"Best" withdrawal decisions

[baldeagle]

We are a couple of 63-year-olds who retired nearly three years ago, and took SS at age 62. I have always felt that financial decisions made in the ten years between when we retired and age 70.5, when we cede withdrawal control to MRDs, could be significant for the rest of our lives. In particular, the decisions of account withdrawal sequencing, Roth conversion, and SS repay/restart seemed most important. But what decision is “best?”

This post describes what we did to figure it out for our situation. Have any of you done something similar? Did you come to similar conclusions?

OUR PORTFOLIO

We have a 60/40 portfolio comprised of about 85% IRA, 15% taxable, and 5% Roth. Our living expenses require a withdrawal rate of about 3.2%. We are well within the 15% FIT bracket. Our current WD approach has been:

• Living expenses from IRA
• Taxes and extraordinary expenses from taxable account (for as long as it survives, then switch to Roth)
• Convert IRA to Roth up to the top of the 15% bracket for as long as we can

Drawing from IRA and converting to Roth has been aimed at getting the IRA values down so that RMD amounts are down when they hit.

ANALYSIS

I build a spreadsheet model to see how different scenarios would play out over the next 40 years, accounting for inflation and taxes.

Evaluation criteria

• Total portfolio value over time (most important)
• Maximum Roth account value
• Minimum excess RMDs (i.e., minimum required WDs over what we needed)
• Minimum incursion into 25% bracket over the years

Criteria 3 and 4 are viewed as hedges against expected future tax increases.

Scenarios - Five were chosen. They are variations of our current approach:

• Continue as is
• Abandon Roth conversion – just stay low in 15% bracket until age 70.5
• Repay SS, restart at 66 and Roth convert to top of 15% bracket
• Repay SS, restart at 70 and Roth convert to top of 15% bracket
• WD expenses from taxable, Roth-convert to top of 15% bracket

Calculations – For every year from now to age 100, the model computes the following, adjusting for inflation and investment return:

• Cash flow (SSI, pension, taxes, spending)
• Conversions, withdrawals, RMDs
• Tax brackets, deductions, exemptions
• Account values (taxable, IRA, Roth, total)

Economic sensitivity – To assess scenario sensitivity to economic conditions, I ran the model in three conditions:

• Mid – 7% return, 3% inflation
• Low – 6% return, 4% inflation
• High – 8% return, 2% inflation

CONCLUSIONS

The charts show that for our specific situation, scenario 5 is always better than our current approach. It is also the best scenario of all for Mid and High economic conditions. Only in Low economic conditions does SS repay/restart become best, but that takes until age 87 to happen.

We will change our approach to scenario 5.

 

[ziggy29]

I think you're definitely on the right track, and that concept is what I plan to do when I get there. Yes, the tax brackets will change but the principle might not.

Given today's tax brackets, I would definitely recommend that anyone with a significant traditional IRA or 401K plan use up as much of the 15% bracket as possible, even if they don't need the money. People with large IRAs could otherwise be forced into RMDs at age 70 1/2 that kick them into what is now the 25% bracket.

And if you have the cash outside of retirement accounts and you don't need the money from the IRA withdrawals, Roth conversions might even be a better choice since they still compound tax-deferred and don't trigger RMDs.

I'm trying to build a three-prong retirement savings/investment plan with traditional 401K/IRAs, Roths and taxable accounts, since I think that gives the maximum flexibility in "engineering" income to maximize withdrawals at low tax brackets while being able to get more money if needed from sources that won't kick you into a higher one.

 

[Rich_by_the_Bay]

Repaying SS may not be an option down the road; once it catches on it may disappear.

You might want to search here and elsewhere about sequence of withdrawals within your nest egg. Several studies seem to support burning through fixed equities and cash almost completely, then reallocating your stock nest egg when all that is gone. It's counterintuitive in some ways, but the references and rationale are buried in some of those threads. I'll post a link tomorrow if I have time to locate them.

 

[Midpack]

You might want to search here and elsewhere about sequence of withdrawals within your nest egg. Several studies seem to support burning through fixed equities and cash almost completely, then reallocating your stock nest egg when all that is gone. It's counterintuitive in some ways, but the references and rationale are buried in some of those threads. I'll post a link tomorrow if I have time to locate them.

If you find it, I'd be interested too. I'm thinking about buying ESPlanner to figure out same...

 

[OAG]

To some extent that is what we have done. Now 70/67 with DW starting RMD this year. We did however, abandon the TIRA to ROTH conversions after getting about 25% of IRA to ROTH, mainly because we could stay in the 15% bracket (unless some drastic change is made to the tax tables). Additionally, a significant COLA'd Retirement Account sort of precludes, to some extent, the ability to "engineer" taxes. The SS redo remains an option but may not be the best approach for us as it would almost guarantee the 25% tax bracket. In any event your plan and options seem very well thought out and should work out just fine for you. I do think, over the next few years, the ability to remain flexible will be a major key to success. BTW am I reading your portfolio wrong (85, 15, 5)? Oh, maybe the 5 Roth is included in the 15 IRA)!

 

[Rich_by_the_Bay]

Is Rebalancing a Portfolio During Retirement Necessary?
by John J. Spitzer, Ph.D., and Sandeep Singh, Ph.D., CFA
Journal of Financial Planning, June 2007

Journalfp.net

 

[REWahoo]

Rich, did you intend to include a link to the article in your previous post? If so, it isn't working.

 

[donheff]

Anybody else who is already at the 25% bracket done these calculations? I have the luxury of a substantial Federal pension - but it is almost all taxable. DW will be coming online with a big 401K/IRA and I have a TSP plan that will also be fully taxable. We assumed the only sensible thing will be to burn through the taxed accounts during the next ten years or so and then just switch to fully taxable income (IRAs/pension) from then on out. We view DWs social security as the "tax equalizer" in our calculations. In other words, SS will pay the additional tax we will get stuck with when we go fully taxable.

Are the attached spreadsheets case specific or are they adaptable enough that someone in a significantly different situation like me should download them?

 

[Rich_by_the_Bay]

I'm having problems getting this to display on my current setup:

Article

If you have the same, click the link for PDF version at the bottom of the window.

 

[baldeagle]

Thanks, all, for your helpful comments.

Zigg29 and RWood -- I like your use of the word "engineer" WRT to income and/or taxes. That is exactly what I am trying to do.

RWood -- Yes, you found a math error in the percentages of portfolio across the three account locations. I should have said 80/15/5. Duh.

Also, I agree with your comment about SS redo not always being a good option. In my spreadsheet model, I will be in the 25% bracket by age 80 no matter what scenario I choose. But if I do a SS redo, about 2.5 times as much income will be in that 25% bracket. Pros and cons, eh? Safer income from SS if you wait/redo, but more expensive income.

Donheff -- The attachments were PDF files made from the spreadsheets, not the actual spreadsheets themselves. I tried to keep the spreadsheets general where possible, but, to keep the formulas simpler, I made some simplifications appropriate to my case -- e.g., never in less than 10% FIT bracket and never more than 25% bracket. Plus, they may not be very
obvious to someone who didn't write them!

 

[FIRE'd@51]

Repaying SS may not be an option down the road; once it catches on it may disappear.

This comment keeps coming up in discussions about restarting SS. My guess is that SS, in allowing this in the first place, has taken into account actuarially the fact that many who restart will die before they reach the new break-even point.

 

[jIMOh]

Anybody else who is already at the 25% bracket done these calculations? I have the luxury of a substantial Federal pension - but it is almost all taxable. DW will be coming online with a big 401K/IRA and I have a TSP plan that will also be fully taxable. We assumed the only sensible thing will be to burn through the taxed accounts during the next ten years or so and then just switch to fully taxable income (IRAs/pension) from then on out. We view DWs social security as the "tax equalizer" in our calculations. In other words, SS will pay the additional tax we will get stuck with when we go fully taxable.

Are the attached spreadsheets case specific or are they adaptable enough that someone in a significantly different situation like me should download them?

One distinct advantage is capping the Roth conversion in 25% tax bracket will leave more wiggle room. Higher taxes, but you might be able to escpae the RMD issue more by converting earlier and converting in 25% bracket.

If you can live off of income in 15% tax bracket (~62k), this would imply you could convert double that (~120k is cap of 25% bracket).

 

[jIMOh]

We are a couple of 63-year-olds who retired nearly three years ago, and took SS at age 62. I have always felt that financial decisions made in the ten years between when we retired and age 70.5, when we cede withdrawal control to MRDs, could be significant for the rest of our lives. In particular, the decisions of account withdrawal sequencing, Roth conversion, and SS repay/restart seemed most important. But what decision is “best?”


We have a 60/40 portfolio comprised of about 85% IRA, 15% taxable, and 5% Roth. Our living expenses require a withdrawal rate of about 3.2%. We are well within the 15% FIT bracket. Our current WD approach has been:

• Living expenses from IRA
• Taxes and extraordinary expenses from taxable account (for as long as it survives, then switch to Roth)
• Convert IRA to Roth up to the top of the 15% bracket for as long as we can

Drawing from IRA and converting to Roth has been aimed at getting the IRA values down so that RMD amounts are down when they hit.

Have you done an analysis which included the following:

1) cap the Roth conversion once or twice in 25% tax bracket. Reason being is if done in current year and maybe one other year, the taxes paid will be lower (relative to 25% tax bracket being increased to a higher percentage or minimum being lowered to a lower income level).

2) The higher conversion amounts now might make RMD so low that you have more room to convert to Roth after 70.5.

 

[baldeagle]

jIMoh -- Good point. I modeled this also, converting deep into the 25% bracket, even though we live comfortably under the 15% bracket. I didn't publish the results in my original post, but I can share them below. The two scenarios I modeled were:

1. Convert as much as necessary as soon as possible to keep MRDs from exceeding required spending beginning at age 70.5 and lasting until age 100. This definitely put me in the 25% bracket for a while.

2. Ditto except goal was to stay out of 25% bracket for life after age 70.5

These two scenarios underperformed my described winners in Mid economic conditions, somewhat overperformed in High conditions, and significantly underperformed in Low conditions. So, for my situation, I eliminated them. Your mileage may vary...

 

[baldeagle]

jIMho -- our posts just crossed. Mine was in response to yours to Donheff. After dinner, I'll read your latest and respond.

 

[baldeagle]

jIMho -- Actually, my earlier answer is somewhat relevant. In particular, I found that for my two "winner" scenarios, I could convert all the way up to age 80 and still avoid the 25% bracket. After that, however, I was stuck in it forever.

I didn't run a scenario of a few heavy early conversions, though it's easy enough to do, but it took really big annual conversions (filling nearly half the 25% bracket) for 7 years before age 70.5 in order to stay out of the 25% bracket thereafter.

 

[megacorp-firee]

Quote:
Originally Posted by Rich_in_Tampa
Repaying SS may not be an option down the road; once it catches on it may disappear.

This comment keeps coming up in discussions about restarting SS. My guess is that SS, in allowing this in the first place, has taken into account actuarially the fact that many who restart will die before they reach the new break-even point.

IMO, there will not be that many takers for repaying and resetting SS. Most people (members of this board excepted) cannot put together enough cash to pay for a short vacation, much less 4 to 8 years of SS payments.
Also, I agree with FIRE'd@51 SS has the math worked out. To them it's a wash.

 

[megacorp-firee]

I am starting to develop a love/hate relationship re: this board.
There are some good thinkers and a lot of great information here. The support is also wonderful.
However, sometimes I get overwhelmed with the amount of things one 'should' be considering as well as the alternatives. This thread is an example. Without exceeding the precision of the model, I have also tried to analyze how to take withdrawals and am getting a bit frustrated.
OK, venting is over ... here is my situation;

If I don't convert 401k to IRA to Roth, then at 70 1/2 (because of fully taxable pension and my current fixed income investments) I will be pushed into the 28% tax bracket (which is where I was when w*rking).
With my pension and fixed income and no conversion I am in the upper 1/3 of the 15% bracket.
I have run a ss with different scenarios (amounts to convert) and have found that if I convert a 'considerable' amount, I will end up in the 25% bracket (about the beginning of the upper 1/3 again). RMD will also continue @ 70 1/2 at approximately the same withdrawal and tax rate.

Benefit of converting is that, by 70 1/2:
I will have about 1/2 my 401k/IRA converted to Roth and therefore growing tax free forever

Drawbacks of converting:
a) I am paying more taxes up front, than I would if I did not convert.
b) Cumulative tax breakeven seems to be at age 90, when my tax savings start kicking in.
c) my (ballpark) analysis shows the value of the portfolio under condition i) 'doing nothing' is always greater than option ii) Roth conversions starting immediately. Even though at age 70 1/2 I am paying a fortune in fed income taxes due to rmd amount.

Conclusion: I think (he says hesitantly) that I have found the same as baldeagle. Roth conversion is not the best bet for those with substantial taxable pensions.

... but I am so unsure of this ....

Here is the part where that is frustrating me. I am not sure I have accounted for all of the RELEVANT (rather than clutter) variables. I did not include inflation, since it is relatively the same for both scenarios. I did not take in consideration NPV, since I don't remember how to ... and I am not sure that it is relevant (again relativity). In any case, NPV would benefit not paying more taxes up front I would think.
I also am thinking I don't know what I don't know.

Critiques please. Thank you.

 

[kaudrey]

megacorp - I am a ways from FIRE and I haven't done super-detailed analysis, but in my reading (here, and elsewhere), and in discussions with my father (FIRED early, has a pension, has a big IRA), I think your conclusion is valid.

Of course everyone's situation is different, but with a sizeable taxable pension, you won't be able to convert much, if any, at the lower tax brackets, so your tax savings won't be very much over the years, and you'll likely still have a large IRA balance when RMDs start.

I plan to look into this more when I get closer to FIRE, but for now, I have high taxes built into my spreadsheet of future income needs.

 

[jIMOh]

jIMoh -- Good point. I modeled this also, converting deep into the 25% bracket, even though we live comfortably under the 15% bracket. I didn't publish the results in my original post, but I can share them below. The two scenarios I modeled were:

1. Convert as much as necessary as soon as possible to keep MRDs from exceeding required spending beginning at age 70.5 and lasting until age 100. This definitely put me in the 25% bracket for a while.

2. Ditto except goal was to stay out of 25% bracket for life after age 70.5

These two scenarios underperformed my described winners in Mid economic conditions, somewhat overperformed in High conditions, and significantly underperformed in Low conditions. So, for my situation, I eliminated them. Your mileage may vary...

I was concentrating more on something between 1 and 2.

Convert as much as necessary as soon as possible.- model this with a 1 year withdraw for example- to cap out 25% bracket- amount x. Then model with a 1 year withdraw which caps out 28% bracket-amount y.

Then compare this to only capping out 25% bracket and how long it would take to get $y converted in 25% bracket.

See next post.

 

[jIMOh]

jIMho -- Actually, my earlier answer is somewhat relevant. In particular, I found that for my two "winner" scenarios, I could convert all the way up to age 80 and still avoid the 25% bracket. After that, however, I was stuck in it forever.

I didn't run a scenario of a few heavy early conversions, though it's easy enough to do, but it took really big annual conversions (filling nearly half the 25% bracket) for 7 years before age 70.5 in order to stay out of the 25% bracket thereafter.

More than likely you did not account for inflating the tax bracket cap? Or did you?

If you could stay in 15% bracket until age 80, I would take those chances.

Compare what you did in second comment ("a few heavy early conversions") and make them heavier, sooner.

Then run another model which does a huge conversion when you are aged 69 (convert at last possible year before turning 70.5) and again make this a single heavy conversion- maybe to extent of capping out 28% tax bracket.

Make a decision- is it better to pay into 28% tax bracket for 1-5 years (to convert) so you can stay in 15% bracket there after, or better to stay in 15% tax bracket now, moving into 25% bracket at age 70.5.

Pay taxes now or pay later?

 

[jIMOh]

I am starting to develop a love/hate relationship re: this board.
There are some good thinkers and a lot of great information here. The support is also wonderful.
However, sometimes I get overwhelmed with the amount of things one 'should' be considering as well as the alternatives. This thread is an example. Without exceeding the precision of the model, I have also tried to analyze how to take withdrawals and am getting a bit frustrated.
OK, venting is over ... here is my situation;

If I don't convert 401k to IRA to Roth, then at 70 1/2 (because of fully taxable pension and my current fixed income investments) I will be pushed into the 28% tax bracket (which is where I was when w*rking).
With my pension and fixed income and no conversion I am in the upper 1/3 of the 15% bracket.
I have run a ss with different scenarios (amounts to convert) and have found that if I convert a 'considerable' amount, I will end up in the 25% bracket (about the beginning of the upper 1/3 again). RMD will also continue @ 70 1/2 at approximately the same withdrawal and tax rate.

Benefit of converting is that, by 70 1/2:
I will have about 1/2 my 401k/IRA converted to Roth and therefore growing tax free forever

Drawbacks of converting:
a) I am paying more taxes up front, than I would if I did not convert.
b) Cumulative tax breakeven seems to be at age 90, when my tax savings start kicking in.
c) my (ballpark) analysis shows the value of the portfolio under condition i) 'doing nothing' is always greater than option ii) Roth conversions starting immediately. Even though at age 70 1/2 I am paying a fortune in fed income taxes due to rmd amount.

Conclusion: I think (he says hesitantly) that I have found the same as baldeagle. Roth conversion is not the best bet for those with substantial taxable pensions.

... but I am so unsure of this ....

Here is the part where that is frustrating me. I am not sure I have accounted for all of the RELEVANT (rather than clutter) variables. I did not include inflation, since it is relatively the same for both scenarios. I did not take in consideration NPV, since I don't remember how to ... and I am not sure that it is relevant (again relativity). In any case, NPV would benefit not paying more taxes up front I would think.
I also am thinking I don't know what I don't know.

Critiques please. Thank you.

megacorp- how from from doing this are you (time wise?). It might make more sense to get money in a taxable account, because income from long term capital gains and dividends do not increase taxable income (tax bracket).

Example 1:

If you withdraw 180k from an IRA, a portion of the 180k is taxed at 28% level.

Example 2 if you withdraw 62k from IRA/Pension, that is taxed at 15% level. If you supplement this with 120k of income from Roths, taxable accounts and similar (which do not increase taxable income) then the taxable accounts are taxed at 5% (because you are in 15% bracket based on taxable income).

Example 2 requires some planning- you need to set aside money in taxable accounts to maintain an income stream which is larger than the IRA/pension stream.

 

[baldeagle]

jIMOh -- Catching up on your comments...

Yes, my model does account for inflation. I can change it for every year if I want, but in the runs I reported I used the same amount every year in each scenario. Then I then ran three scenarios.

From a tax point of view, every year I inflated all three tax brackets -- 10, 15, and 25% -- and inflated deductions and exemptions as well. I also inflated annual income requirements and SS payments. I don't think I left anything out.

I tried your idea of a few big conversions, one early and one late. The later one at age 69 was deep into the 28% bracket, near the top. However, it did not improve over taking smaller, steady conversions near the top of the 25% bracket from age 63 until age 70.5. By "improve" I mean I eventually got back into the 25% bracket again and stayed there.

My success metrics for this exercise were to determine what conversion amounts, and when, gave portfolio the greatest value in good times and bad. For my particular situation, there seems to be a "sweet spot" in all this. It seems to be made of:

1. Convert as much as possible as early as possible
2. But not over the 15% bracket
3. Because the additional taxes (nearly double the rate) take too big a bite too early
4. Making it take too long for the portfolio to recover

 

[megacorp-firee]

megacorp- how from from doing this are you (time wise?). It might make more sense to get money in a taxable account, because income from long term capital gains and dividends do not increase taxable income (tax bracket).

Example 1:

If you withdraw 180k from an IRA, a portion of the 180k is taxed at 28% level.

Example 2 if you withdraw 62k from IRA/Pension, that is taxed at 15% level. If you supplement this with 120k of income from Roths, taxable accounts and similar (which do not increase taxable income) then the taxable accounts are taxed at 5% (because you are in 15% bracket based on taxable income).

Example 2 requires some planning- you need to set aside money in taxable accounts to maintain an income stream which is larger than the IRA/pension stream.

JiMoh, I'm now FIREd (since last June). I was looking at doing conversions starting this year. However, as I look at this closer, it seems that with my taxable pension, I can't convert much without driving myself into higher tax brackets (25%).

I don't fully understand your example #2. How do you get a 5% tax if you're taking 120K out of ROTH (no tax), taxable accounts?

My 401k represents about 2/3 of my portfolio. I don't have the luxury of large taxable account funds. So it looks like I'm locked into a NON Roth conversion strategy.

 

[jIMOh]

JiMoh, I'm now FIREd (since last June). I was looking at doing conversions starting this year. However, as I look at this closer, it seems that with my taxable pension, I can't convert much without driving myself into higher tax brackets (25%).

I don't fully understand your example #2. How do you get a 5% tax if you're taking 120K out of ROTH (no tax), taxable accounts?

My 401k represents about 2/3 of my portfolio. I don't have the luxury of large taxable account funds. So it looks like I'm locked into a NON Roth conversion strategy.

Long term capital gains and dividends are taxed at 5% (under present tax code) for people in 10% and 15% tax brackets.

As for the conversions- it's a tax trade off/ task risk issue. Maybe conversions now have you pay a 25% tax- what tax will be paid on RMDs?

If RMDs push income into 28% tax bracket, then paying 25% now isn't so bad.