Surprising "when should we take SS" exercise outcome

[Lisa99]

Using Firecalc I just estimated our 'level of spending' with a 95% success rate using SS benefits starting at 62, 67 and 70.

I'm shocked to learn that our highest level of spending ability will occur if we both take SS at 62. There is a slight drop in spending level if we wait to 65 and 67, but an overall 5% drop in spending ability if we wait until 70.

I would have thought it would have been the opposite, but I will have a very small pension and DH doesn't have one so I guess spending from the portfolio only for too many years really takes a toll.

Wahoo for Firecalc!

[Gumby]

Very clever idea. I hadn't thought to do it that way, but it makes sense.

In our case, the decision is driven by the Government Pension Offset. Given that my young wife will not be eligible for SS on her own, and her survivor benefit would be eliminated by the GPO, it makes sense for me to take SS at 62 to minimize draws on the nest egg, since I will in all likelihood predecease her.

[Ken11]

Firecalc has you input a given time horizon. So, the results do not account for residual longevity risk. IF you live longer than than the number of years inpout to Firecalc then the higher SS payments resulting from a later start to SS could change your conslusions.

Cheers

[Brett_Cameron]

in my many projection models, I had some interim results similar to yours. In my case I have a non-cola pension, SS, dividend income and 401k to model as income sources. After I projected expenses with associated life changes and escalation and figured out how to estimate future income taxes (with escalation of brackets) , my results changed. If I wait until 70 to draw SS, then I will not need anything other than SS and pension to exceed expenses by a wide margin, however, 401k balance will be much lower at age 86. I can maintain the same spending profile (year by year discrete sums rather than % SWR) and leave a much larger 401k balance to DW who is 23 years younger than me.

In my current projection model, maximum spend rate is by taking SS at 70 and minimum is taking SS at 62. I use FIRECALC to input my planned annual withdrawals to see what the historical failure rates would have been. So far age 62 has a 98% success rate, age 66-2 has an 84% success rate and age 70 has a 72% success rate.

My plan is to monitor things and change as needed based on market returns, inflation rate and expense changes. For example, I will recast the projections annually to help me decide when to take SS, if and when to buy an annuity with a portion of the 401k and when and how to change expenses. YMMV.

[ERD50]

Quote:

Originally Posted by Ken11 Firecalc has you input a given time horizon. So, the results do not account for residual longevity risk. IF you live longer than than the number of years inpout to Firecalc then the higher SS payments resulting from a later start to SS could change your conslusions. Cheers

Agreed - Lisa99, can you provide your other inputs?

Actually, after thinking a bit (and I'm just sipping my first cup, so I'm slower than normal), the results may not be surprising at all. When you delay you are essentially buying 'longevity insurance'. There is a price to pay for that, and you pay it early (with lost income & higher WDR), but if you live beyond your actuarial life expectancy, you have a larger income stream for those late years. A 30 year period would not have much 'recovery time'.

On top of that, remember that the failures in FIRECALC are usually because it throws the worst years right at the beginning which depletes your portfolio before it has any chance for growth. So delaying SS or pensions amplifies this. It won't be as bad in most cases, and you have some time to react, since it happens early rather than late.

I'd expect the moral of the story to be: Plan to delay if longevity is a likely concern, but take it early if you hit a trough right away. You might miss a few years SS income while waiting to decide if a downturn is a dip or a DIP, but it should help.

-ERD50

[MasterBlaster]

You should try the same analysis with the ORP calculator. When you include taxes, the picture changes and the outcome may be very different.

you can find the ORP calculator here:

Optimal Retirement Calculator and Retirement Decision Support System

If you do run the analysis on ORP, Please post the results here for discussion.

[rescueme]

Quote:

Originally Posted by Lisa99 I'm shocked to learn that our highest level of spending ability will occur if we both take SS at 62.

Did you run a calc based upon a "split SS" decison?

DW will be taking her SS in two years at FRA. I'll be claiming 50% of her benefit, while delaying my SS till age 70 (we're the same age).

Calculations showed (using Fidelity's RIP program) that this works out best, for out situation when measuing our age 62/66/70 options.

What seems to be a slam-dunk as indivudials is greatly changed if you take advanage of the combined/marriage options...

I'm not saying that you/spouse should not take SS at age 62; however depending on age, earnings, and SS credits (along with SS options and taxes - especially related to your investments), it could make an impact on your decision.

Not all circumstances are the same, and there is no "perfect rule", IMHO.

[Midpack]

I thought this was another good tool, just a different POV Social Security calculator: retirement options for you and spouse., someone here brought it to our attention a few weeks ago.

[ejman]

Quote:

Originally Posted by Lisa99 Using Firecalc I just estimated our 'level of spending' with a 95% success rate using SS benefits starting at 62, 67 and 70. I'm shocked to learn that our highest level of spending ability will occur if we both take SS at 62. There is a slight drop in spending level if we wait to 65 and 67, but an overall 5% drop in spending ability if we wait until 70. I would have thought it would have been the opposite, but I will have a very small pension and DH doesn't have one so I guess spending from the portfolio only for too many years really takes a toll. Wahoo for Firecalc!

That is a clever way of using Firecalc - thank you for pointing it out. Using my personal data confirms your findings. I had already decided to start my SS at 62 (wife already started SS 2 years ago) based on other factors but this confirmation from Firecalc certainly helps.

[W2R]

Quote:

Originally Posted by ejman That is a clever way of using Firecalc - thank you for pointing it out. Using my personal data confirms your findings. I had already decided to start my SS at 62 (wife already started SS 2 years ago) based on other factors but this confirmation from Firecalc certainly helps.

FIRECalc tells me I can spend $2K/year more if I claim SS at age 70, than I can if I claim it this summer at age 64. Although I don't plan to provide my input figures on this thread, I wonder if perhaps the assumed investment plan and yield might have something to do with it.

On the other hand, $2K/year is only $167/month, so not a huge difference....

[rodi]

We've been running the calcs for split outcomes also. I'm younger than my husband. We have minor children. If he starts collecting at age 62, the kids can get a benefit similar to the spousal benefit that rescueme mentioned. If he waits to full retirement (age 66), there are fewer years getting this kid benefit. So the math strongly favors him taking SS at 62. The math is a bit muddier for whether I take it at 62, 67 (my FRA), or 70. Need to run a few more scenarios.

It's definitely worth plugging in the numbers for the auxiliary recipients to see if that changes the dates of when to collect.
Social Security Publications

[ejman]

Quote:

Originally Posted by W2R FIRECalc tells me I can spend $2K/year more if I claim SS at age 70, than I can if I claim it this summer at age 64. Although I don't plan to provide my input figures on this thread, I wonder if perhaps the assumed investment plan and yield might have something to do with it.

I'm not sure what all of the drivers are but I noticed in my case the plan duration seemed to have a significant impact. For example if I assume I will have a plan duration of 40 years (I wish!) I get a result with a minimal advantage to the 70 year SS start of about 2k a year. If I assume a plan duration of 30 years or less then the advantage goes to the start 62 side.

This is an interesting theoretical exercise but I feel that for myself the actual relevance of such minimal $ differences so many years in the future is rather limited. My crystal ball totally fails to cooperate when asked questions about longevity, future state of the country and world, taxes etc etc for the next year let alone the next 30 or 40 years.

[GrayHare]

Quote:

Originally Posted by ERD50 I'd expect the moral of the story to be: Plan to delay if longevity is a likely concern, but take it early if you hit a trough right away. You might miss a few years SS income while waiting to decide if a downturn is a dip or a DIP, but it should help.

IMO that's the most important factor. In all actuarial matters, payouts are keyed to the average expected lifespan. So, if an individual has longevity genes, a delayed start with higher later payouts is usually the best choice. Starting receipt of payouts sooner is usually best if your genes and/or health suggest you will have a shorter-than-average lifespan.

[W2R]

Quote:

Originally Posted by ejman I'm not sure what all of the drivers are but I noticed in my case the plan duration seemed to have a significant impact. For example if I assume I will have a plan duration of 40 years (I wish!) I get a result with a minimal advantage to the 70 year SS start of about 2k a year. If I assume a plan duration of 30 years or less then the advantage goes to the start 62 side.

That makes sense. I did run my plan to age 95, so that was probably the source of some of this discrepancy.

Quote:

Originally Posted by ejman This is an interesting theoretical exercise but I feel that for myself the actual relevance of such minimal $ differences so many years in the future is rather limited. My crystal ball totally fails to cooperate when asked questions about longevity, future state of the country and world, taxes etc etc for the next year let alone the next 30 or 40 years.

You can say that again, and it makes the decision of when to claim SS even more difficult to resolve than I imagined.

[Lisa99]

Quote:

Originally Posted by ejman I'm not sure what all of the drivers are but I noticed in my case the plan duration seemed to have a significant impact. For example if I assume I will have a plan duration of 40 years (I wish!) I get a result with a minimal advantage to the 70 year SS start of about 2k a year. If I assume a plan duration of 30 years or less then the advantage goes to the start 62 side.

All of our scenarios are run w/a 40 year duration and leaving a $200k estate. 40 years will put me at 95, which I don't think I'll outlast, but the $200k estate is the cushion.

I don't bother to run 30 year scenarios since that puts me at 85 and several close relatives have outlived that age.

[Chuckanut]

Just ran FireCalc with a 40 year scenario. The only difference is taking SS at 62 or at 66. The success rate for age 62 was 97%, for age 66 100%. Moving SS to age 70 does not increase the success rate past 100% (interesting if it did!), but my average balance does increase by about 200,000 dollars.

[Independent]

I just tried this with the default FireCalc $750k portfolio.

Instead of a spending target of $30k I entered $60k. Then I entered at $30k SS benefit starting in year 2013. That gave me 8 failures over a 30 year horizon.

For the alternate run, I entered $52,800 SS benefit starting in year 2021. With the same $60k spending, I got 4 failures.

Some years ago we debated this. IIRC, the person who found that taking SS sooner was the better option was running Firecalc with a significant number of years of earning/saving before 62.

[youbet]

Quote:

Originally Posted by Lisa99 I don't bother to run 30 year scenarios since that puts me at 85 and several close relatives have outlived that age.

Try doing both and noting both results. The 30 year test uses additional data not included in the 40 year test and is sometimes worthwhile to check.

For example, the run I do which tests my current situation has a 100% success rate with either 30 or 40 years. But, the 30 year scenario shows lower early year portfolio balances than the 40 year scenario.

Just another way to get yourself confused I guess. But do try both. And note that the end result on a 30 year run is different than the 30 year point on a 40 year run, all other things being equal.

[Lisa99]

Quote:

Originally Posted by Independent IIRC, the person who found that taking SS sooner was the better option was running Firecalc with a significant number of years of earning/saving before 62.

DH and I both fit that description. We've hit the max SS contributions every year for the last 10 years and still have 4 years to work, so our SS payments will meet more than half of our 'income' requirements even if we begin payments at 62.

And thanks for the tip, YouBet. I didn't know the 30 year test used addtl data.

[GSMAN]

Quote:

Originally Posted by Midpack I thought this was another good tool, just a different POV Social Security calculator: retirement options for you and spouse., someone here brought it to our attention a few weeks ago.

This tool confirms my analysis that I put together last year when doing some planning. If you look at where the three lines intersect, the red line at 62 is always above the other lines until you reach 86 or 87. After that, you see that the line starting SS at 70 will exceed the cumulative benefit value of the other two start dates. That "break even" point was my determing factor on when I plan to take SS. Now, seeing how I am not too confident that I will live much beyond 86 or 87, my folks are still alive in their 70's and grandparents died in their 80s, These odds seem about right for me. If I were to take a dirt nap earlier than 86 or 87 then I would have recieved a higher pay out starting at 62 than the other two dates.