is in today's dollars, aren't all the benefit amounts already computed in today's dollars? Why the need to discount back to today's dollars?
Discounting has to do with more than just inflation. There's an opportunity cost whenever dollars are received later rather than sooner. By default the calculator uses the yield on 20-year TIPS at the discount rate. The following article and Bogleheads discussion explain the rationale for such.

https://obliviousinvestor.com/claiming- ... we-assume/
viewtopic.php?p=5362672#p5362672
iceport wrote: Wed Sep 22, 2021 5:53 pm Even so, I still find the magnitude of the potential shortfalls from optimum expected present values that occur in the event of deaths at particularly sub-optimum ages to be disconcerting. And the probability of death coming within a range of particularly sub-optimum ages is not at all trivial; it's actually substantial. I'm referring to the risk of a person claiming at 62 living an especially long life, or of a person deciding to claim at 70 dying within a few years on either side of 70.

In your opinion, do you think it's worthwhile to consider such risks in conjunction with the present value analysis?

And if so, depending on one's particular circumstances, do you think it might be reasonable to take steps to hedge against some of the worst possible (relative) outcomes?
Short answer, probably not -- at least not in the way you are thinking.

It may be helpful to back up a step and ask what the goal is here.

For many people, the goal is to reduce the likelihood of outliving their money, so delaying is a relatively easy decision since it's a "win" from that point of view while also being (usually) the best from the standpoint of maximizing expected PV.

I believe you stated up-thread that longevity risk is not really a concern. That is, you're confident you're unlikely to run out of money during retirement. So what is the goal then? For most people I work with as clients, the goal then becomes one of maximizing the expected amount to be left to heirs. (Note that this is not the same as maximizing the amount of Social Security collected.)

In other words, if you want to look at a distribution of outcomes, you likely should be using a different metric (because if you die at, e.g., age 93 as opposed to age 73 there are other things that will be different between those two scenarios than just the amount of Social Security collected).

As a separate point the analysis should be on an after-tax basis (which is often a point in favor of delaying).
Mike Piper | Roth is a name, not an acronym. If you type ROTH, you're just yelling about retirement accounts.
User avatar
grabiner
Advisory Board
Posts: 35307
Joined: Tue Feb 20, 2007 10:58 pm
Location: Columbia, MD

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by grabiner »

bradpevans wrote: Wed Sep 22, 2021 12:52 pm as a rough rule, the "breakeven" point is about 13-14 years after the start date.
It's easiest to estimate this for a single year. If your FRA is 66 and you claim at 67 instead, your benefit is 8% more, say $20,000 versus $21,600. If your investments match inflation, it would take 12.5 years after age 67 to make up the $1600 per year, so the break-even is 13.5 years after the age-66 start. Likewise, the break-even is 16.5 years for $24,800 at age 69 versus $26,400 at age 70.
if you delay to 70, you need to live to 83-84 to have made the right choice (cashflow, not considering gains/losses)
For 69 versus 70, the break-even is age 85-86, which is less than the life expectancy for a single male.
At least 15% of SS is never taxed, so that helps a bit too for the "you should delay" crowd
If the full 85% of SS is taxed, this is essentially a wash. If you are in the 22% tax bracket, you lose 18.7% of your SS to taxes, so your after-tax SS benefit is 81.3% of your pre-tax benefit whenever you take it.

If you are in the middle of the phase-in, there is usually a small benefit for waiting. Suppose that $20,000 of SS puts you in the 85% phase-in range. If you delay one year and have $21,600 in SS at age 67, the extra $1600 in SS counts as only $800 of income for making SS taxable, so you pay tax on only an additional $680 of SS, which is 42.5% of the increase. Thus you pay tax on the same fraction of SS if 42.5% of the original $20,000 was taxable, or on a smaller fraction if it was more than 42.5%.

Conversely, if you are at the bottom of the phase-in, there might be a benefit for not waiting (although the lack of inflation adjustments on the phase-in means that you won't stay at the bottom). If you can take up to $20,000 of SS and pay no tax on it, then delaying one year to 67 and taking $21,600 means that you will pay tax on $400 of SS.
Wiki David Grabiner
randomguy
Posts: 11295
Joined: Wed Sep 17, 2014 9:00 am

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by randomguy »

iceport wrote: Wed Sep 22, 2021 10:55 am .

I was ribbed at the start of this thread for being willing to leave $65k on the table by claiming early. Well, how do all the folks claiming at 70 feel about the possibility of leaving $183k on the table, instead, if they go early?

Still confused why nobody seems to care about that, I started wondering if it was because the chances of dying so much earlier than expected are too slim to worry about. But that didn't make intuitive sense. So I wanted to get some kind of handle on how likely it would be to die in the years from, say, 65 to 75. I still don't know how to compute the probability, but just eyeballing the area under the 62 YO curve leads me to estimate that it could be somewhere in the vicinity of 25%.

If you alive at 90, you care about money. If your dead at 70, you don't care about money....

SS life expectancy stats are pretty useless when looking at your case. Go run one of the life insurance ones that factors in weight, smoking, and prior health issues. It is easy to pick up(or lose) half a decade of life expectancy.

And studies have shown genetics don't matter a ton. You need to look at what did people who died early and decide how much applies to you. There is always the risk of going early(or living to 110).

In the end the the strongest arguments for delaying SS isn't that you will end up with more money. It is that longevity component if you live 95+. It isn't very likely but it also isn't a rare event...
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by HomerJ »

I know way too many people who have died in their 50s and 60s to worry about living to 90
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
Nestegg_User
Posts: 2112
Joined: Wed Aug 05, 2009 1:26 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by Nestegg_User »

(well, Homer, I've also known those that passed in their 30's, mid-40's, and turning 50...and with neither parent making it to collect SS I do understand some peoples mindset...
fortunately, we've saved enough to be able to defer, and I'm already "medicare age" :? )

iceport
not everyone has a legacy desire {see Pfau : https://retirementincomejournal.com/art ... cess-tool/} certainly we don't, we were DINKs

What we DO desire was a sufficient stream of income, from various sources, that allows both or either of us when one passes to continue to enjoy a decent income and the lifestyle that it entails.

Towards that end having a decent SS for the first to file, together with a pension and income from a portfolio (taxable, 401k/IRA, and Roths), will allow a good income (even before SS, we've been 2-2 1/2 % wr). Roth conversions before SS started allows some of the 401k/IRA to be preferentially converted at lower rates (for us, once SS is started there would be no way to stay in the lower rate).

After SS for first is started, then plan on having SS plus pension and withdrawals from the 401k/IRA to further drop its level (limiting future RMD's) as once the second SS starts we'd be getting closer to the IRMAA level (with any reasonable interest/cap gain). { if so, it would be 2X the IRMAA surcharge...until the first passes (and likely for the duration) - - so the desire is to avoid being in the IRMAA crosshairs in the first place.}
{The second to file, at age 70 (expected under our IPS) has the slightly higher PIA. If the first filer passes before the second reaches 70, then there's still a good SS (as survivor benefit) plus (at least half) pension___ and the remaining portfolio! That's the incentive to not more extensively draw down the portfolio but instead start the first SS before 70. {most have the lower PIA file at 62... but, for us, that would have precluded significant conversions (at decent tax savings) hence the decision for just past FRA (somewhat of a compromise, but it achieves the desired outcome)}
vested1
Posts: 3496
Joined: Wed Jan 04, 2012 3:20 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by vested1 »

Nestegg_User wrote: Wed Sep 22, 2021 11:05 pm (well, Homer, I've also known those that passed in their 30's, mid-40's, and turning 50...and with neither parent making it to collect SS I do understand some peoples mindset...
fortunately, we've saved enough to be able to defer, and I'm already "medicare age" :? )

iceport
not everyone has a legacy desire {see Pfau : https://retirementincomejournal.com/art ... cess-tool/} certainly we don't, we were DINKs

What we DO desire was a sufficient stream of income, from various sources, that allows both or either of us when one passes to continue to enjoy a decent income and the lifestyle that it entails.

Towards that end having a decent SS for the first to file, together with a pension and income from a portfolio (taxable, 401k/IRA, and Roths), will allow a good income (even before SS, we've been 2-2 1/2 % wr).
Roth conversions before SS started allows some of the 401k/IRA to be preferentially converted at lower rates (for us, once SS is started there would be no way to stay in the lower rate).

After SS for first is started, then plan on having SS plus pension and withdrawals from the 401k/IRA to further drop its level (limiting future RMD's) as once the second SS starts we'd be getting closer to the IRMAA level (with any reasonable interest/cap gain). { if so, it would be 2X the IRMAA surcharge...until the first passes (and likely for the duration) - - so the desire is to avoid being in the IRMAA crosshairs in the first place.}
{The second to file, at age 70 (expected under our IPS) has the slightly higher PIA. If the first filer passes before the second reaches 70, then there's still a good SS (as survivor benefit) plus (at least half) pension___ and the remaining portfolio! That's the incentive to not more extensively draw down the portfolio but instead start the first SS before 70. {most have the lower PIA file at 62... but, for us, that would have precluded significant conversions (at decent tax savings) hence the decision for just past FRA (somewhat of a compromise, but it achieves the desired outcome)}
And as with Homer, those who died prior to the average life expectancy were the exception, not the rule, which the OP seems to have ignored.

This is our plan in a nutshell, and because of that our income, rather than decreasing every year will continue to increase every year. At my wife's age and my own respectively of 71/72, our income will surpass the highest amount ever made while we were employed, but still keep us under IRMAA thresholds. Income will continue to rise unless and until RMD's lower the amount required to withdraw, which is unlikely. The largest portion of that income (SS) at over 70k will be inflation protected by COLA.

This yearly increase in income is primarily due to the delay of SS. My wife filed at age 65, when I filed simultaneously for a restricted application on her benefit at my FRA. I will receive my first maximum benefit check in 11 months, 1 month after I turn 70. The survivor, should one of us die immediately, will still be better off than if we had filed earlier.

Contrary to statements that identify a definite age for breakeven, those dates vary for every individual. Mine is about age 77.5 because of the restricted application money being received for over 3 years now. While this method isn't available to younger recipients it should not be ignored by those eligible, if breakeven is their primary concern. Breakeven has no meaning for me, as SS benefits are income, not a savings account. I'm not keeping score with the government, which IMHO would only lead to disappointment.
BigJohn
Posts: 2627
Joined: Wed Apr 02, 2014 11:27 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by BigJohn »

Like many of the topics that get debated here frequently, I think it's all about what risk you choose to manage. Just like deciding on an asset allocation or whether you should hold TIPS, you can't manage all the risks equally so choices have to be made.....
randomguy wrote: Wed Sep 22, 2021 9:18 pm If you alive at 90, you care about money. If your dead at 70, you don't care about money....

In the end the the strongest arguments for delaying SS isn't that you will end up with more money. It is that longevity component if you live 95+. It isn't very likely but it also isn't a rare event...
Some people don't care about breakeven points or NPV analysis....
iceport wrote: Wed Sep 22, 2021 5:53 pm Even so, I still find the magnitude of the potential shortfalls from optimum expected present values that occur in the event of deaths at particularly sub-optimum ages to be disconcerting. And the probability of death coming within a range of particularly sub-optimum ages is not at all trivial; it's actually substantial. I'm referring to the risk of a person claiming at 62 living an especially long life, or of a person deciding to claim at 70 dying within a few years on either side of 70.
To others, it's just the opposite....
ObliviousInvestor wrote: Wed Sep 22, 2021 7:10 pm It may be helpful to back up a step and ask what the goal is here.
So take Mike's advice and decide which risk is most important to you and then manage it in the best way possible for your circumstances recognizing two things. First, you are unlikely to change the mind of anyone who's decided that the other risk is more important. Second, see my signature.
"The greatest enemy of a good plan is the dream of a perfect plan" - Carl Von Clausewitz
Angst
Posts: 2968
Joined: Sat Jun 09, 2007 11:31 am

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by Angst »

Hasn't this thread pretty much "jumped the shark"?
There are plenty of questions in life in life for which there is no definitive answer, and yours may be one them.

If you're one to feel better about having the insurance of a slightly larger real life income annuity from SS rather than a slightly lower one, then delay a bit. If you're one who really looks forward to the money from SS, take it earlier. And if your doctor says you could be dead in 12 months, that's worth considering too, but if you think you'll live to 100, you might decide to delay.

Here's a thought experiment for everyone:
The actuaries at SS could hypothetically extend the early & delayed claiming ages - it's not rocket science. What would you do if you could claim as early as 55 and as late at 75?

Would both the knee-jerk "take it at 62" crowd and the knee-jerk "delay it 'til 70" crowd just move their respective goalposts? At some point, I think not! It's easy to just commit to one crowd or the other, without really thinking about it all too much. (And what about the FRA crowd?!)

I suggest the choice of when to claim largely boils down to a guess about how long you'll live and how that works with whatever worries you do or don't have about your current financial resources. This thread has probably already provided all the information it can, including some nice spreadsheets for those marginal age decisions.
JackoC
Posts: 4714
Joined: Sun Aug 12, 2018 11:14 am

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by JackoC »

randomguy wrote: Wed Sep 22, 2021 9:18 pm
iceport wrote: Wed Sep 22, 2021 10:55 am .

I was ribbed at the start of this thread for being willing to leave $65k on the table by claiming early. Well, how do all the folks claiming at 70 feel about the possibility of leaving $183k on the table, instead, if they go early?

Still confused why nobody seems to care about that, I started wondering if it was because the chances of dying so much earlier than expected are too slim to worry about. But that didn't make intuitive sense. So I wanted to get some kind of handle on how likely it would be to die in the years from, say, 65 to 75. I still don't know how to compute the probability, but just eyeballing the area under the 62 YO curve leads me to estimate that it could be somewhere in the vicinity of 25%.

If you alive at 90, you care about money. If your dead at 70, you don't care about money....

SS life expectancy stats are pretty useless when looking at your case. Go run one of the life insurance ones that factors in weight, smoking, and prior health issues. It is easy to pick up(or lose) half a decade of life expectancy.

And studies have shown genetics don't matter a ton. You need to look at what did people who died early and decide how much applies to you. There is always the risk of going early(or living to 110).

In the end the the strongest arguments for delaying SS isn't that you will end up with more money. It is that longevity component if you live 95+. It isn't very likely but it also isn't a rare event...
Exactly, but I (and others, maybe you too) have posted basically the same thing many times and 'brick wall' with some people. I guess the people saying "it's really unlikely 'you' live to your 90's" or mention all the people they know who have died young actually in many cases may know they have factors and 'comorbidities' that make the SS table applicable to them, or even make it overstate their life expectancy. But understandably they might not want to parade those personal situations in front of the forum even quasi-anonymously. I get that, if so. But if (what I believe are) typical higher income/education than average people of this forum (with which tends to go w/ lifestyle habits and situations which extend LE) think in general the SS table applies to them or it's really unlikely they live past 90, they're wrong.

And also worth beating this dead horse: life expectancy has little directly to do with older relatives' lifespans. Put your particulars into several different online personalized calculators and see what the results are and if relatively consistent among them.
User avatar
Cash is King
Posts: 578
Joined: Sun Dec 10, 2017 8:04 am

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by Cash is King »

Angst wrote: Thu Sep 23, 2021 8:23 am (And what about the FRA crowd?!)

Great question. The right fighters on both sides are only interested in convincing everyone to either take it early at 62 or delay until 70. :oops:
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

ObliviousInvestor wrote: Wed Sep 22, 2021 7:10 pm
iceport wrote: Wed Sep 22, 2021 5:53 pm Also, since the PIA is in today's dollars, aren't all the benefit amounts already computed in today's dollars? Why the need to discount back to today's dollars?
Discounting has to do with more than just inflation. There's an opportunity cost...
You can't see it, but my head is bowed in shame. (And here I was trying to look at the big picture... :oops: )

ObliviousInvestor wrote: Wed Sep 22, 2021 7:10 pm
iceport wrote: Wed Sep 22, 2021 5:53 pm Even so, I still find the magnitude of the potential shortfalls from optimum expected present values that occur in the event of deaths at particularly sub-optimum ages to be disconcerting. And the probability of death coming within a range of particularly sub-optimum ages is not at all trivial; it's actually substantial. I'm referring to the risk of a person claiming at 62 living an especially long life, or of a person deciding to claim at 70 dying within a few years on either side of 70.

In your opinion, do you think it's worthwhile to consider such risks in conjunction with the present value analysis?

And if so, depending on one's particular circumstances, do you think it might be reasonable to take steps to hedge against some of the worst possible (relative) outcomes?
Short answer, probably not -- at least not in the way you are thinking.

It may be helpful to back up a step and ask what the goal is here.

For many people, the goal is to reduce the likelihood of outliving their money, so delaying is a relatively easy decision since it's a "win" from that point of view while also being (usually) the best from the standpoint of maximizing expected PV.

I believe you stated up-thread that longevity risk is not really a concern. That is, you're confident you're unlikely to run out of money during retirement. So what is the goal then? For most people I work with as clients, the goal then becomes one of maximizing the expected amount to be left to heirs. (Note that this is not the same as maximizing the amount of Social Security collected.)

In other words, if you want to look at a distribution of outcomes, you likely should be using a different metric (because if you die at, e.g., age 93 as opposed to age 73 there are other things that will be different between those two scenarios than just the amount of Social Security collected).

As a separate point the analysis should be on an after-tax basis (which is often a point in favor of delaying).
Thank you very much, Mike, for your explanation. All you say makes perfect sense — all except the first line.

I was actually asking from the standpoint of assuming all of that, taking for granted that there are several potential objectives, some of which could conflict with and override any PV optimization endeavor. So it was from the standpoint of all other things being equal.

One of the basic premises of the thread is that the PV analysis is of limited interest or consequence for me. The funny thing is, though, during the course of the discussion I came to realize that that's not entirely true, and that claiming at either extreme end of the possible range subjects me to the risk of the poorest potential PV outcomes. And the dollar values involved are actually large enough to get me to sit up and take notice. And that led to the notion that I'd be glad to give up a shot at the best possible PV outcomes in the interest of preventing the worst. Claiming sometime in the middle years is a way of narrowing the dispersion of PV outcomes.

I'm certainly not requesting or expecting any specific personal advice from you through this medium. But my perspective aligns very closely with yours on other goals and other metrics likely being more important. Unfortunately, it all seems like a fairly crude balancing act between sometimes competing objectives.

The factors now under consideration include:

1) portfolio preservation — essentially, this is maximizing what's left for heirs, and it favors claiming early
2) longevity risk — maximizing SS income in the outer years, which favors claiming late
3) maximizing expected present value of benefits — though I believe some hedging here against the worst possible outcomes makes a great deal of sense, the highest *expected* value also tends to favor claiming late
4) psychological factors — this primarily takes the form of wanting (maybe even craving, after a lifetime of delaying...) the immediate gratification of claiming early, but it's also a major driver in objective 1

Where I'm ending up is probably somewhere in the middle, after considering all of the above.

But perhaps more importantly, this discussion has confirmed to a certain degree that — in my personal circumstances — the choice is not a particularly crucial one.

Thank you again, Mike, for lending a helping hand in my stumble towards greater clarity on this complex problem.
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
User avatar
ObliviousInvestor
Posts: 4212
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by ObliviousInvestor »

iceport wrote: Thu Sep 23, 2021 12:24 pm Claiming sometime in the middle years is a way of narrowing the dispersion of PV outcomes.
But why would this be a useful goal? It strikes me as akin to trying to narrow the dispersion of outcomes of just your international stock fund rather than of the whole portfolio.

If what you care about is bequest, then there's a directly offsetting item here (dollars spent, which increases as lifespan increases). By delaying, you narrow the dispersion of the net amount left to heirs.
Mike Piper | Roth is a name, not an acronym. If you type ROTH, you're just yelling about retirement accounts.
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

vested1 wrote: Thu Sep 23, 2021 5:31 am And as with Homer, those who died prior to the average life expectancy were the exception, not the rule, which the OP seems to have ignored.
Huh? By definition, doesn't half the population die before their life expectancy?

I'm interested in both ends of the mortality spectrum. Using #Cruncher's excellent spreadsheet, it's easy to set the age of Pers1 to 62 in Cell I3, and sum up the probabilities of dying in any age range of interest in Column P.

Some randomly selected age ranges, and the probabilities of dying within that range:

62—69: 13%
65—74: 20%
62—74: 24%

86—90: 20%
91—110: 18% (eye-opening!)

I don't consider any of those probabilities small enough to be classified as "exceptions."
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by vanbogle59 »

https://www.johnhancockinsurance.com/li ... lator.html
I don't know if I should be suspicious because they are trying to convince me I will live forever or that I will die tomorrow, but....
I put in my numbers;
62, male....93
DW... 97
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

ObliviousInvestor wrote: Thu Sep 23, 2021 12:34 pm
iceport wrote: Thu Sep 23, 2021 12:24 pm Claiming sometime in the middle years is a way of narrowing the dispersion of PV outcomes.
But why would this be a useful goal? It strikes me as akin to trying to narrow the dispersion of outcomes of just your international stock fund rather than of the whole portfolio.

If what you care about is bequest, then there's a directly offsetting item here (dollars spent, which increases as lifespan increases). By delaying, you narrow the dispersion of the net amount left to heirs.
My initial reaction is that it's just as interesting as the basic PV analysis is. It almost seems as though you are downplaying the value of performing the PV analysis in the first place.

As I see it, if PV is of any interest at all, then identifying and factoring in the full range of *possible* PV outcomes is every bit as interesting as identifying only the highest *expected* PV. Why go through the effort of performing the massive number-crunching of a PV analysis and then ignore much of the resulting data?

Let me ponder this more, though. I must surely be missing something...
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
User avatar
Cash is King
Posts: 578
Joined: Sun Dec 10, 2017 8:04 am

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by Cash is King »

It's clear after 6 pages that analysis paralysis is a real problem for some indivudals.
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

Cash is King wrote: Thu Sep 23, 2021 1:01 pm It's clear after 6 pages that analysis paralysis is a real problem for some indivudals.
Ya think???

Actually, I've made enormous progress. The plan is firming up around claiming at 66, with all things considered. The thread has been an enormous help already.

:beer 8-)
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
User avatar
Cash is King
Posts: 578
Joined: Sun Dec 10, 2017 8:04 am

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by Cash is King »

iceport wrote: Thu Sep 23, 2021 1:12 pm
Cash is King wrote: Thu Sep 23, 2021 1:01 pm It's clear after 6 pages that analysis paralysis is a real problem for some indivudals.
Ya think???

Actually, I've made enormous progress. The plan is firming up around claiming at 66, with all things considered. The thread has been an enormous help already.

:beer 8-)
:thumbsup
User avatar
bobcat2
Posts: 6076
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by bobcat2 »

From Center for Retirement Research at Boston College's Squared Away Blog. (9/23/21)
The Consequences of Current Benefit Adjustments for Early and Delayed Claiming
The option to claim early was introduced just over 60 years ago, when Congress set 62 as the program’s earliest eligibility age. The option to claim between 65 and 70 on an actuarially fair basis stems from the 1983 Social Security amendments, which gradually increased the annual “delayed retirement credit” from 3 percent to 8 percent. Also in 1983, reductions for early claiming were changed in tandem with the gradual increase in the full retirement age from 65 to 67.

The goal of actuarial adjustments to the monthly benefits has always been to ensure that retirees with average life expectancy could expect to get the same total lifetime benefits, regardless of when they started. But calculating lifetime benefits requires assumptions about how long people will live and assumptions about interest rates. The current calculations are based on life expectancy and interest rates in the early 1960s or 1980s. Much has changed since those dates: life expectancy has increased dramatically and interest rates have declined. Longer life expectancy and, to a lesser extent, lower interest rates would each call for a smaller penalty for early claiming and a smaller reward for delaying claiming.

Consider what this means for baby boomers whose full retirement age is 67. Under the current system, if they claim at 62, they receive 70 percent of their age-67 benefit. However, to reflect decades of increasing life spans and falling interest rates, the researchers calculated that the accurate monthly benefit would be 77.5 percent of the age-67 benefit. That is, early claimers are penalized too much. For workers who delay claiming, a discrepancy also exists between the current and accurate delayed retirement credits, though the difference is smaller since the credit was initially too small. Specifically, workers who wait until 70 to start Social Security today receive 124 percent of the benefit they would’ve gotten at 67, whereas 120 percent of the age-67 benefit would be more accurate.
Link to underlying paper - https://crr.bc.edu/working-papers/the-c ... -claiming/

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
User avatar
ObliviousInvestor
Posts: 4212
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by ObliviousInvestor »

iceport wrote: Thu Sep 23, 2021 12:55 pm
ObliviousInvestor wrote: Thu Sep 23, 2021 12:34 pm
iceport wrote: Thu Sep 23, 2021 12:24 pm Claiming sometime in the middle years is a way of narrowing the dispersion of PV outcomes.
But why would this be a useful goal? It strikes me as akin to trying to narrow the dispersion of outcomes of just your international stock fund rather than of the whole portfolio.

If what you care about is bequest, then there's a directly offsetting item here (dollars spent, which increases as lifespan increases). By delaying, you narrow the dispersion of the net amount left to heirs.
My initial reaction is that it's just as interesting as the basic PV analysis is. It almost seems as though you are downplaying the value of performing the PV analysis in the first place.

As I see it, if PV is of any interest at all, then identifying and factoring in the full range of *possible* PV outcomes is every bit as interesting as identifying only the highest *expected* PV. Why go through the effort of performing the massive number-crunching of a PV analysis and then ignore much of the resulting data?

Let me ponder this more, though. I must surely be missing something...
In a situation in which portfolio depletion is not a significant possibility, if what we care about is bequest, maximizing expected PV also has the effect of maximizing expected bequest (on a before-tax basis).

An analogous thing cannot be said regarding distribution of expected PV (for various life expectancies, with claiming age held constant) vs distribution of expected bequest.
Mike Piper | Roth is a name, not an acronym. If you type ROTH, you're just yelling about retirement accounts.
vested1
Posts: 3496
Joined: Wed Jan 04, 2012 3:20 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by vested1 »

iceport wrote: Thu Sep 23, 2021 12:40 pm
vested1 wrote: Thu Sep 23, 2021 5:31 am And as with Homer, those who died prior to the average life expectancy were the exception, not the rule, which the OP seems to have ignored.
Huh? By definition, doesn't half the population die before their life expectancy?


I don't consider any of those probabilities small enough to be classified as "exceptions."
No, 50% die below their life expectancy and 50% die above it. Those who die earlier or later are beating the odds, which is why they are the exception.

Life expectancy increases gradually with every year of life because the individual has avoided fatal accidents and/or illnesses.

The following table is listed on the SS website, for life expectancy at every age, both male and female, (50% above and 50% below). A newborn's life expectancy of a male is 76.23 and for a female newborn is 81.28. The life expectancy of a 62 year old male is 82.28 and for a 62 year old female it's 85.14. After attaining the age of 67, for a male it's 83.67 and a female it's 86.1. For a 70 year old a male it's 84.6 and for a female 86.76. Taken to the extreme end of these examples, a male 70 year old, on average will live 8.37 years longer than a newborn.

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

A hypothetical couple who are both age 65 has a 50% chance of one of them living to age 89, according to this article by Michael Kitces. In fact, joint life expectancy is longer for a couple than for a single according to his chart in the article.

https://www.kitces.com/blog/life-expect ... survivors/

I'm not trying to convince you to file early or to delay, just pointing out that the odds of living longer are better than you may be assuming they are. Playing the odds is what we all do to a certain degree, which is the motive for obtaining insurance in order to hedge that bet. In this case, the hedge is longevity insurance.
Tom_T
Posts: 4836
Joined: Wed Aug 29, 2007 2:33 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by Tom_T »

Life expectancy is generally expressed as an average, not a median. The life expectancy for a newborn U.S. male is 76, but I will bet that the median is over 80. The SSA tables are averages. If the SSA table says that a 65-year-old male has a life expectancy of 18 years, that doesn't mean that half the 65-year-olds will die before then. It will be less than half.
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

ObliviousInvestor wrote: Thu Sep 23, 2021 3:50 pm
iceport wrote: Thu Sep 23, 2021 12:55 pm
ObliviousInvestor wrote: Thu Sep 23, 2021 12:34 pm
iceport wrote: Thu Sep 23, 2021 12:24 pm Claiming sometime in the middle years is a way of narrowing the dispersion of PV outcomes.
But why would this be a useful goal? It strikes me as akin to trying to narrow the dispersion of outcomes of just your international stock fund rather than of the whole portfolio.

If what you care about is bequest, then there's a directly offsetting item here (dollars spent, which increases as lifespan increases). By delaying, you narrow the dispersion of the net amount left to heirs.
My initial reaction is that it's just as interesting as the basic PV analysis is. It almost seems as though you are downplaying the value of performing the PV analysis in the first place.

As I see it, if PV is of any interest at all, then identifying and factoring in the full range of *possible* PV outcomes is every bit as interesting as identifying only the highest *expected* PV. Why go through the effort of performing the massive number-crunching of a PV analysis and then ignore much of the resulting data?

Let me ponder this more, though. I must surely be missing something...
In a situation in which portfolio depletion is not a significant possibility, if what we care about is bequest, maximizing expected PV also has the effect of maximizing expected bequest (on a before-tax basis).

An analogous thing cannot be said regarding distribution of expected PV (for various life expectancies, with claiming age held constant) vs distribution of expected bequest.
If only the expected PV at the expected age of death is considered, it usually results in claiming at 70, or very close to it, in order to maximize the expected PV.

However, for each possible age of claiming, there is a range of expected PVs associated with it, depending upon when death actually occurs. Using the same table I put up before of expected PVs for ages of death from 71 to 90 and claiming ages from 62 to 70, someone claiming at 70 could expect PVs ranging from $975,953 if they go at 90, down to a low of a paltry $46,283 if they go at 71.

On the other hand, the range of expected PVs for someone claiming at 67 goes from a high of $900,715 at 90, down to $145,982 at 71.

If maximizing expected PV is the objective, surely $145,982 is better than $46,283, right?

In fact, there is a range of ages of death that extends from maybe 71 up to 75 for which the expected PVs if claiming at 70 are substantially lower than the corresponding PVs if claiming at 67.

So why wouldn't it be reasonable for a person to conclude that the worst possible outcomes associated with claiming at 70 and dying early are just too poor for comfort? And that a reasonable alternative would be to claim at 67 instead, thereby forgoing the maximum expected PVs associated with a longer life in order to limit the worst of the deficits in PVs associated with a shorter life?

Reaching for the maximum expected PV also involves greater risk of very low PVs. Why is that of no interest?

There's a roughly 1 in 8 chance for a 62 YO single male to die between 71 and 75, so we're not talking about an especially rare event here.

In some respects, I see this as analogous to the classic risk/reward trade-off.

Please help me see the flaw(s) in this line of reasoning, assuming expected PV is being used as a metric of interest in the first place. (Disregard longevity risk, a separate matter.)

(I've got questions about how maximizing expected PV also maximizes expected bequest too, but I'd like to address this main question first.)

Imageuploading pictures
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

bobcat2 wrote: Thu Sep 23, 2021 3:39 pm Link to underlying paper - https://crr.bc.edu/working-papers/the-c ... -claiming/

BobK
Thanks BobK! I saw that in my inbox yesterday, too. Thanks for posting. And it is very interesting.

(Interesting collaborators on the project, also.)
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by vanbogle59 »

iceport wrote: Fri Sep 24, 2021 3:03 pm So why wouldn't it be reasonable for a person to conclude that the worst possible outcomes associated with claiming at 70 and dying early are just too poor for comfort?

There ain't no arguing what one person thinks is comfortable.
de gustibus non disputandum est
User avatar
ObliviousInvestor
Posts: 4212
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by ObliviousInvestor »

iceport wrote: Fri Sep 24, 2021 3:03 pm [...]

Reaching for the maximum expected PV also involves greater risk of very low PVs. Why is that of no interest?

[...]

Please help me see the flaw(s) in this line of reasoning, assuming expected PV is being used as a metric of interest in the first place. (Disregard longevity risk, a separate matter.)

(I've got questions about how maximizing expected PV also maximizes expected bequest too, but I'd like to address this main question first.)
We kind of have to start with this last part first. The reason that maximizing expected PV is of interest is because it's also what maximizes expected bequest, which is what we really care about (in cases in which longevity risk is not considered to be particularly important). We don't care about expected PV of Social Security itself. (At least I don't care about it. And I'm truly struggling to understand why anybody would.)

That is, in the context of maximizing expected PV it works because it functions as a good stand-in for a thing we care about. In the context of range of outcomes, it doesn't.
Mike Piper | Roth is a name, not an acronym. If you type ROTH, you're just yelling about retirement accounts.
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

ObliviousInvestor wrote: Fri Sep 24, 2021 3:12 pm
iceport wrote: Fri Sep 24, 2021 3:03 pm [...]

Reaching for the maximum expected PV also involves greater risk of very low PVs. Why is that of no interest?

[...]

Please help me see the flaw(s) in this line of reasoning, assuming expected PV is being used as a metric of interest in the first place. (Disregard longevity risk, a separate matter.)

(I've got questions about how maximizing expected PV also maximizes expected bequest too, but I'd like to address this main question first.)
We kind of have to start with this last part first. The reason that maximizing expected PV is of interest is because it's also what maximizes expected bequest, which is what we really care about (in cases in which longevity risk is not considered to be particularly important). We don't care about expected PV of Social Security itself. (At least I don't care about it. And I'm truly struggling to understand why anybody would.)

That is, in the context of maximizing expected PV it works because it functions as a good stand-in for a thing we care about. In the context of range of outcomes, it doesn't.
Well, I'm in complete agreement with you that maximizing expected PV is not a particularly relevant objective for its own sake. But I got clobbered for saying so up front, so I have been assuming it's important to people.

As far as it being a stand-in for maximum expected bequest, I'm confused about that, too.

When I run the claim at 62 and claim at 70 scenarios through cFIREsim, the highest portfolio values are associated with claiming early, not late.

What assumptions are being made that reverse that order?
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
User avatar
ObliviousInvestor
Posts: 4212
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by ObliviousInvestor »

iceport wrote: Fri Sep 24, 2021 3:27 pm
ObliviousInvestor wrote: Fri Sep 24, 2021 3:12 pm
iceport wrote: Fri Sep 24, 2021 3:03 pm [...]

Reaching for the maximum expected PV also involves greater risk of very low PVs. Why is that of no interest?

[...]

Please help me see the flaw(s) in this line of reasoning, assuming expected PV is being used as a metric of interest in the first place. (Disregard longevity risk, a separate matter.)

(I've got questions about how maximizing expected PV also maximizes expected bequest too, but I'd like to address this main question first.)
We kind of have to start with this last part first. The reason that maximizing expected PV is of interest is because it's also what maximizes expected bequest, which is what we really care about (in cases in which longevity risk is not considered to be particularly important). We don't care about expected PV of Social Security itself. (At least I don't care about it. And I'm truly struggling to understand why anybody would.)

That is, in the context of maximizing expected PV it works because it functions as a good stand-in for a thing we care about. In the context of range of outcomes, it doesn't.
Well, I'm in complete agreement with you that maximizing expected PV is not a particularly relevant objective for its own sake. But I got clobbered for saying so up front, so I have been assuming it's important to people.

As far as it being a stand-in for maximum expected bequest, I'm confused about that, too.

When I run the claim at 62 and claim at 70 scenarios through cFIREsim, the highest portfolio values are associated with claiming early, not late.

What assumptions are being made that reverse that order?
I have not used cFIREsim, so I don't know. I would guess that it's a high assumption as to expected returns (discount rate, in the context of a PV analysis).

Edited to add: my second guess would be differences in the mortality assumptions used.
Last edited by ObliviousInvestor on Fri Sep 24, 2021 3:36 pm, edited 1 time in total.
Mike Piper | Roth is a name, not an acronym. If you type ROTH, you're just yelling about retirement accounts.
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

vested1 wrote: Fri Sep 24, 2021 4:50 am I'm not trying to convince you to file early or to delay, just pointing out that the odds of living longer are better than you may be assuming they are. Playing the odds is what we all do to a certain degree, which is the motive for obtaining insurance in order to hedge that bet. In this case, the hedge is longevity insurance.
I'm not looking to argue with you on this, I just think you are making assumptions about what I am assuming. I'm using the SSA 2017 Period table for reference.

I came into this thread without any knowledge at all about the mortality probabilities at various ages. I know more now. And one of the things that became apparent is that the probability of what I consider a very long life is far higher than I had assumed. I know better now.

Similarly, now I have a better handle on the probability of dying relatively younger.

I consider it valuable information to know that there's a roughly 1 in 4 chance of a 62 YO dying from 62—74, and a roughly 1 in 5 chance from 91—110.
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

ObliviousInvestor wrote: Fri Sep 24, 2021 3:30 pm
iceport wrote: Fri Sep 24, 2021 3:27 pm
ObliviousInvestor wrote: Fri Sep 24, 2021 3:12 pm
iceport wrote: Fri Sep 24, 2021 3:03 pm [...]

Reaching for the maximum expected PV also involves greater risk of very low PVs. Why is that of no interest?

[...]

Please help me see the flaw(s) in this line of reasoning, assuming expected PV is being used as a metric of interest in the first place. (Disregard longevity risk, a separate matter.)

(I've got questions about how maximizing expected PV also maximizes expected bequest too, but I'd like to address this main question first.)
We kind of have to start with this last part first. The reason that maximizing expected PV is of interest is because it's also what maximizes expected bequest, which is what we really care about (in cases in which longevity risk is not considered to be particularly important). We don't care about expected PV of Social Security itself. (At least I don't care about it. And I'm truly struggling to understand why anybody would.)

That is, in the context of maximizing expected PV it works because it functions as a good stand-in for a thing we care about. In the context of range of outcomes, it doesn't.
Well, I'm in complete agreement with you that maximizing expected PV is not a particularly relevant objective for its own sake. But I got clobbered for saying so up front, so I have been assuming it's important to people.

As far as it being a stand-in for maximum expected bequest, I'm confused about that, too.

When I run the claim at 62 and claim at 70 scenarios through cFIREsim, the highest portfolio values are associated with claiming early, not late.

What assumptions are being made that reverse that order?
I have not used cFIREsim, so I don't know. I would guess that it's a high assumption as to expected returns (discount rate, in the context of a PV analysis).

Edited to add: my second guess would be differences in the mortality assumptions used.
That could very well be. But it also seems realistic, because whatever isn't spent will remain in the portfolio with a fixed AA. And claiming early reduces what needs to be spent from the portfolio up front for a given spending level.

On the mortality, I think I've run it over 30 years (to age 89) and over 26 years (to age 85), and found similar results.
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
User avatar
ObliviousInvestor
Posts: 4212
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by ObliviousInvestor »

iceport wrote: Fri Sep 24, 2021 3:41 pm
ObliviousInvestor wrote: Fri Sep 24, 2021 3:30 pm
iceport wrote: Fri Sep 24, 2021 3:27 pm
ObliviousInvestor wrote: Fri Sep 24, 2021 3:12 pm
iceport wrote: Fri Sep 24, 2021 3:03 pm [...]

Reaching for the maximum expected PV also involves greater risk of very low PVs. Why is that of no interest?

[...]

Please help me see the flaw(s) in this line of reasoning, assuming expected PV is being used as a metric of interest in the first place. (Disregard longevity risk, a separate matter.)

(I've got questions about how maximizing expected PV also maximizes expected bequest too, but I'd like to address this main question first.)
We kind of have to start with this last part first. The reason that maximizing expected PV is of interest is because it's also what maximizes expected bequest, which is what we really care about (in cases in which longevity risk is not considered to be particularly important). We don't care about expected PV of Social Security itself. (At least I don't care about it. And I'm truly struggling to understand why anybody would.)

That is, in the context of maximizing expected PV it works because it functions as a good stand-in for a thing we care about. In the context of range of outcomes, it doesn't.
Well, I'm in complete agreement with you that maximizing expected PV is not a particularly relevant objective for its own sake. But I got clobbered for saying so up front, so I have been assuming it's important to people.

As far as it being a stand-in for maximum expected bequest, I'm confused about that, too.

When I run the claim at 62 and claim at 70 scenarios through cFIREsim, the highest portfolio values are associated with claiming early, not late.

What assumptions are being made that reverse that order?
I have not used cFIREsim, so I don't know. I would guess that it's a high assumption as to expected returns (discount rate, in the context of a PV analysis).

Edited to add: my second guess would be differences in the mortality assumptions used.
That could very well be. But it also seems realistic, because whatever isn't spent will remain in the portfolio with a fixed AA. And claiming early reduces what needs to be spent from the portfolio up front for a given spending level.

On the mortality, I think I've run it over 30 years (to age 89) and over 26 years (to age 85), and found similar results.
You can use whatever discount rate/ROI assumptions you want. But as long as they (and other assumptions) are consistent between the two calculators, the option that maximizes expected PV from Social Security should also maximize expected bequest.
Mike Piper | Roth is a name, not an acronym. If you type ROTH, you're just yelling about retirement accounts.
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

ObliviousInvestor wrote: Fri Sep 24, 2021 3:54 pm You can use whatever discount rate/ROI assumptions you want. But as long as they (and other assumptions) are consistent between the two calculators, the option that maximizes expected PV from Social Security should also maximize expected bequest.
cFIREsim, like FIRECalc, uses historic market and economic data. I don't impose any assumptions or limitations over the historic data.

Are ending portfolio balances the appropriate metric for identifying maximum bequest?
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
User avatar
ObliviousInvestor
Posts: 4212
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by ObliviousInvestor »

iceport wrote: Fri Sep 24, 2021 4:05 pm
ObliviousInvestor wrote: Fri Sep 24, 2021 3:54 pm You can use whatever discount rate/ROI assumptions you want. But as long as they (and other assumptions) are consistent between the two calculators, the option that maximizes expected PV from Social Security should also maximize expected bequest.
cFIREsim, like FIRECalc, uses historic market and economic data. I don't impose any assumptions or limitations over the historic data.

Are ending portfolio balances the appropriate metric for identifying maximum bequest?
Yes, ending portfolio balance = bequest, if you are projecting out to a particular age at death. And if you use the same rate of return assumption and mortality assumption, the strategy that maximizes in one calculator should maximize in the other calculator as well.

To back up a step though, you don't need to run any calculators to understand this concept. The more dollars you receive from Social Security over your lifetime (after accounting for the fact that dollars received earlier can be invested sooner), the greater the bequest. (Again, this is all in pre-tax terms.)
Mike Piper | Roth is a name, not an acronym. If you type ROTH, you're just yelling about retirement accounts.
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

ObliviousInvestor wrote:Fri Sep 24, 2021 4:14 pm
iceport wrote: Fri Sep 24, 2021 4:05 pm
ObliviousInvestor wrote: Fri Sep 24, 2021 3:54 pm You can use whatever discount rate/ROI assumptions you want. But as long as they (and other assumptions) are consistent between the two calculators, the option that maximizes expected PV from Social Security should also maximize expected bequest.
cFIREsim, like FIRECalc, uses historic market and economic data. I don't impose any assumptions or limitations over the historic data.

Are ending portfolio balances the appropriate metric for identifying maximum bequest?
Yes, ending portfolio balance = bequest, if you are projecting out to a particular age at death. And if you use the same rate of return assumption and mortality assumption, the strategy that maximizes in one calculator should maximize in the other calculator as well.
Well, retirement already started at 55. So there will be 8 more years of larger portfolio withdrawals if claiming at 70 vs. 62, and those larger withdrawals come in the early years. So there is possibly some SORR showing up in the results?

This is what I found in one comparison run:
27 year horizon, claim SS @62 vs. claim SS @ 70:

1.81% more annual spending capacity w/SS @70 than @62
6.0% larger average ending portfolio balance w/SS @62 than @70
11.8% larger median ending portfolio balance w/SS @62 than @70
(smaller numbers in all denominators)

.
ObliviousInvestor wrote:Fri Sep 24, 2021 4:14 pm To back up a step though, you don't need to run any calculators to understand this concept. The more dollars you receive from Social Security over your lifetime (after accounting for the fact that dollars received earlier can be invested sooner), the greater the bequest. (Again, this is all in pre-tax terms.)
Well if that's the assumption, then the first question of the day has not yet been resolved. ("...surely $145,982 is better than $46,283, right?") When I asked that question, you said PV was not an important metric. But here you say clearly: greater PV = better outcome.

So why no concern whatsoever about the risk of much lower PVs? It's still not making sense...




Thank you, Mike. I am greatly appreciative of your input. But please don't let me monopolize any more of your time...
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
wrongfunds
Posts: 3187
Joined: Tue Dec 21, 2010 2:55 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by wrongfunds »

iceport wrote: Fri Sep 24, 2021 3:08 pm
bobcat2 wrote: Thu Sep 23, 2021 3:39 pm Link to underlying paper - https://crr.bc.edu/working-papers/the-c ... -claiming/

BobK
Thanks BobK! I saw that in my inbox yesterday, too. Thanks for posting. And it is very interesting.

(Interesting collaborators on the project, also.)
Why did you say that? The names did NOT sound familiar to me. At least those were NOT the usual suspects in this field. also I thought BU is the retirement research capital and NOT BC.
grok87
Posts: 10512
Joined: Tue Feb 27, 2007 8:00 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by grok87 »

wrongfunds wrote: Sat Sep 25, 2021 8:34 am
iceport wrote: Fri Sep 24, 2021 3:08 pm
bobcat2 wrote: Thu Sep 23, 2021 3:39 pm Link to underlying paper - https://crr.bc.edu/working-papers/the-c ... -claiming/

BobK
Thanks BobK! I saw that in my inbox yesterday, too. Thanks for posting. And it is very interesting.

(Interesting collaborators on the project, also.)
Why did you say that? The names did NOT sound familiar to me. At least those were NOT the usual suspects in this field. also I thought BU is the retirement research capital and NOT BC.
interesting about BU- do you have a iink to their research? The BC folks have a blog
https://squaredawayblog.bc.edu/

cheers,
grok
RIP Mr. Bogle.
bsteiner
Posts: 9208
Joined: Sat Oct 20, 2012 9:39 pm
Location: NYC/NJ/FL

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by bsteiner »

It depends on what interest rate you use to compare the choices. Do you discount the expected future payments to present value at a low interest rate because they're safe and they're indexed for inflation? Or do you use a higher rate because they're not not marketable (you can't sell your right to the future payments or borrow against them)?
JackoC
Posts: 4714
Joined: Sun Aug 12, 2018 11:14 am

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by JackoC »

Tom_T wrote: Fri Sep 24, 2021 6:11 am Life expectancy is generally expressed as an average, not a median. The life expectancy for a newborn U.S. male is 76, but I will bet that the median is over 80. The SSA tables are averages. If the SSA table says that a 65-year-old male has a life expectancy of 18 years, that doesn't mean that half the 65-year-olds will die before then. It will be less than half.
True, but the distribution of ages of death at birth has a big long left hand tail of people who die decades before 'normal' age of death, so the median and mean are around 4 years different (median age of death, at birth, has been around 80 recently). Once and if you reach your 60's the distribution of remaining years is much less skewed, mean and median are more like 1-1.5 years different.

IOW once you're close to actually deciding to take SS at 62 or 70, the question of whether your own LE prospects are well represented by the SSA table is a much bigger one than median v mean. The upside of SSA life table: you can delve into quite specific details but a personalized calculator elsewhere will typically only give you mean (or sometimes median, 50%-tile, one I used gave a 75%-tile also). But just above somebody said they got 93 at age 62 on a John Hancock personalized calculator, mine generally come out 93-94 for male 64 on various online calculators, different than SSA by much more than median v mean, and renders it highly questionable to use SSA just because it has more detail.

To beat another dead horse, the high LE estimate doesn't mean I won't get a terminal diagnosis next time I don't feel well or won't get hit by a bus tomorrow. But it also doesn't mean there's no such thing as a distribution more suitable to people in your general situation. Would anyone use a life table for all people on earth to plan their retirement? No, because it includes a large % of people on earth in quite different situations than people in the US generally. Same goes for table for a whole large, highly diverse country like the US, plus just individual history, SSA table includes smokers, already very sick people, etc. vs. one's own situation. But there will still be millions of people in one's own general situation, many times more than enough to construct a meaningful table, as insurance companies do every day. Delving into the distribution of SSA numbers is easy via their table, but false precision unless you can determine from personalized calculators that your expectancy falls pretty close to SSA's expectancy.
User avatar
sapphire96
Posts: 193
Joined: Fri Jun 16, 2017 8:08 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by sapphire96 »

It is interesting to me how this is a “take at 62 or 70” argument… why not take at 67 and split the difference, best of both worlds?
Keep interest as your friend, not your foe. | Use money as a tool for bettering your life, not squandering it. | Stay the course, don’t deviate from it.
jello_nailer
Posts: 571
Joined: Sun Apr 07, 2019 10:20 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by jello_nailer »

[/quote]

Really? I'm surprised you need the money. I would have thought you saved enough money so you could wait until you're 70 to maximize your longevity insurance. :mrgreen:
[/quote] No. I waited to only 69 years, 8 months. I'd just appreciate the loose change from iceport.
BobK
[/quote]

Interesting. BobK - What was your thought process to pick that age? I would think you could have eaten fish heads and rice for 4 months if you were short income and were struggling to make it to 70 and receive the full load max. I'm asking because I'm open to understanding other perspectives.
Thanks
User avatar
bobcat2
Posts: 6076
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by bobcat2 »

jello_nailer wrote: Sat Sep 25, 2021 10:55 am

Really? I'm surprised you need the money. I would have thought you saved enough money so you could wait until you're 70 to maximize your longevity insurance. :mrgreen:
No. I waited to only 69 years, 8 months. I'd just appreciate the loose change from iceport.
Interesting. BobK - What was your thought process to pick that age? I would think you could have eaten fish heads and rice for 4 months if you were short income and were struggling to make it to 70 and receive the full load max. I'm asking because I'm open to understanding other perspectives.
Thanks
Large and unexpected home repairs were suddenly needed. I am not a fanatic about rules - there also needs to be good sense applied. :happy

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
wrongfunds
Posts: 3187
Joined: Tue Dec 21, 2010 2:55 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by wrongfunds »

grok87 wrote: Sat Sep 25, 2021 8:39 am
wrongfunds wrote: Sat Sep 25, 2021 8:34 am
iceport wrote: Fri Sep 24, 2021 3:08 pm
bobcat2 wrote: Thu Sep 23, 2021 3:39 pm Link to underlying paper - https://crr.bc.edu/working-papers/the-c ... -claiming/

BobK
Thanks BobK! I saw that in my inbox yesterday, too. Thanks for posting. And it is very interesting.

(Interesting collaborators on the project, also.)
Why did you say that? The names did NOT sound familiar to me. At least those were NOT the usual suspects in this field. also I thought BU is the retirement research capital and NOT BC.
interesting about BU- do you have a iink to their research? The BC folks have a blog
https://squaredawayblog.bc.edu/

cheers,
grok
May be I am the one who is mistaken? I am trying figure out the big names from BU but now my brain is going blank! Name Larry Kxxxx sounds vaguely familiar.
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

wrongfunds wrote: Sat Sep 25, 2021 12:11 pm
grok87 wrote: Sat Sep 25, 2021 8:39 am
wrongfunds wrote: Sat Sep 25, 2021 8:34 am
iceport wrote: Fri Sep 24, 2021 3:08 pm
bobcat2 wrote: Thu Sep 23, 2021 3:39 pm Link to underlying paper - https://crr.bc.edu/working-papers/the-c ... -claiming/

BobK
Thanks BobK! I saw that in my inbox yesterday, too. Thanks for posting. And it is very interesting.

(Interesting collaborators on the project, also.)
Why did you say that? The names did NOT sound familiar to me. At least those were NOT the usual suspects in this field. also I thought BU is the retirement research capital and NOT BC.
interesting about BU- do you have a iink to their research? The BC folks have a blog
https://squaredawayblog.bc.edu/

cheers,
grok
May be I am the one who is mistaken? I am trying figure out the big names from BU but now my brain is going blank! Name Larry Kxxxx sounds vaguely familiar.
wrongfunds,

I had heard Alicia Munnell's name before, in radio interviews and quotes in random articles. Then in 2015 our governor commissioned Munnell and the folks at the Boston College CRR to perform a study on our state pension system. That's when I looked into her background and found a she has rich career experience, including a long stint at the Federal Reserve in Boston, and other high profile appointments and recognitions. (The study ended up being an *outstanding* resource that included a forensic analysis of what went wrong and several creative ideas on how to address funding shortfalls. Though for various reasons most sides in the debate elected to distance themselves from the study recommendations, in reality they formed the backbone of the reforms ultimately enacted.) In terms of public policy, she tends towards the middle, often advocating a mix of benefit reductions and revenue raising, but she has a steadfast interest in protecting financial safety nets for seniors. I don't know a whole lot about Andrew Biggs other than what I read in the press, including lots of Op-Eds by Biggs. While my impression of him is that he is very well-respected and knowledgeable, and that he bases his work in valid research and facts, he comes across to me as an almost militant advocate for cutting taxes and slashing public benefits — of all kinds. I just found it interesting that Munnell and Biggs would end up collaborating on anything. But perhaps that fact that they did is an indication of intellectual honesty in both of them.
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

JackoC wrote: Sat Sep 25, 2021 10:21 am
Tom_T wrote: Fri Sep 24, 2021 6:11 am Life expectancy is generally expressed as an average, not a median. The life expectancy for a newborn U.S. male is 76, but I will bet that the median is over 80. The SSA tables are averages. If the SSA table says that a 65-year-old male has a life expectancy of 18 years, that doesn't mean that half the 65-year-olds will die before then. It will be less than half.
True, but the distribution of ages of death at birth has a big long left hand tail of people who die decades before 'normal' age of death, so the median and mean are around 4 years different (median age of death, at birth, has been around 80 recently). Once and if you reach your 60's the distribution of remaining years is much less skewed, mean and median are more like 1-1.5 years different.

IOW once you're close to actually deciding to take SS at 62 or 70, the question of whether your own LE prospects are well represented by the SSA table is a much bigger one than median v mean. The upside of SSA life table: you can delve into quite specific details but a personalized calculator elsewhere will typically only give you mean (or sometimes median, 50%-tile, one I used gave a 75%-tile also). But just above somebody said they got 93 at age 62 on a John Hancock personalized calculator, mine generally come out 93-94 for male 64 on various online calculators, different than SSA by much more than median v mean, and renders it highly questionable to use SSA just because it has more detail.

To beat another dead horse, the high LE estimate doesn't mean I won't get a terminal diagnosis next time I don't feel well or won't get hit by a bus tomorrow. But it also doesn't mean there's no such thing as a distribution more suitable to people in your general situation. Would anyone use a life table for all people on earth to plan their retirement? No, because it includes a large % of people on earth in quite different situations than people in the US generally. Same goes for table for a whole large, highly diverse country like the US, plus just individual history, SSA table includes smokers, already very sick people, etc. vs. one's own situation. But there will still be millions of people in one's own general situation, many times more than enough to construct a meaningful table, as insurance companies do every day. Delving into the distribution of SSA numbers is easy via their table, but false precision unless you can determine from personalized calculators that your expectancy falls pretty close to SSA's expectancy.
Thank you for this explanation. It's very helpful.
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
Tom_T
Posts: 4836
Joined: Wed Aug 29, 2007 2:33 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by Tom_T »

iceport wrote: Sat Sep 25, 2021 12:54 pm
JackoC wrote: Sat Sep 25, 2021 10:21 am
Tom_T wrote: Fri Sep 24, 2021 6:11 am Life expectancy is generally expressed as an average, not a median. The life expectancy for a newborn U.S. male is 76, but I will bet that the median is over 80. The SSA tables are averages. If the SSA table says that a 65-year-old male has a life expectancy of 18 years, that doesn't mean that half the 65-year-olds will die before then. It will be less than half.
True, but the distribution of ages of death at birth has a big long left hand tail of people who die decades before 'normal' age of death, so the median and mean are around 4 years different (median age of death, at birth, has been around 80 recently). Once and if you reach your 60's the distribution of remaining years is much less skewed, mean and median are more like 1-1.5 years different.

IOW once you're close to actually deciding to take SS at 62 or 70, the question of whether your own LE prospects are well represented by the SSA table is a much bigger one than median v mean. The upside of SSA life table: you can delve into quite specific details but a personalized calculator elsewhere will typically only give you mean (or sometimes median, 50%-tile, one I used gave a 75%-tile also). But just above somebody said they got 93 at age 62 on a John Hancock personalized calculator, mine generally come out 93-94 for male 64 on various online calculators, different than SSA by much more than median v mean, and renders it highly questionable to use SSA just because it has more detail.

To beat another dead horse, the high LE estimate doesn't mean I won't get a terminal diagnosis next time I don't feel well or won't get hit by a bus tomorrow. But it also doesn't mean there's no such thing as a distribution more suitable to people in your general situation. Would anyone use a life table for all people on earth to plan their retirement? No, because it includes a large % of people on earth in quite different situations than people in the US generally. Same goes for table for a whole large, highly diverse country like the US, plus just individual history, SSA table includes smokers, already very sick people, etc. vs. one's own situation. But there will still be millions of people in one's own general situation, many times more than enough to construct a meaningful table, as insurance companies do every day. Delving into the distribution of SSA numbers is easy via their table, but false precision unless you can determine from personalized calculators that your expectancy falls pretty close to SSA's expectancy.
Thank you for this explanation. It's very helpful.
Yes, +1.

As someone just said, this doesn't have to be a 62-vs.-70 debate. I can claim in any month from 62 all the way to 70. There are a lot of points in between! My guess is that I'll play it by ear. I am nearly 63 and my health is A-plus so far, so I don't feel like I'm under the gun to start collecting "my" money. Let's say I reach 66, and my health is great, and my withdrawals are covering my expenses with no problems. I might think "let's go another year and see how things are."
User avatar
Topic Author
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by iceport »

bsteiner wrote: Sat Sep 25, 2021 9:59 am It depends on what interest rate you use to compare the choices. Do you discount the expected future payments to present value at a low interest rate because they're safe and they're indexed for inflation? Or do you use a higher rate because they're not not marketable (you can't sell your right to the future payments or borrow against them)?
We know what discount rate is being used in the Open Social Security calculations, and there is a logical reason for it.

However, when I run my combination of asset base and income streams through an historical database analysis like cFIREsim, I'm assuming the "discount rate" used in the analysis ends up being whatever the combined effects of inflation and market returns happened to actually produce, in real life, in the sequences analyzed.

So yes, the difference in assumed discount rates could be large.

But is there any reason to alter the analysis that uses historic data? The major appeal of that kind of analysis is that it is generally more realistic than many others. I think it models the circumstances appropriately. Is that a bad assumption?

Intuitively, it does make sense that larger early withdrawals from a portfolio needed with claiming SS late produces smaller ending portfolio balances. Mike's analysis apparently reaches the opposite conclusion.

:?
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
User avatar
ObliviousInvestor
Posts: 4212
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: I’m Planning to Claim SS @62… Well, Why Not?

Post by ObliviousInvestor »

iceport wrote: Sat Sep 25, 2021 1:05 pm
bsteiner wrote: Sat Sep 25, 2021 9:59 am It depends on what interest rate you use to compare the choices. Do you discount the expected future payments to present value at a low interest rate because they're safe and they're indexed for inflation? Or do you use a higher rate because they're not not marketable (you can't sell your right to the future payments or borrow against them)?
We know what discount rate is being used in the Open Social Security calculations, and there is a logical reason for it.

However, when I run my combination of asset base and income streams through an historical database analysis like cFIREsim, I'm assuming the "discount rate" used in the analysis ends up being whatever the combined effects of inflation and market returns happened to actually produce, in real life, in the sequences analyzed.

So yes, the difference in assumed discount rates could be large.

But is there any reason to alter the analysis that uses historic data? The major appeal of that kind of analysis is that it is generally more realistic than many others. I think it models the circumstances appropriately. Is that a bad assumption?

Intuitively, it does make sense that larger early withdrawals from a portfolio needed with claiming SS late produces smaller ending portfolio balances. Mike's analysis apparently reaches the opposite conclusion.

:?
Open Social Security is assuming that you are spending down the safest part of your portfolio in order to delay Social Security (if you delay Social Security), because that is what in most cases it makes sense to do. This discussion has more on that:
viewtopic.php?p=5362672#p5362672

Hence the selection of the yield on 20-year TIPS as the default discount rate. Using stock returns only makes sense if you're planning to spend down stocks, which you generally shouldn't unless you're 100%-stock already. I will leave the question of historical returns (for stocks and/or bonds) vs something based on current circumstances to others. There are countless threads on those topics already. (Though of course there are also countless threads on when to file for Social Security, yet here we are participating in another...)
Mike Piper | Roth is a name, not an acronym. If you type ROTH, you're just yelling about retirement accounts.
Post Reply