Open Social Security strategy results question

Have a question about your personal investments? No matter how simple or complex, you can ask it here.
Topic Author
markcoop
Posts: 1159
Joined: Fri Mar 02, 2007 8:36 am

Open Social Security strategy results question

Post by markcoop » Sat May 16, 2020 11:10 am

I was playing around with the Open Social Security calculator to understand my optimal strategy. I have made a decent amount more than my wife. The optimal strategy was for her was to start taking her benefits early (Age 62 and 7 months) and for me to wait till as late as possible (age 70). I understand conceptually why that makes sense given the spousal benefit bump my wife would receive when I start taking withdrawals at the highest payout for me. What I don't understand is under what circumstances it concludes she should start her early withdrawals at age 62 and 7 months and not age 62. When I plugged in my alternative strategy (starting her benefits at age 62), the present value difference was really small - $235 worse in my alternative. So, really not a big deal either way, but just wanted to understand why it would be the case. My only thought was that the rate of change in SS benefits is not linear going from age 62 to other ages and at 62 and 7 months there is a small spike??
Mark

Silk McCue
Posts: 4712
Joined: Thu Feb 25, 2016 7:11 pm

Re: Open Social Security strategy results question

Post by Silk McCue » Sat May 16, 2020 12:36 pm

This might help. It’s just math - note the bolded words below.

https://opensocialsecurity.com/about
Open Social Security is a free, open-source Social Security strategy calculator.

The calculator runs the math for each possible claiming age (or, if you're married, each possible combination of claiming ages) and reports back, telling you which strategy is expected to provide the most total spendable dollars over your lifetime.

Please note that this calculator should not be the only analysis you do, as there are various factors that it does not consider, such as:

The fact that delaying benefits reduces longevity risk and therefore may be preferable even in some cases in which it is not the strategy that maximizes expected total spending, or
Tax planning reasons or other unrelated reasons why it might be better for you to file earlier or later than the calculator suggests.

If you're interested in learning more about Social Security, you may want to read my book: Social Security Made Simple.
Cheers

vtMaps
Posts: 687
Joined: Tue Oct 14, 2008 12:05 pm
Location: central Vermont

Re: Open Social Security strategy results question

Post by vtMaps » Sat May 16, 2020 12:51 pm

The calculator is making (possibly unwarranted) assumptions about your lifespans. If you ask the calculator to give you the optimal strategy assuming you both live to be 110, the calculator will likely have your wife waiting until at least her full retirement age before claiming.

Likewise, if you ask the calculator to give you the optimal strategy assuming you both die in your 70's, the calculator will have both of you starting at age 62.

-vtMaps
Historical Fact: Justin Smith Morrill represented Vermont in congress, had a dog named 'Trump', and wrote legislation establishing the Land Grant Colleges.

Silk McCue
Posts: 4712
Joined: Thu Feb 25, 2016 7:11 pm

Re: Open Social Security strategy results question

Post by Silk McCue » Sat May 16, 2020 1:02 pm

vtMaps wrote:
Sat May 16, 2020 12:51 pm
The calculator is making (possibly unwarranted) assumptions about your lifespans. If you ask the calculator to give you the optimal strategy assuming you both live to be 110, the calculator will likely have your wife waiting until at least her full retirement age before claiming.

Likewise, if you ask the calculator to give you the optimal strategy assuming you both die in your 70's, the calculator will have both of you starting at age 62.

-vtMaps
The program doesn’t make any assumptions about your lifespan although you can enter assumptions about death dates. It is far more sophisticated than that.

If you would like to learn about the tool please take a few minutes to read the following.

https://opensocialsecurity.com/about

Cheers

N.Y.Cab
Posts: 42
Joined: Wed Feb 07, 2018 8:46 pm

Re: Open Social Security strategy results question

Post by N.Y.Cab » Sat May 16, 2020 1:19 pm

I think the problem is with the negative yield on 20-years TIPS that is being used as input. I found the results to be more reasonable if the real discount rate is above .76%.

ModifiedDuration
Posts: 112
Joined: Sat Dec 05, 2015 4:33 pm

Re: Open Social Security strategy results question

Post by ModifiedDuration » Sat May 16, 2020 1:22 pm

N.Y.Cab wrote:
Sat May 16, 2020 1:19 pm
I think the problem is with the negative yield on 20-years TIPS that is being used as input. I found the results to be more reasonable if the real discount rate is above .76%.
Great minds think alike. I ball parked an estimate for recent years and was using 0.80%.

User avatar
Watty
Posts: 19767
Joined: Wed Oct 10, 2007 3:55 pm

Re: Open Social Security strategy results question

Post by Watty » Sat May 16, 2020 1:27 pm

markcoop wrote:
Sat May 16, 2020 11:10 am
So, really not a big deal either way, but just wanted to understand why it would be the case.
Ideally Social Security would be actuarial neutral so that whenever you start it you would get the same value adjusted for your life expectancy and interest rates. This would be like how a commercial annuity payout would very depending on if you bought it when you were 62 or 62 and 7 months.

In practice Social Security is not that exact since it;

1) Does not adjust for gender
2) Does not change when interest rates change

That web side does adjust for gender and the current interest rates but otherwise goes through a similar calculation. The $235 difference could just be accounted for by those adjustments.

marcopolo
Posts: 3077
Joined: Sat Dec 03, 2016 10:22 am

Re: Open Social Security strategy results question

Post by marcopolo » Sat May 16, 2020 2:23 pm

Silk McCue wrote:
Sat May 16, 2020 1:02 pm
vtMaps wrote:
Sat May 16, 2020 12:51 pm
The calculator is making (possibly unwarranted) assumptions about your lifespans. If you ask the calculator to give you the optimal strategy assuming you both live to be 110, the calculator will likely have your wife waiting until at least her full retirement age before claiming.

Likewise, if you ask the calculator to give you the optimal strategy assuming you both die in your 70's, the calculator will have both of you starting at age 62.

-vtMaps
The program doesn’t make any assumptions about your lifespan although you can enter assumptions about death dates. It is far more sophisticated than that.

If you would like to learn about the tool please take a few minutes to read the following.

https://opensocialsecurity.com/about

Cheers
Huh? How could it possibly make any suggestions without some assumptions about lifespan?!?

The link you provided clearly explains the methodology, which states that survival probability for each year computed from SSA period life table, which produces results that approximates average life span.

How did you conclude from that description that the tool does not make any assumptions about one's life span?
Once in a while you get shown the light, in the strangest of places if you look at it right.

Silk McCue
Posts: 4712
Joined: Thu Feb 25, 2016 7:11 pm

Re: Open Social Security strategy results question

Post by Silk McCue » Sat May 16, 2020 2:29 pm

marcopolo wrote:
Sat May 16, 2020 2:23 pm
Silk McCue wrote:
Sat May 16, 2020 1:02 pm
vtMaps wrote:
Sat May 16, 2020 12:51 pm
The calculator is making (possibly unwarranted) assumptions about your lifespans. If you ask the calculator to give you the optimal strategy assuming you both live to be 110, the calculator will likely have your wife waiting until at least her full retirement age before claiming.

Likewise, if you ask the calculator to give you the optimal strategy assuming you both die in your 70's, the calculator will have both of you starting at age 62.

-vtMaps
The program doesn’t make any assumptions about your lifespan although you can enter assumptions about death dates. It is far more sophisticated than that.

If you would like to learn about the tool please take a few minutes to read the following.

https://opensocialsecurity.com/about

Cheers
Huh? How could it possibly make any suggestions without some assumptions about lifespan?!?

The link you provided clearly explains the methodology, which states that survival probability for each year computed from SSA period life table, which produces results that approximates average life span.

How did you conclude from that description that the tool does not make any assumptions about one's life span?
I don’t consider the methodology to be an assumption. I consider it to be a statistical analysis.

Cheers

marcopolo
Posts: 3077
Joined: Sat Dec 03, 2016 10:22 am

Re: Open Social Security strategy results question

Post by marcopolo » Sat May 16, 2020 2:38 pm

Silk McCue wrote:
Sat May 16, 2020 2:29 pm
marcopolo wrote:
Sat May 16, 2020 2:23 pm
Silk McCue wrote:
Sat May 16, 2020 1:02 pm
vtMaps wrote:
Sat May 16, 2020 12:51 pm
The calculator is making (possibly unwarranted) assumptions about your lifespans. If you ask the calculator to give you the optimal strategy assuming you both live to be 110, the calculator will likely have your wife waiting until at least her full retirement age before claiming.

Likewise, if you ask the calculator to give you the optimal strategy assuming you both die in your 70's, the calculator will have both of you starting at age 62.

-vtMaps
The program doesn’t make any assumptions about your lifespan although you can enter assumptions about death dates. It is far more sophisticated than that.

If you would like to learn about the tool please take a few minutes to read the following.

https://opensocialsecurity.com/about

Cheers
Huh? How could it possibly make any suggestions without some assumptions about lifespan?!?

The link you provided clearly explains the methodology, which states that survival probability for each year computed from SSA period life table, which produces results that approximates average life span.

How did you conclude from that description that the tool does not make any assumptions about one's life span?
I don’t consider the methodology to be an assumption. I consider it to be a statistical analysis.

Cheers
It is a statistical analysis with certain assumtions. The main one being average life span as defined by the SSA tables. If the life span was also being statistically modelled, it would NOT need the SSA table as an input.

EDIT: Actually, taking a closer look at the methodology, i don't think there is any statiscal analysis at all. It seems quite deterministic calculation, with a given set of assumptions.
Once in a while you get shown the light, in the strangest of places if you look at it right.

ralph124cf
Posts: 2507
Joined: Tue Apr 01, 2014 11:41 am

Re: Open Social Security strategy results question

Post by ralph124cf » Sat May 16, 2020 3:23 pm

There a number of reasons that I find this analysis incomplete. It does not take non-wage non-SS income into account, such as interest, dividends, annuities or royalties. It also does not consider the opportunity to do ROTH conversions at lower tax rates before starting SS.

When I hit full retirement age of 66, I filed and suspended my benefit. Then a couple of months later when DW hit 66, she filed a restricted application for ONLY spousal benefits. Then when we hit age 70, I started collecting, and she switched from spousal to her own age 70 full benefit.

I know that the law has been changed to eliminate this strategy, but I am not sure of the effective date. It may be that no one under today's FRA is eligible for this strategy.

Ralph

Silk McCue
Posts: 4712
Joined: Thu Feb 25, 2016 7:11 pm

Re: Open Social Security strategy results question

Post by Silk McCue » Sat May 16, 2020 3:33 pm

ralph124cf wrote:
Sat May 16, 2020 3:23 pm
There a number of reasons that I find this analysis incomplete. It does not take non-wage non-SS income into account, such as interest, dividends, annuities or royalties. It also does not consider the opportunity to do ROTH conversions at lower tax rates before starting SS.

When I hit full retirement age of 66, I filed and suspended my benefit. Then a couple of months later when DW hit 66, she filed a restricted application for ONLY spousal benefits. Then when we hit age 70, I started collecting, and she switched from spousal to her own age 70 full benefit.

I know that the law has been changed to eliminate this strategy, but I am not sure of the effective date. It may be that no one under today's FRA is eligible for this strategy.
It doesn't need to do any of those things mentioned in your first paragraph, nor does it purport to do so. It isn't a retirement financial planner. It is a single tool focused on a single topic analysis. It is up to the user to determine how that information provided fits, or doesn't fit, in their individual plans and goals.

Cheers

User avatar
FiveK
Posts: 9336
Joined: Sun Mar 16, 2014 2:43 pm

Re: Open Social Security strategy results question

Post by FiveK » Sat May 16, 2020 4:10 pm

ralph124cf wrote:
Sat May 16, 2020 3:23 pm
There a number of reasons that I find this analysis incomplete. It does not take non-wage non-SS income into account, such as interest, dividends, annuities or royalties. It also does not consider the opportunity to do ROTH conversions at lower tax rates before starting SS.
There is no tool - either free or commercially available - that will solve the non-linear optimization problem of "maximum spendable" using all degrees of freedom available to most retirees, including
- SS start dates
- Roth conversion amounts and timing
- accurate federal and state tax calculations
- health insurance (e.g., ACA and Medicare)

There are programs such as Open Social Security, Retiree Portfolio Model, Personal Finance Toolbox, I-ORP, etc., that do very well for the portion of that problem they do address. In other words, "some assembly required"....

Topic Author
markcoop
Posts: 1159
Joined: Fri Mar 02, 2007 8:36 am

Re: Open Social Security strategy results question

Post by markcoop » Sun May 17, 2020 9:49 am

Thanks for the replies. I am just starting to look into SS. So, let me back up a bit and ask how this works in general. My hunch was my death date, according to the tables, was the break even point. In other words, no matter when I started to take SS, if I died on that day I would have received the same amount (using present value dollars). So, my decision was if I think I would outlive the tables, then it is better to take SS later. Conversely, if I didn't outlive the table, it would be better to take SS earlier. But on the exact date, it didn't matter. Is that correct?
Mark

User avatar
ObliviousInvestor
Posts: 3699
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: Open Social Security strategy results question

Post by ObliviousInvestor » Sun May 17, 2020 10:10 am

markcoop wrote:
Sun May 17, 2020 9:49 am
Thanks for the replies. I am just starting to look into SS. So, let me back up a bit and ask how this works in general. My hunch was my death date, according to the tables, was the break even point. In other words, no matter when I started to take SS, if I died on that day I would have received the same amount (using present value dollars). So, my decision was if I think I would outlive the tables, then it is better to take SS later. Conversely, if I didn't outlive the table, it would be better to take SS earlier. But on the exact date, it didn't matter. Is that correct?
You're on the right path, but there's a bit more to it than that, because we're talking about a couple here.

If you find that the filing date for the lower earning spouse doesn't matter very much (i.e., the decision is roughly actuarially neutral for that person), that means that the first-to-die joint life expectancy (i.e., expected length of time before one person dies -- either person) roughly matches the breakeven point for the lower earning spouse.

That is, for the lower earning spouse we are concerned with how long we expect it to be before one person dies. And for the higher earning spouse we are concerned with how long we expect it to be until both people have died.

But yes, the longer those two expected periods, the more advantageous it becomes for the applicable person to delay.
Mike Piper, author/blogger

vtMaps
Posts: 687
Joined: Tue Oct 14, 2008 12:05 pm
Location: central Vermont

Re: Open Social Security strategy results question

Post by vtMaps » Sun May 17, 2020 12:21 pm

markcoop wrote:
Sun May 17, 2020 9:49 am
if I think I would outlive the tables, then it is better to take SS later. Conversely, if I didn't outlive the table, it would be better to take SS earlier. But on the exact date, it didn't matter. Is that correct?
I think that's not quite right. When you delay SS you are living on non SS money. The calculator assumes that if you start SS early, you will invest that non SS money. You can change the investment growth rate, and if you assume a high enough growth rate, the calculator will tell both of you to start benefits at 62.

--vtMaps
Historical Fact: Justin Smith Morrill represented Vermont in congress, had a dog named 'Trump', and wrote legislation establishing the Land Grant Colleges.

Topic Author
markcoop
Posts: 1159
Joined: Fri Mar 02, 2007 8:36 am

Re: Open Social Security strategy results question

Post by markcoop » Mon May 18, 2020 2:43 pm

Sounds to me this calculator is more complicated than I originally thought (especially if it assumes when you start SS early, you will invest that non SS money). I am beginning to understand the impact of the death rates for the married couple, who dies first and who earns more money. Thank you Mike for the calculator and responding to my post.

Getting back to my original question, in starting my journey to understand how to use SS and when to retire, I had assumed from my limited knowledge that my spouse would be taking SS as early as possible and I would be waiting as long as possible (I earned more than her, but not enough for her to have any annual spousal benefit when I start my SS). I was a bit surprised at the results suggesting my spouse start her SS a few months later. I guess if the difference is so small (couple hundred dollar difference on the present value of SS), it probably won't matter much and other things would probably dictate the decision. For example, if my 401K is large enough and my RMDs will generate a decent amount of taxes, perhaps taking her SS and 62+1month would be better overall since it will lower our taxes a bit when those RMDs start to come in. I still have a few more years to make any decision, but I think I'm starting to understand my homework to do.
Mark

User avatar
ObliviousInvestor
Posts: 3699
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: Open Social Security strategy results question

Post by ObliviousInvestor » Mon May 18, 2020 2:49 pm

markcoop wrote:
Mon May 18, 2020 2:43 pm
Sounds to me this calculator is more complicated than I originally thought (especially if it assumes when you start SS early, you will invest that non SS money).
Yes, it's assuming that if you file early you invest the money (or alternatively that you spend the benefits received, and thereby have more of your portfolio that remains invested).

markcoop wrote:
Mon May 18, 2020 2:43 pm
I had assumed from my limited knowledge that my spouse would be taking SS as early as possible and I would be waiting as long as possible (I earned more than her, but not enough for her to have any annual spousal benefit when I start my SS). I was a bit surprised at the results suggesting my spouse start her SS a few months later. I guess if the difference is so small (couple hundred dollar difference on the present value of SS), it probably won't matter much and other things would probably dictate the decision.
Your original understanding (higher earner at 70, lower earner early) is good. That's a great rough draft plan for most people. And it's often the case that it doesn't matter very much when the lower earner files. It's generally far less impactful of a decision than when the higher earner files. (Both because we're talking about the smaller benefit amount -- so there are simply smaller dollar amounts involved -- and because the decision is closer to neutral for this person, in most cases.)
Mike Piper, author/blogger

User avatar
FiveK
Posts: 9336
Joined: Sun Mar 16, 2014 2:43 pm

Re: Open Social Security strategy results question

Post by FiveK » Mon May 18, 2020 2:50 pm

markcoop wrote:
Mon May 18, 2020 2:43 pm
For example, if my 401K is large enough and my RMDs will generate a decent amount of taxes, perhaps taking her SS and 62+1month would be better overall since it will lower our taxes a bit when those RMDs start to come in.
Or perhaps delay her SS by several years so more traditional to Roth conversions are possible at lower marginal rates, thus decreasing the size of later RMDs.

Topic Author
markcoop
Posts: 1159
Joined: Fri Mar 02, 2007 8:36 am

Re: Open Social Security strategy results question

Post by markcoop » Mon May 18, 2020 3:18 pm

FiveK wrote:
Mon May 18, 2020 2:50 pm
markcoop wrote:
Mon May 18, 2020 2:43 pm
For example, if my 401K is large enough and my RMDs will generate a decent amount of taxes, perhaps taking her SS and 62+1month would be better overall since it will lower our taxes a bit when those RMDs start to come in.
Or perhaps delay her SS by several years so more traditional to Roth conversions are possible at lower marginal rates, thus decreasing the size of later RMDs.
I do plan to do Roth conversions between retirement and age 70. The difference in Roth conversions would be pretty small whether my spouse starts SS at 62+1 month or 63 or 64. By delaying her SS a couple of years, I have more room for Roth conversions initially, but less room for Roth conversions during the last few years till 70.
Mark

User avatar
FiveK
Posts: 9336
Joined: Sun Mar 16, 2014 2:43 pm

Re: Open Social Security strategy results question

Post by FiveK » Mon May 18, 2020 3:35 pm

markcoop wrote:
Mon May 18, 2020 3:18 pm
FiveK wrote:
Mon May 18, 2020 2:50 pm
markcoop wrote:
Mon May 18, 2020 2:43 pm
For example, if my 401K is large enough and my RMDs will generate a decent amount of taxes, perhaps taking her SS and 62+1month would be better overall since it will lower our taxes a bit when those RMDs start to come in.
Or perhaps delay her SS by several years so more traditional to Roth conversions are possible at lower marginal rates, thus decreasing the size of later RMDs.
I do plan to do Roth conversions between retirement and age 70. The difference in Roth conversions would be pretty small whether my spouse starts SS at 62+1 month or 63 or 64. By delaying her SS a couple of years, I have more room for Roth conversions initially, but less room for Roth conversions during the last few years till 70.
Yes, there are trade-offs. ;)

If you are looking at some aggressive conversion amounts, better to do so before she reaches age 63 and a high AGI affects Medicare premiums at age 65 - unless her birthday is toward the end of the year and the IRMAA effect will apply for only a few months, in which case you get an "extra" year.

Topic Author
markcoop
Posts: 1159
Joined: Fri Mar 02, 2007 8:36 am

Re: Open Social Security strategy results question

Post by markcoop » Mon May 18, 2020 10:17 pm

FiveK wrote:
Mon May 18, 2020 3:35 pm
markcoop wrote:
Mon May 18, 2020 3:18 pm
FiveK wrote:
Mon May 18, 2020 2:50 pm
markcoop wrote:
Mon May 18, 2020 2:43 pm
For example, if my 401K is large enough and my RMDs will generate a decent amount of taxes, perhaps taking her SS and 62+1month would be better overall since it will lower our taxes a bit when those RMDs start to come in.
Or perhaps delay her SS by several years so more traditional to Roth conversions are possible at lower marginal rates, thus decreasing the size of later RMDs.
I do plan to do Roth conversions between retirement and age 70. The difference in Roth conversions would be pretty small whether my spouse starts SS at 62+1 month or 63 or 64. By delaying her SS a couple of years, I have more room for Roth conversions initially, but less room for Roth conversions during the last few years till 70.
Yes, there are trade-offs. ;)

If you are looking at some aggressive conversion amounts, better to do so before she reaches age 63 and a high AGI affects Medicare premiums at age 65 - unless her birthday is toward the end of the year and the IRMAA effect will apply for only a few months, in which case you get an "extra" year.
I haven't planned this all out yet, but I would imagine I would convert to the top of the current 12% bracket (AGI $78,950 MFJ) per year for 10 years (or whatever that tax bracket turns into when I retire). That is well below the Medicare high-income surcharge (IRMAA) currently at $170,000 for MFJ. So, if my spouse's SS is something like $14k per year, I guess for those years till 70, I'd convert $78,950-$14K (or $64,950)/yr. I also have some I-bonds that will mature that will affect some conversion years as well. Seems if I do about 10 years of conversions with SS for 8 years and selling some I-bonds, up to the top of 12% bracket, will end up being around $600K of conversions.

Any SS calculators allow you to play around with Roth conversions/I-bond redemptions?

Edit: I do understand that not all SS is taxed, so my numbers above are not exact.
Last edited by markcoop on Mon May 18, 2020 10:29 pm, edited 1 time in total.
Mark

User avatar
tooluser
Posts: 617
Joined: Sat Oct 01, 2011 7:04 pm

Re: Open Social Security strategy results question

Post by tooluser » Mon May 18, 2020 10:27 pm

OpenSocialSecurity is a wonderful tool. Everyone should run sensitivity analyses on the various age(s) of retirement(s) and see what the difference is.

As a single, maxed-out for many of my 35+ contribution years person, I found that the optimal claiming age was very close to either extreme (age 62 or 70). As a result it allowed me to consider optimizing for longevity insurance or bad luck instead of maximum payout, really not much difference. The net result for me is that I will claim at age 70, or earlier if I need the money or think I am soon to pass.

I can imagine scenarios where this makes sense for married couples too.

User avatar
Eagle33
Posts: 714
Joined: Wed Aug 30, 2017 3:20 pm

Re: Open Social Security strategy results question

Post by Eagle33 » Tue May 19, 2020 3:56 am

markcoop wrote:
Mon May 18, 2020 10:17 pm
I haven't planned this all out yet, but I would imagine I would convert to the top of the current 12% bracket (AGI $78,950 MFJ) per year for 10 years (or whatever that tax bracket turns into when I retire). That is well below the Medicare high-income surcharge (IRMAA) currently at $170,000 for MFJ. So, if my spouse's SS is something like $14k per year, I guess for those years till 70, I'd convert $78,950-$14K (or $64,950)/yr. I also have some I-bonds that will mature that will affect some conversion years as well. Seems if I do about 10 years of conversions with SS for 8 years and selling some I-bonds, up to the top of 12% bracket, will end up being around $600K of conversions.

Any SS calculators allow you to play around with Roth conversions/I-bond redemptions?

Edit: I do understand that not all SS is taxed, so my numbers above are not exact.
Your AGI # is incorrect. $78,950 is 2019 "taxable" income after your deduction for the top of MFJ 12% tax bracket. AGI is income before taking your deduction. Assuming standard deduction, then you have another $24,800 of available income in the 12% bracket.
Rocket science is not “rocket science” to a rocket scientist, just as personal finance is not “rocket science” to a Boglehead.

User avatar
FiveK
Posts: 9336
Joined: Sun Mar 16, 2014 2:43 pm

Re: Open Social Security strategy results question

Post by FiveK » Tue May 19, 2020 4:44 am

markcoop wrote:
Mon May 18, 2020 10:17 pm
Any SS calculators allow you to play around with Roth conversions/I-bond redemptions?
Any of the better ones mentioned in US and state income tax calculator - Bogleheads.org should get the SS benefit taxability correct.

Topic Author
markcoop
Posts: 1159
Joined: Fri Mar 02, 2007 8:36 am

Re: Open Social Security strategy results question

Post by markcoop » Tue May 19, 2020 7:07 am

Eagle33 wrote:
Tue May 19, 2020 3:56 am
markcoop wrote:
Mon May 18, 2020 10:17 pm
I haven't planned this all out yet, but I would imagine I would convert to the top of the current 12% bracket (AGI $78,950 MFJ) per year for 10 years (or whatever that tax bracket turns into when I retire). That is well below the Medicare high-income surcharge (IRMAA) currently at $170,000 for MFJ. So, if my spouse's SS is something like $14k per year, I guess for those years till 70, I'd convert $78,950-$14K (or $64,950)/yr. I also have some I-bonds that will mature that will affect some conversion years as well. Seems if I do about 10 years of conversions with SS for 8 years and selling some I-bonds, up to the top of 12% bracket, will end up being around $600K of conversions.

Any SS calculators allow you to play around with Roth conversions/I-bond redemptions?

Edit: I do understand that not all SS is taxed, so my numbers above are not exact.
Your AGI # is incorrect. $78,950 is 2019 "taxable" income after your deduction for the top of MFJ 12% tax bracket. AGI is income before taking your deduction. Assuming standard deduction, then you have another $24,800 of available income in the 12% bracket.

Thank you. I just played around with Turbotax and the equations are pretty simple. The standard deduction is $24,400 for MFJ. Oddly enough, if I enter the max income for a Roth conversion of $103,350 ($78,950+24,400) to max out the 12% bracket, Turbotax comes out with a total tax of $9,092 which would be $6 more than what the top of the 12% bracket should be. Not going to worry why Turbotax is slightly off from what I think it should be. I then played around with adding SS for my wife and it seems like a straight 85% of SS would be taxable. So, it seems pretty simple to compute. Adding in I-bonds would also seem pretty easy to do. So, the big question in my head now is if it is generally better to delay both spouse's SS to do more conversions (for 10 years) or is it better to do less conversions and start her SS early (assuming mine wouldn't start till 70). I think I would need to play around more to figure that out. If anyone has a rough opinion based on $14K early SS for spouse (with no annual spousal benefit jump at my SS start) and say a starting tax-deferred balance of $2 million, that would be appreciated.
Mark

PatrickA5
Posts: 531
Joined: Mon Jul 28, 2014 1:55 pm

Re: Open Social Security strategy results question

Post by PatrickA5 » Tue May 19, 2020 11:36 am

When I run Open Social Security, I get me filing at 64 and DW (higher earner) at 70. Changing the death dates or discount rate just a little results in me filing at 62 - which is what it said in previous years.

I also bought Maximize My Social Security ($40) and was surprised to see it suggested me waiting until 70 to file. A little digging into the program reveals that is heavily influenced by the "longevity insurance" thoughts on when to take SS. I was more than surprised to see a $325K increase in NPV by waiting until 70. BUT, upon further investigation, I figured out that it has as a default both of us living to 100! Obviously, if you plan to each live to 100 then you'd want to delay and the change in NPV would be huge (as compared to 62). It's discount rate is programed at a 0% real rate, which also makes it more advantageous to delay. After plugging in some more realistic death ages (85 and 95), it lowered my filing date down to 64 and the difference in NPV was minimal.

So, after a decade of running calculators and fretting over this stuff, I filed for SS yesterday and will start collecting at 62. DW will wait until 70.

Big Dog
Posts: 1854
Joined: Mon Sep 07, 2015 4:12 pm

Re: Open Social Security strategy results question

Post by Big Dog » Tue May 19, 2020 11:40 am

Watty wrote:
Sat May 16, 2020 1:27 pm
markcoop wrote:
Sat May 16, 2020 11:10 am
So, really not a big deal either way, but just wanted to understand why it would be the case.
Ideally Social Security would be actuarial neutral so that whenever you start it you would get the same value adjusted for your life expectancy and interest rates. This would be like how a commercial annuity payout would very depending on if you bought it when you were 62 or 62 and 7 months.

In practice Social Security is not that exact since it;

1) Does not adjust for gender
2) Does not change when interest rates change

That web side does adjust for gender and the current interest rates but otherwise goes through a similar calculation. The $235 difference could just be accounted for by those adjustments.
It's not designed to be actuarially neutral. SS includes a spousal benefit, for example. Ditto survivor benefit. It is also based on a unisex death table.

User avatar
#Cruncher
Posts: 2991
Joined: Fri May 14, 2010 2:33 am
Location: New York City
Contact:

Re: Open Social Security strategy results question

Post by #Cruncher » Tue May 19, 2020 12:24 pm

ObliviousInvestor wrote:
Mon May 18, 2020 2:49 pm
… it's often the case that it doesn't matter very much when the lower earner files. It's generally far less impactful of a decision than when the higher earner files.
Correct. This is because the smaller payment generally continues only while both members of a couple live, while the larger payment continues while either lives. This difference can be seen in the following two tables. Both show the present value (PV) at age 62 per $100 of Primary Insurance Amount (PIA) when discounted at -0.28%. Benefits are calculated for nine different claiming ages: 62+1 mo, 63, 64, …, 69, and 70 for someone with a Normal Retirement Age of 67. The first table weights benefits according to the survival probability for a man / woman couple of the same age both being alive. The second table is the same except benefits are weighted by the probability that either Is alive.

Of the nine examined claiming ages, 66 provides the largest PV in the first case (while both live). But claiming at age 62 would produces almost as large a PV, only 1.6% less. In the second case (while either lives) the largest PV occurs with a claiming age of 70; and to instead start at age 62+1 month would produce a much smaller PV, 20.0% less.

Code: Select all

Claim age         62.083   63.000   64.000   65.000   66.000   67.000   68.000   69.000   70.000
Monthly benefit   70.417   75.000   80.000   86.667   93.333  100.000  108.000  116.000  124.000
Present value     13,493   13,561   13,541   13,690   13,712   13,613   13,570   13,394   13,095
Less than best       219      151      171       22                98      142      318      616
Percent less        1.6%     1.1%     1.2%     0.2%              0.7%     1.0%     2.3%     4.5%

Code: Select all

Claim age         62.083   63.000   64.000   65.000   66.000   67.000   68.000   69.000   70.000
Monthly benefit   70.417   75.000   80.000   86.667   93.333  100.000  108.000  116.000  124.000
Present value     23,241   23,927   24,557   25,556   26,391   27,063   27,917   28,576   29,040
Less than best     5,799    5,113    4,483    3,484    2,649    1,977    1,123      465
Percent less       20.0%    17.6%    15.4%    12.0%     9.1%     6.8%     3.9%     1.6%
The tables are from the SS sheet of my Longevity Estimator Excel workbook and use the 2016 SSA Period Life Table [1] for a man / woman couple of the same age (selected on the Alive sheet). This mortality table and the -0.28% discount rate [2] correspond to the defaults in Mike Piper's Open Social Security (OSS) claiming age optimizer.
  1. The 2017 SSA Period Life Table has just been released; but it is little changed from the 2016 Period Life Table OSS uses as the default.
  2. OSS uses the yield shown in the 20 YR column of the Daily Treasury Real Yield Curve Rates. -0.28% is the figure for 5/15/2020. It appears however to be the average of the 10 YR and 30 YR figures, not the yield of the actual TIPS maturing in about 20 years, the 2.125% of Feb 2040 which currently yields about 0.14% points more per WSJ TIPS Quotes.

Topic Author
markcoop
Posts: 1159
Joined: Fri Mar 02, 2007 8:36 am

Re: Open Social Security strategy results question

Post by markcoop » Tue May 19, 2020 12:31 pm

PatrickA5 wrote:
Tue May 19, 2020 11:36 am
When I run Open Social Security, I get me filing at 64 and DW (higher earner) at 70. Changing the death dates or discount rate just a little results in me filing at 62 - which is what it said in previous years.
So, right now the default discount rate is -0.26%. Maybe that's the best estimate for the future, but if I plug in any value above 0.4% I get the result I expected. Although I understand why using 20 year TIPS rate makes sense, it does seem odd to me that the results of the calculator I am using to help me with future plans can change from day to day.
Mark

Chuck
Posts: 2113
Joined: Thu May 21, 2009 12:19 pm

Re: Open Social Security strategy results question

Post by Chuck » Tue May 19, 2020 1:08 pm

markcoop wrote:
Tue May 19, 2020 7:07 am
Oddly enough, if I enter the max income for a Roth conversion of $103,350 ($78,950+24,400) to max out the 12% bracket, Turbotax comes out with a total tax of $9,092 which would be $6 more than what the top of the 12% bracket should be.
Because tax tables go in chunks of $100, so $103,350 is rounded up to $103,400 before calculating the tax.

Topic Author
markcoop
Posts: 1159
Joined: Fri Mar 02, 2007 8:36 am

Re: Open Social Security strategy results question

Post by markcoop » Tue May 19, 2020 1:20 pm

Chuck wrote:
Tue May 19, 2020 1:08 pm
markcoop wrote:
Tue May 19, 2020 7:07 am
Oddly enough, if I enter the max income for a Roth conversion of $103,350 ($78,950+24,400) to max out the 12% bracket, Turbotax comes out with a total tax of $9,092 which would be $6 more than what the top of the 12% bracket should be.
Because tax tables go in chunks of $100, so $103,350 is rounded up to $103,400 before calculating the tax.
Ah...Thanks. It does bother me when things don't work exactly.
Mark

User avatar
#Cruncher
Posts: 2991
Joined: Fri May 14, 2010 2:33 am
Location: New York City
Contact:

Re: Open Social Security strategy results question

Post by #Cruncher » Tue May 19, 2020 3:41 pm

Chuck wrote:
Tue May 19, 2020 1:08 pm
markcoop wrote:
Tue May 19, 2020 7:07 am
Oddly enough, if I enter the max income for a Roth conversion of $103,350 ($78,950+24,400) to max out the 12% bracket, Turbotax comes out with a total tax of $9,092 which would be $6 more than what the top of the 12% bracket should be.
Because tax tables go in chunks of $100, so $103,350 is rounded up to $103,400 before calculating the tax.
No. The tax tables (see page 12 of Tax and Earned Income Credit Tables) are in $50 increments. So markcoop's $78,950 taxable income falls in the row ["At least"] 78,950 ["But less than"] 79,000. For each row the tax is computed on the midpoint, in this case 78,975 -- which is $25 into the 22% bracket.

Code: Select all

1,940.00 = 10% *  19400
7,146.00 = 12% * (78950 - 19400)
    5.50 = 22% * (78975 - 78950)
--------
9,091.50 = total tax, rounds to 9,092

User avatar
celia
Posts: 10842
Joined: Sun Mar 09, 2008 6:32 am
Location: SoCal

Re: Open Social Security strategy results question

Post by celia » Tue May 19, 2020 4:45 pm

markcoop wrote:
Sat May 16, 2020 11:10 am
What I don't understand is under what circumstances it concludes she should start her early withdrawals at age 62 and 7 months and not age 62. When I plugged in my alternative strategy (starting her benefits at age 62), the present value difference was really small - $235 worse in my alternative. So, really not a big deal either way, but just wanted to understand why it would be the case.
You answered your own question that there was hardly a difference in DW starting at 62 or at 62+7 mo. The difference was really small, but 62+7 mo gave a slight advantage of $235. This is based on the cumulative benefit received (7 fewer months of benefits, but bigger monthly pay-out if starting at 62+7 mo) and doesn't take other things like Roth conversions into account.

Not only does taking SS early use up some of the room that Roth conversions would use in the same tax bracket, but the conversions MAY make more of the SS be taxed. Paying more tax because of the SS is the same as getting SS minus that increase in tax. That would completely change your Open Social Security calculations.

The general concept that many of us use is to keep our Taxable Income fairly level each year from retirement until you die, so the taxes would be relatively level each year. It makes no sense to have low income (and low taxes) the first years of retirement only to have your tax bracket jump after age 70 or 72 due to bigger RMDs, SS, and other income streams you may have.

What I did when getting ready to retire was to make a chart of our expected income streams for each year and estimate the taxes. I tried to get the income to increase (using more Roth conversions) to use up all the space in my marginal tax bracket each year. Some years we went a little into the next bracket, but only the "little" extra was taxed at the higher tax rate. Not a big deal to me.
A dollar in Roth is worth more than a dollar in a taxable account. A dollar in taxable is worth more than a dollar in a tax-deferred account.

vtMaps
Posts: 687
Joined: Tue Oct 14, 2008 12:05 pm
Location: central Vermont

Re: Open Social Security strategy results question

Post by vtMaps » Tue May 19, 2020 8:03 pm

celia wrote:
Tue May 19, 2020 4:45 pm
The general concept that many of us use is to keep our Taxable Income fairly level each year from retirement until you die, so the taxes would be relatively level each year.
That is not always the best strategy. For example:
A single taxpayer with SS = $35,000 and RMD = $30,000 wishes to take (in addition to their RMD) taxable distributions of $16,000 per year.

If he takes $16,000 per year for 4 years ($64,000 total), his total tax bill for the four years is $37,346.

If he takes $50,500 one year and $4,500 the other three years ($64,000 total), his total tax bill for the four years is $31,150.

The reason for this is the so called social security tax hump.

--vtMaps
Historical Fact: Justin Smith Morrill represented Vermont in congress, had a dog named 'Trump', and wrote legislation establishing the Land Grant Colleges.

User avatar
David Jay
Posts: 8675
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Open Social Security strategy results question

Post by David Jay » Tue May 19, 2020 8:10 pm

tooluser wrote:
Mon May 18, 2020 10:27 pm
OpenSocialSecurity is a wonderful tool. Everyone should run sensitivity analyses on the various age(s) of retirement(s) and see what the difference is.

As a single, maxed-out for many of my 35+ contribution years person, I found that the optimal claiming age was very close to either extreme (age 62 or 70). As a result it allowed me to consider optimizing for longevity insurance or bad luck instead of maximum payout, really not much difference. The net result for me is that I will claim at age 70, or earlier if I need the money or think I am soon to pass.

I can imagine scenarios where this makes sense for married couples too.
Probably not, the big difference between single and married is that, after the death of one spouse, the survivor receives a total benefit equal to benefit of the higher earner. So in almost every case it makes sense for the higher earner to defer claiming to maximize the survivor's benefit. When you look at the mortality tables for (2) individuals, it dramatically extends the payout period as compared to (1) person.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius

User avatar
tooluser
Posts: 617
Joined: Sat Oct 01, 2011 7:04 pm

Re: Open Social Security strategy results question

Post by tooluser » Tue May 19, 2020 9:00 pm

David Jay wrote:
Tue May 19, 2020 8:10 pm
tooluser wrote:
Mon May 18, 2020 10:27 pm

I can imagine scenarios where this makes sense...
...in almost every case...
So we agree! I didn't say it would be common, but that people should check. Peace!

User avatar
celia
Posts: 10842
Joined: Sun Mar 09, 2008 6:32 am
Location: SoCal

Re: Open Social Security strategy results question

Post by celia » Tue May 19, 2020 10:01 pm

vtMaps wrote:
Tue May 19, 2020 8:03 pm
celia wrote:
Tue May 19, 2020 4:45 pm
The general concept that many of us use is to keep our Taxable Income fairly level each year from retirement until you die, so the taxes would be relatively level each year.
That is not always the best strategy. For example:
A single taxpayer with SS = $35,000 and RMD = $30,000 wishes to take (in addition to their RMD) taxable distributions of $16,000 per year.

If he takes $16,000 per year for 4 years ($64,000 total), his total tax bill for the four years is $37,346.

If he takes $50,500 one year and $4,500 the other three years ($64,000 total), his total tax bill for the four years is $31,150.

The reason for this is the so called social security tax hump.

--vtMaps
Interesting! In the year the individual takes the extra $50,500, do they enter a higher tax bracket?

User avatar
FiveK
Posts: 9336
Joined: Sun Mar 16, 2014 2:43 pm

Re: Open Social Security strategy results question

Post by FiveK » Wed May 20, 2020 1:53 am

celia wrote:
Tue May 19, 2020 10:01 pm
vtMaps wrote:
Tue May 19, 2020 8:03 pm
celia wrote:
Tue May 19, 2020 4:45 pm
The general concept that many of us use is to keep our Taxable Income fairly level each year from retirement until you die, so the taxes would be relatively level each year.
That is not always the best strategy. For example:
A single taxpayer with SS = $35,000 and RMD = $30,000 wishes to take (in addition to their RMD) taxable distributions of $16,000 per year.

If he takes $16,000 per year for 4 years ($64,000 total), his total tax bill for the four years is $37,346.

If he takes $50,500 one year and $4,500 the other three years ($64,000 total), his total tax bill for the four years is $31,150.

The reason for this is the so called social security tax hump.

--vtMaps
Interesting! In the year the individual takes the extra $50,500, do they enter a higher tax bracket?
Yes, but even more deleterious another IRMAA tier just comes into play. Stay just under that tier (e.g., $49K in the big year, and $5K in each of the small years), and the total of federal tax plus Medicare premiums is $39,135, vs. $40,156 for the $50.5K/$4.5K combo. Either is better than the $44,244 from 4 $16K years.

At least, that's what the personal finance toolbox calculates. Might be off by a few dollars per year due to using formulas instead of tables when taxable income <$100K, but close enough....

Topic Author
markcoop
Posts: 1159
Joined: Fri Mar 02, 2007 8:36 am

Re: Open Social Security strategy results question

Post by markcoop » Wed May 20, 2020 8:07 am

celia wrote:
Tue May 19, 2020 4:45 pm
Not only does taking SS early use up some of the room that Roth conversions would use in the same tax bracket, but the conversions MAY make more of the SS be taxed. Paying more tax because of the SS is the same as getting SS minus that increase in tax. That would completely change your Open Social Security calculations.
Assuming I plan to max out the 12% tax bracket every year, perhaps go into the 22% bracket a little bit, I think that means 85% of my SS will be taxed no matter when I take it and I will be under the IRMAA. Although it is important to keep these limits in mind, I expect my income to always be greater than $44,000 (threshold for SS of 85% taxed but less than $174,000 (IRMAA begins).

The issue of one starting SS early and a little less Roth conversions and neither starting SS early and more conversions is certainly an interesting one involving many factors. My quick analysis at this point has me leaning toward one starting SS early and a little less conversions. A couple of factors that figures into my thinking is the ending balance in the tax-deffered accounts (want some flexibility) and the ratio of tax-deffered to Roth (would have more than 50% Roth) at 70.
Mark

User avatar
David Jay
Posts: 8675
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Open Social Security strategy results question

Post by David Jay » Wed May 20, 2020 8:28 am

tooluser wrote:
Tue May 19, 2020 9:00 pm
David Jay wrote:
Tue May 19, 2020 8:10 pm
tooluser wrote:
Mon May 18, 2020 10:27 pm

I can imagine scenarios where this makes sense...
...in almost every case...
So we agree! I didn't say it would be common, but that people should check. Peace!
People should certainly check. We were planning to have my wife wait to FRA (66y 4mo) to get full spousal but opensocialsecurity said it was a better deal for her to claim @62 and collect an additional 52 months of payments.

I said “almost always“ because I have never seen an alternate recommendation for similar-age couples but I acknowledge that I haven’t seen all possible inputs.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius

User avatar
celia
Posts: 10842
Joined: Sun Mar 09, 2008 6:32 am
Location: SoCal

Re: Open Social Security strategy results question

Post by celia » Wed May 20, 2020 12:13 pm

markcoop wrote:
Wed May 20, 2020 8:07 am
The issue of one starting SS early and a little less Roth conversions and neither starting SS early and more conversions is certainly an interesting one involving many factors. My quick analysis at this point has me leaning toward one starting SS early and a little less conversions. A couple of factors that figures into my thinking is the ending balance in the tax-deffered accounts (want some flexibility) and the ratio of tax-deffered to Roth (would have more than 50% Roth) at 70.
Other factors to consider:

* Future changes in tax rates and the direction you would expect them to go. (Consider that we are currently at historically low tax rates and they are scheduled to increase in 2026 unless Congress should pass something else before then. The country has also just incurred a lot of new debt due to covid payments.)

* When either you or DW die, the survivor will then file as Single. One SS income stream will go away, but the same total RMDs will be required (assuming you are close in age). Tax brackets for Singles are half as much for MFJ, making it likely the survivor could pop up to the next tax bracket.

* If your estate is left to relatives after both of you die, the heirs could be in their high earning years when they need to withdraw from the Inherited IRA. Would you prefer to leave them more tax-deferred or Roth?
A dollar in Roth is worth more than a dollar in a taxable account. A dollar in taxable is worth more than a dollar in a tax-deferred account.

User avatar
#Cruncher
Posts: 2991
Joined: Fri May 14, 2010 2:33 am
Location: New York City
Contact:

Re: Open Social Security strategy results question

Post by #Cruncher » Thu May 21, 2020 6:30 am

celia wrote:
Tue May 19, 2020 10:01 pm
vtMaps wrote:
Tue May 19, 2020 8:03 pm
...That [equalizing taxable income every year] is not always the best strategy. For example:
A single taxpayer with SS = $35,000 and RMD = $30,000 wishes to take (in addition to their RMD) taxable distributions of $16,000 per year. If he takes $16,000 per year for 4 years ($64,000 total), his total tax bill for the four years is $37,346. If he takes $50,500 one year and $4,500 the other three years ($64,000 total), his total tax bill for the four years is $31,150.

The reason for this is the so called social security tax hump.
Interesting! In the year the individual takes the extra $50,500, do they enter a higher tax bracket?
In vtMaps example, the tax bracket does increase. But in spite of this the total tax bill decreases because much less Social Security benefits are being taxed. This happens because the better option has income either below the SS hump [1] or well over it; while the worse option has income within the hump every year. This can be better illustrated with a simpler example:
  • 2020 Single tax return with age 65+ standard deduction and $35,000 SS benefit.
  • Each year taxpayer has $34,432 RMD. This alone would put him exactly at the start of the 22% tax bracket.
  • Taxpayer needs an additional $11,774 each year to cover expenses.
  • Case A: He withdraws the additional $11,774 every year from his IRA for a total withdrawal of $46,206 (34432 + 11774) . This puts his income exactly where the maximum 85% of SS is taxed, i.e., right at the top of the hump.
  • Case B: He withdraws twice $11,774 every other year for a total withdrawal of $57,980 (34432 + 11774 + 11774). The tables [2] below show how this results in tax savings of $2,202 every two years. (11744 * 85% * 22%)

Code: Select all

Social Security 50% threshhold     25,000 ------------------>
Social Security 85% threshhold     34,000 ------------------>
Floor: ord income 10% bracket           0 ------------------>
Floor: ord income 12% bracket       9,875 ------------------>
Floor: ord income 22% bracket      40,125 ------------------>
                                   Case A    ---- Case B ----
                                   Yr 1&2    Year 1    Year 2
Social Security Benefit            35,000    35,000    35,000
Total IRA withdrawal               46,206    34,432    57,980
SS "Relevant" Income               63,706    51,932    75,480 = IRA withdrawals + 1/2 SS
50% SS taxable                      4,500     4,500     4,500 = 50% * (34000-25000)
85% SS taxable                     25,250    15,242    25,250 (see Case A & B detail below)
Total SS taxable                   29,750    19,742    29,750

Code: Select all

Adjusted gross income              75,956    54,174    87,730
Std Deduction                      14,050    14,050    14,050
Taxable Income                     61,906    40,124    73,680

Tax: ord income 10% bracket           988       988       988
Tax: ord income 12% bracket         3,630     3,630     3,630
Tax: ord income 22% bracket         4,792       -       7,382
Total tax                           9,409     4,617    12,000
                                   ---- Totals ----
                                   Case A    Case B     Diff
SS taxed                           59,500    49,492    10,008 = 11774 * 85%
Taxable income                    123,812   113,804    10,008
Total tax                          18,819    16,617     2,202 = 10008 * 22%

Code: Select all

Maximum SS taxable in 50% to 85% bracket
  25,250 =  35000 * 85% - 4500
Case A SS taxable each year
  63,706 =  46206 + 35000 / 2
  25,250 = (63706 - 34000) * 85%
       0 =  25250 - 25250 = no savings since at 85% maximum
Case B 85% SS taxable year 1
  51,932 =  34432 + 35000 / 2
  15,242 = (51932 - 34000) * 85%
Case B 85% SS taxable year 2
  75,480 =  57980 + 35000 / 2
  35,258 = (75480 - 34000) * 85%
  10,008 =  35258 - 25250 = savings since past 85% maximum
  1. To understand the so-called SS taxation "hump", see the Wiki article, Taxation of Social Security benefits. Here is an update of the example from that article for a single taxpayer with $35,000 of SS benefits instead of $20,000:

    Code: Select all

                                                  Additional
                      Adjusted                      SS Taxed
    Non SS   Taxable     Gross   Taxable      Tax   for each  Marginal
    Income        SS    Income    Income  Bracket  $1 Income  Tax Rate
    ------   -------    ------    ------  -------  ---------  --------
     7,500         0     7,500         0       0%       0.50      0.0%
    11,867     2,183    14,050         0      10%       0.50     15.0%
    16,500     4,500    21,000     6,950      10%       0.85     18.5%
    18,081     5,844    23,925     9,875      12%       0.85     22.2%
    34,432    19,743    54,175    40,125      22%       0.85     40.7% <== start of 22% bracket
    46,206    29,750    75,956    61,906      22%       0.00     22.0% <== reach max 85% SS taxed
  2. The tables were prepared using the Compare sheet of my Marginal Tax Rates Excel workbook.

Topic Author
markcoop
Posts: 1159
Joined: Fri Mar 02, 2007 8:36 am

Re: Open Social Security strategy results question

Post by markcoop » Thu May 21, 2020 11:07 am

I've been giving myself a crash course on SS and Roth conversions the past few days. Some of the info here (the wiki as well) has been great. I thank anyone who has made contributions. I do understand the SS hump in essence caused by your next dollar of income bringing in more taxable SS income. For the married filing jointly case, my plan (at least at this point) is to have enough taxable income to max out the 12% bracket (or whatever it will be in the future). If one does that, the hump considerations go away. Correct? I understand that part of my SS will be made taxable maxing to the top of the 12% bracket, but that seems like a bit of a sweet spot for me to pull out enough income and to keep taxes down. I'm expecting about $58K in SS benefits at 70 (maybe more if I delay my spouse's SS). I am hoping between SS, RMDs (I estimate between $40K and $55k) and perhaps a little more (to the top of the 12% bracket or maybe some Roth withdrawals), I can stay in the 12% bracket and withdraw enough for my living expenses. I have alot more planning to do (as I said I rally am just starting out here), but does that seem reasonable?
Mark

vtMaps
Posts: 687
Joined: Tue Oct 14, 2008 12:05 pm
Location: central Vermont

Re: Open Social Security strategy results question

Post by vtMaps » Thu May 21, 2020 12:10 pm

markcoop wrote:
Thu May 21, 2020 11:07 am
For the married filing jointly case, my plan (at least at this point) is to have enough taxable income to max out the 12% bracket (or whatever it will be in the future). If one does that, the hump considerations go away. Correct?
No, not correct. The hump is still there. In the example I gave earlier (and much refined by #Cruncher) in the high income year you blast through the hump to a lower marginal bracket, and in the low income years you may enter the hump, but not go too far into it.

--vtMaps
Historical Fact: Justin Smith Morrill represented Vermont in congress, had a dog named 'Trump', and wrote legislation establishing the Land Grant Colleges.

User avatar
#Cruncher
Posts: 2991
Joined: Fri May 14, 2010 2:33 am
Location: New York City
Contact:

Re: Open Social Security strategy results question

Post by #Cruncher » Thu May 21, 2020 1:52 pm

markcoop wrote:
Thu May 21, 2020 11:07 am
... my plan ... is to have enough taxable income to max out the 12% bracket .. If one does that, the hump considerations go away. Correct?
As vtMaps says, this isn't correct. If one has long term capital gains (LTCG) or qualified dividend income (QDI), it's even possible for additional non-SS ordinary income to be taxed at a marginal rate of almost 50% without leaving the 12% tax bracket. For example, consider the case of a 2020 Single tax return with $35,000 of SS benefits, $20,000 of LTCG and QDI, and a 65+ standard deduction. In such a case a $10,811 increase in non-SS ordinary income from $14,365 to $25,176 would increase federal tax by $5,400, a marginal rate of 49.95%.

And this happens entirely within the 12% tax bracket for ordinary income. Each $100 of non-SS ordinary income first makes $85 of SS taxable. This $185 increase in taxable ordinary income then pushes $185 of LTCG and QDI into the 15% bracket. This makes a marginal rate of 49.95% (1.85 * 12% + 1.85 * 15%).

Code: Select all

Social Security 50% threshhold    25,000 ------->
Social Security 85% threshhold    34,000 ------->
Floor: ord income 10% bracke           0 ------->
Floor: ord income 12% bracket      9,875 ------->
Floor: ord income 22% bracket     40,125 ------->
Floor: LTCG & QDI 15% bracket     40,000 ------->
Non-SS Ordinary Income            14,365   25,176
LTCG & QDI                        20,000   20,000
Social Security Benefit           35,000   35,000
SS Relevant Income                51,865   62,676
50% SS taxable                     4,500    4,500
85% SS taxable                    15,185   24,374
Total SS taxable                  19,685   28,874

Code: Select all

Adjusted gross income             54,050   74,050
Deductions plus Exemptions        14,050   14,050
Taxable Income                    40,000   60,000
LTCG & QDI Taxable                20,000   20,000
Ordinary income taxable           20,000   40,000
Taxable: ord income 12% bracket   10,125   30,125
Taxable: ord income 10% bracket    9,875    9,875
Taxable: LTCG & QDI 15% bracket      -     20,000
Taxable: LTCG & QDI  0% bracket   20,000      -  
Tax: ord income 12% bracket        1,215    3,615
Tax: ord income 10% bracket          988      988
Tax: LTCG & QDI 15% bracket          -      3,000
Total tax                          2,203    7,603

Code: Select all

Increased non-SS ord income           10,811
Increased taxable SS                   9,189
Increased tax                          5,400
Marginal SS taxable                   85.00% = 9189 / 10811
Marginal tax rate                     49.95% = 5400 / 10811
Edited 5/24/2020 to add following for the amount of non-SS ordinary income at which the 49.95% marginal tax rate begins and the amount of additional non-SS ordinary income before it ends (the "Extent").
2020 Single tax return with 65+ standard deduction for $10K, $20K, and $30K of LTCG + QDI

Code: Select all

Row              COL A       COL B   COL D       COL E   COL G       COL H   COL J
  3    Single or Joint      Single              Single              Single        
  4         LTCG & QDI      10,000              20,000              30,000        
  5   50% SS threshold      25,000              25,000              25,000        
  6   85% SS threshold      34,000              34,000              34,000        
  7     15% LTCG floor      40,000              40,000              40,000        
  8   Std deduct (65+)      14,050              14,050              14,050        
  9          AGI Begin      54,050              54,050              54,050        
 10            AGI End      64,050              74,050              84,050        
 11         SS Benefit       Begin  Extent       Begin  Extent       Begin  Extent

Code: Select all

 12             16,000      30,450              20,450              10,450        
 13             18,000      28,750              18,750               8,750        
 14             20,000      27,811     895      17,811     895       7,811     895
 15             22,000      27,351   2,355      17,351   2,355       7,351   2,355
 16             24,000      26,892   3,814      16,892   3,814       6,892   3,814
 17             26,000      26,432   5,273      16,432   5,273       6,432   5,273
 18             28,000      25,973   5,405      15,973   6,733       5,973   6,733
 19             30,000      25,514   5,405      15,514   8,192       5,514   8,192
 20             32,000      25,054   5,405      15,054   9,652       5,054   9,652
 21             34,000      24,595   5,405      14,595  10,811       4,595  11,111
 22             36,000      24,135   5,405      14,135  10,811       4,135  12,571
 23             38,000      23,676   5,405      13,676  10,811       3,676  14,030
 24             40,000      23,216   5,405      13,216  10,811       3,216  15,490
 25             42,000      22,757   5,405      12,757  10,811       2,757  16,216
 26             44,000      22,297   5,405      12,297  10,811       2,297  16,216
 27             46,000      21,838   5,405      11,838  10,811       1,838  16,216
2020 Joint tax return with two 65+ standard deductions for $10K, $20K, and $30K of LTCG + QDI

Code: Select all

Row              COL A       COL B   COL D       COL E   COL G       COL H   COL J
  3    Single or Joint       Joint               Joint               Joint        
  4         LTCG & QDI      10,000              20,000              30,000        
  5   50% SS threshold      32,000              32,000              32,000        
  6   85% SS threshold      44,000              44,000              44,000        
  7     15% LTCG floor      80,000              80,000              80,000        
  8   Std deduct (65+)      27,400              27,400              27,400        
  9          AGI Begin     107,400             107,400             107,400        
 10            AGI End     117,400             127,400             137,400        
 11         SS Benefit       Begin  Extent       Begin  Extent       Begin  Extent

Code: Select all

 12             50,000      54,900              44,900              34,900        
 13             52,000      53,200              43,200              33,200        
 14             54,000      52,622   1,320      42,622   1,320      32,622   1,320
 15             56,000      52,162   2,779      42,162   2,779      32,162   2,779
 16             58,000      51,703   4,238      41,703   4,238      31,703   4,238
 17             60,000      51,243   5,405      41,243   5,698      31,243   5,698
 18             62,000      50,784   5,405      40,784   7,157      30,784   7,157
 19             64,000      50,324   5,405      40,324   8,617      30,324   8,617
 20             66,000      49,865   5,405      39,865  10,076      29,865  10,076
 21             68,000      49,405   5,405      39,405  10,811      29,405  11,536
 22             70,000      48,946   5,405      38,946  10,811      28,946  12,995
 23             72,000      48,486   5,405      38,486  10,811      28,486  14,455
 24             74,000      48,027   5,405      38,027  10,811      28,027  15,914
 25             76,000      47,568   5,405      37,568  10,811      27,568  16,216
 26             78,000      47,108   5,405      37,108  10,811      27,108  16,216
 27             80,000      46,649   5,405      36,649  10,811      26,649  16,216
To use this for other amounts of LTCG + QDI, Select All, Copy, and Paste [ * ] the following at cell A3 of a blank Excel sheet and then copy down the formulas in cells A14:D14.

Code: Select all

Single or Joint	Single
LTCG & QDI	10000
50% SS threshold	=IF(B$3="Single",25000,32000)
85% SS threshold	=IF(B$3="Single",34000,44000)
15% LTCG floor	=IF(B$3="Single",40000,80000)
Std deduct (65+)	=IF(B$3="Single",14050,27400)
AGI Begin	=B7+B8
AGI End	=B9+B4
SS Benefit	Begin	End	Extent
18000	=IF($A12>2*(B$6-B$9+0.5*(B$6-B$5)),IF(0.5*(B$6-B$5)+0.85*((B$9+0.5*B$5+0.35*B$6-0.425*$A12)/1.85+$A12/2-B$6)>0.85*$A12,B$9-0.85*$A12,(B$9+0.5*B$5+0.35*B$6-0.425*$A12)/1.85),IF($A12>2*(B$5-B$9),(B$9-$A12*0.25+0.5*B$5)/1.5,B$9))-B$4	=MIN((0.425*$A12-(B$6-B$5)/2+0.85*B$6)/0.85,IF($A12>2*(B$6-B$10+0.5*(B$6-B$5)),IF(0.5*(B$6-B$5)+0.85*((B$10+0.5*B$5+0.35*B$6-0.425*$A12)/1.85+$A12/2-B$6)>0.85*$A12,B$10-0.85*$A12,(B$10+0.5*B$5+0.35*B$6-0.425*$A12)/1.85),IF($A12>2*(B$5-B$10),(B$10-$A12*0.25+0.5*B$5)/1.5,B$10)))-B$4	=MAX(0,C12-B12)
20000	=IF($A13>2*(B$6-B$9+0.5*(B$6-B$5)),IF(0.5*(B$6-B$5)+0.85*((B$9+0.5*B$5+0.35*B$6-0.425*$A13)/1.85+$A13/2-B$6)>0.85*$A13,B$9-0.85*$A13,(B$9+0.5*B$5+0.35*B$6-0.425*$A13)/1.85),IF($A13>2*(B$5-B$9),(B$9-$A13*0.25+0.5*B$5)/1.5,B$9))-B$4	=MIN((0.425*$A13-(B$6-B$5)/2+0.85*B$6)/0.85,IF($A13>2*(B$6-B$10+0.5*(B$6-B$5)),IF(0.5*(B$6-B$5)+0.85*((B$10+0.5*B$5+0.35*B$6-0.425*$A13)/1.85+$A13/2-B$6)>0.85*$A13,B$10-0.85*$A13,(B$10+0.5*B$5+0.35*B$6-0.425*$A13)/1.85),IF($A13>2*(B$5-B$10),(B$10-$A13*0.25+0.5*B$5)/1.5,B$10)))-B$4	=MAX(0,C13-B13)
=2*A13-A12	=IF($A14>2*(B$6-B$9+0.5*(B$6-B$5)),IF(0.5*(B$6-B$5)+0.85*((B$9+0.5*B$5+0.35*B$6-0.425*$A14)/1.85+$A14/2-B$6)>0.85*$A14,B$9-0.85*$A14,(B$9+0.5*B$5+0.35*B$6-0.425*$A14)/1.85),IF($A14>2*(B$5-B$9),(B$9-$A14*0.25+0.5*B$5)/1.5,B$9))-B$4	=MIN((0.425*$A14-(B$6-B$5)/2+0.85*B$6)/0.85,IF($A14>2*(B$6-B$10+0.5*(B$6-B$5)),IF(0.5*(B$6-B$5)+0.85*((B$10+0.5*B$5+0.35*B$6-0.425*$A14)/1.85+$A14/2-B$6)>0.85*$A14,B$10-0.85*$A14,(B$10+0.5*B$5+0.35*B$6-0.425*$A14)/1.85),IF($A14>2*(B$5-B$10),(B$10-$A14*0.25+0.5*B$5)/1.5,B$10)))-B$4	=MAX(0,C14-B14)
* If you have trouble pasting, try "Paste Special" and "Text".
Last edited by #Cruncher on Sun May 24, 2020 6:02 am, edited 1 time in total.

Topic Author
markcoop
Posts: 1159
Joined: Fri Mar 02, 2007 8:36 am

Re: Open Social Security strategy results question

Post by markcoop » Thu May 21, 2020 10:00 pm

Let me try to make my point a different way. My amount of non-SS other income needed to make the maximum 85% of Social Security taxable for my SS amount of $58K is around $66K (if my wife takes her SS at 62) or around $71K (if my wife takes her SS at 70). So if I have that much non-SS other income, I am past the hump. My RMD is between $40K - $65K. However, if I expect I will need more money for living expenses, my options would be to (a) take more money from 401K (possibly to the top of the 12% bracket), (b) remove from taxable if I have it or (c) possibly withdraw from Roth sources. Depending on the size of the RMD, I may already be close to being past the hump. Also, depending on income from taxable, that alone could push me past the point of the hump. So, I was trying to say in my previous post that if I'm going to be already past the point of the hump, it may make sense to just withdraw (or convert to Roth) to the top of the 12% bracket. Does that make sense? If I didn't have any other income besides the RMD and the RMD was closer to $40K, then perhaps it would make sense not to penetrate any deeper into the 12% bracket if possible because I would be going right into the hump.
Mark

User avatar
FiveK
Posts: 9336
Joined: Sun Mar 16, 2014 2:43 pm

Re: Open Social Security strategy results question

Post by FiveK » Thu May 21, 2020 10:26 pm

markcoop wrote:
Thu May 21, 2020 10:00 pm
So, I was trying to say in my previous post that if I'm going to be already past the point of the hump, it may make sense to just withdraw (or convert to Roth) to the top of the 12% bracket. Does that make sense?
Yes.

If your "unavoidable" income puts you past the hump, such that your marginal rate on "optional" income is 12%, opting to pay 12% is likely a good idea.

vtMaps
Posts: 687
Joined: Tue Oct 14, 2008 12:05 pm
Location: central Vermont

Re: Open Social Security strategy results question

Post by vtMaps » Fri May 22, 2020 4:17 am

markcoop wrote:
Thu May 21, 2020 10:00 pm
So, I was trying to say in my previous post that if I'm going to be already past the point of the hump, it may make sense to just withdraw (or convert to Roth) to the top of the 12% bracket. Does that make sense?
No. With SS = $58k, if you are past the hump, you are past the 12% tax bracket.

With SS = $58k, the worst of the hump (40.7%) is so narrow that it is not much of a consideration. Your non SS income above about $20k will be taxed at about 22%. Non SS income up to about $20k is untaxed. If you can do enough Roth conversion to get your RMD down to $20k, you may never pay taxes on any income again.

I think you should convert up to the top of the 22% bracket. That means converting now at 22% to avoid paying 22% later. The reason to do this is because you are married. Look what happens to the tax hump after one of you dies, and the single survivor, with just a survivor SS benefit, has a RMD of $50k. There's a lot of unavoidable 40.7% marginal bracket!

--vtMaps
Historical Fact: Justin Smith Morrill represented Vermont in congress, had a dog named 'Trump', and wrote legislation establishing the Land Grant Colleges.

petulant
Posts: 1329
Joined: Thu Sep 22, 2016 1:09 pm

Re: Open Social Security strategy results question

Post by petulant » Fri May 22, 2020 7:14 am

FiveK wrote:
Thu May 21, 2020 10:26 pm
markcoop wrote:
Thu May 21, 2020 10:00 pm
So, I was trying to say in my previous post that if I'm going to be already past the point of the hump, it may make sense to just withdraw (or convert to Roth) to the top of the 12% bracket. Does that make sense?
Yes.

If your "unavoidable" income puts you past the hump, such that your marginal rate on "optional" income is 12%, opting to pay 12% is likely a good idea.
This is true in any specific year, as FiveK agrees with you, but it's not necessarily the correct view of the big picture. In the big picture, what you're saying is you expect in most years to have taxable income that puts you close to or over the SS hump. In that view, the tax consequences of SS matter little, so you can take it whenever you want. However, vtMaps is saying you have an opportunity to structure your income so that you don't have to have taxable income putting you close to or over the SS hump each year. Imagine you had $1,000,000 in tax-deferred accounts, with both you and your spouse retired at age 62. You will actually minimize taxes by both delaying SS to age 70 and withdrawing or Roth converting all of your tax-deferred accounts each year to the top of the 12% bracket or even into the 22% bracket below IRMAA until age 70 with the result that after age 70, your RMDs would be much lower and SS may not be anywhere close to the hump. The result would be no or little income taxes owed during your highest SS years, with any additional needs taken from the (now large) Roth account. Although your specifics might be a little different, some version of this is the optimal strategy for the majority of Americans. It is not to your benefit to keep "flexibility" or anything with a large tax-deferred account.

Post Reply