[**Update] - Metrics to Compare Roth Conversion Scenarios ??

Non-investing personal finance issues including insurance, credit, real estate, taxes, employment and legal issues such as trusts and wills.
User avatar
Topic Author
WoodSpinner
Posts: 3499
Joined: Mon Feb 27, 2017 12:15 pm

[**Update] - Metrics to Compare Roth Conversion Scenarios ??

Post by WoodSpinner »

[**Update - Aug 3, 2021]
I have found the interchange in this thread very valuable to help crystalize my thinking around the appropriate metrics and to narrow down my alternatives. I hope this update proves helpful and will allow readers to more quickly parse the discussion and its results.

Key Learnings:
  • Start with a well defined understanding of your Roth Conversion Goals
  • A robust tool or model is essential for understanding the nuances of your Income, Expenses, Cash-Flow, Taxes, Asset Growth, Asset Location etc. there are a lot of moving parts that need to be considered.
  • Conversion planning needs to be an iterative process. Taxes, Asset Growth, Unexpected Death, Inflation, Health issues etc. can all change your plan. Suggest reviewing the plan at least annually to fine tune and align.
  • A simple line plot of the Cumulative Taxes Paid serves as a great way to compare various scenarios and to determine how closely Tax Equilibrium is achieved.
  • The Widow(er) Tax Penalty is real and you need to understand what if any impact this will have on your Retirement Plans.
  • When in doubt follow McQuarrie's Rule of Thumb for conservative Roth Conversions
Current info:
  • Ages: 62 (me), Spouse (61), MFJ, currently retired.
  • 2 Pension(s)
  • Marginal Brackets (ages 62-71, before RMDs)
    • Federal 12% (15% after 2026), SS taxed at 85%
    • State 4%
  • Portfolio (Taxable 0%, Roth 18%, IRA 82%)
Roth Conversion Goals (prioritized)
  • Provide Resiliency for our retirement to deal with unexpected opportunities or expenses
  • Fund LTC needs for my wife (hopefully never needed) from our IRA.
  • Inheritance, my Heirs are likely to be in the 12% Marginal Bracket
  • Fund Charitable Giving via QCDs and Bequests from our IRA.
Scenarios Considered:
  1. No Conversions
    Rejected since it fails the Tax Equilibrium test.
  2. Top of IRMAA Tier0 (Selected Alternative)
    The Scenario is very attractive. It will meet my Roth Conversion Goals and is Tax Costs are Lower than IRMAA Tier1-3 or No Conversions. In addition this passes the Tax Equilibrium Test and will deliver a Total Adjusted Portfolio that will meet my Roth Goals
  3. Top of IRMAA Tier1
    The scenario is very attractive. Tax Costs are lower as well as Total Adjusted Portfolio Value. OTOH it will meet my Roth Conversion goals.
  4. Top of IRMAA Tier2
    The scenario is very attractive. Tax Costs are higher as well as Total Adjusted Portfolio Value. It will also meet my Roth Conversion goals.
  5. Top of IRMAA Tier3
    Rejected since it fails the Tax Equilibrium test.
  6. Top of 24% Bracket
    Rejected since it fails the Tax Equilibrium test.
  7. Balance AGI
    Rejected since it would require Conversions on top of RMDs and comes at a slightly higher Tax Cost (as compared to IRMAA Tier1 or Tier2).
  8. Balance Effective Tax
    Rejected since it fails the Tax Equilibrium test.
Metrics Considered:
  1. Marginal Tax Rates vs. Future Marginal Rates (Fed + CA State)
    This is a general useful metric -- it rarely makes sense to pay Higher Taxes now when you can pay Lower Taxes later. This is especially true if you will be leaving the IRA to beneficiaries in a lower Tax Bracket or to Charities.
  2. Tax Equilibrium across Retirement (see Kitces Blog
    This is a very useful metric to consider. My approach to using this metric is to plot the graphs of Total Cumulative Taxes paid over time (Fed, IRMAA, NIT, State etc.) and look at the slopes of the lines and any obvious bend points. This will highlight Conversion scenarios that are wasting lower marginal tax space or paying more in taxes than future income years.
  3. Total Tax Adjusted Portfolio Value
    This is a very useful metric with some caveats. You need to know the Marginal Tax Rate for each scenario on a Year-by-Year basis for useful results. This makes adjusting the values of an IRA fairly easy. However adjusting the value of the Taxable accounts is difficult to model -- this was not implemented in my Retirement Model.
  4. Value of IRA over time
    This can be a useful metric depending on your Retirement goals (e.g. you may want to leave Assets in the IRA):
    1. Self Funding Long Term Care from your IRA
    2. Charitable bequests from your IRA
    3. Inheritance by beneficiaries in a lower marginal bracket than you are
  5. Total Portfolio Value
    Not generally a useful metric. It does not properly take into consideration the deferred taxes in the IRA.
  6. Total Taxes paid (Fed +IRMAA + State)
    Not generally a useful metric. However it can be used to better understand at what age various scenarios break-even from a taxes paid perspective. My findings showed this was in the 84-88 year old range -- so its a long-term tax arbitrage play.
  7. Return on Investment (Tax Adjusted Portfolio Delta/ Total Taxes Paid Delta (Scenario compared to No Conversions)
    I had high hopes for using this as a metric and the more I investigated the less I believed in it. I think it muddied the waters more than it helped clearly identify a solution.
  8. Internal Rate of Return ( Tax Adjusted Portfolio Delta/ Total Taxes Paid Delta (Scenario compared to No Conversions)
    I had high hopes for using this as a metric as well and the more I investigated the less I believed in it as well.
Midway through this discussion I shifted my approach to better focus on the Metrics that I believe have the most value.

Consider starting here first to avoid some confusion.

A special thanks to all of the participants who posted questions, concerns, challenged my thinking and helped generate a useful discussion. I really appreciate the help!

********************* End of Update ****************


I have been doing quite a bit of modeling on Roth Conversions with a goal of identifying the most optimal scenario (given the constraints of predicting the future). A bit puzzled on which metrics to use to compare the various scenarios to provide some insights into which are better than others.

What I am seeing is that the various Metrics simply aren't working as an effective tool to make a selection. There is significant variation based on time period (e.g. 71-80 vs. 81-90. I am also getting conflicting signals from the various metrics as to which is the better scenario.

At this point I need some advice on which metric(s) I should really be prioritizing.

Background:
  • Current info:
Ages: 62 (me), Spouse (61), MFJ
Marginal Brackets (ages 62-71, before RMDs)
Federal 12% (15% after 2026), SS taxed at 85%
State 4%

Scenarios:
1. No Conversions
2. Top of IRMAA Tier2
3. Top of IRMAA Tier3
4. Top of 24% Bracket
5. Balance AGI
6. Balance Effective Tax


Metrics Considered:
  • Marginal Tax Rates vs. Future Marginal Rates (Fed + CA State)
  • Total Portfolio Value
  • Total Tax Adjusted Portfolio Value
  • Total Taxes (Fed +IRMAA + State) paid
  • Return on Investment (Tax Adjusted Portfolio Delta/ Total Taxes Paid Delta (Scenario compared to No Conversions)
  • Internal Rate of Return ( Tax Adjusted Portfolio Delta/ Total Taxes Paid Delta (Scenario compared to No Conversions)
I have looked at these metrics at various time periods (62-70, 71-80, 81-90, 62-90).

Would appreciate any insights or thoughts on how to make a selection!

Thanks in Advance

WoodSpinner
Last edited by WoodSpinner on Thu Aug 05, 2021 9:40 am, edited 11 times in total.
WoodSpinner
User avatar
Nestegg_User
Posts: 2109
Joined: Wed Aug 05, 2009 1:26 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Nestegg_User »

WoodSpinner wrote: Sat Jul 24, 2021 8:42 pm I have been doing quite a bit of modeling on Roth Conversions with a goal of identifying the most optimal scenario (given the constraints of predicting the future). A bit puzzled on which metrics to use to compare the various scenarios to provide some insights into which are better than others.

What I am seeing is that the various Metrics simply aren't working as an effective tool to make a selection. There is significant variation based on time period (e.g. 71-80 vs. 81-90. I am also getting conflicting signals from the various metrics as to which is the better scenario.

At this point I need some advice on which metric(s) I should really be prioritizing.

Background:
  • Current info:
Ages: 62 (me), Spouse (61), MFJ
Marginal Brackets (ages 62-71, before RMDs)
Federal 12% (15% after 2026), SS taxed at 85%
State 4%

Scenarios:
1. No Conversions
2. Top of IRMAA Tier2
3. Top of IRMAA Tier3
4. Top of 24% Bracket
5. Balance AGI
6. Balance Effective Tax


Metrics Considered:
  • Marginal Tax Rates vs. Future Marginal Rates (Fed + CA State)
  • Total Portfolio Value
  • Total Tax Adjusted Portfolio Value
  • Total Taxes (Fed +IRMAA + State) paid
  • Return on Investment (Tax Adjusted Portfolio Delta/ Total Taxes Paid Delta (Scenario compared to No Conversions)
  • Internal Rate of Return ( Tax Adjusted Portfolio Delta/ Total Taxes Paid Delta (Scenario compared to No Conversions)
I have looked at these metrics at various time periods (62-70, 71-80, 81-90, 62-90).

Would appreciate any insights or thoughts on how to make a selection!

Thanks in Advance

WoodSpinner


Woodspinner

at a 12% bracket, noted in some earlier thread (which I can't remember title), there's really not much benefit for conversions as the payback (resultant tax free growth vs tax paid and its associated growth) takes at least a decade

usually the concern is when either growth puts you into a higher bracket or if it's a condition of a surviving spouse being kicked into a higher bracket
(often state taxes start at lower income thresholds so little is gained in that realm**, so it's really future federal marginal rate (including NIIT and IRMAA costs) vs current)

the 71-85(or 90) captures the most likely condition, as that's when you have to start withdrawals (tax becomes real) while still having a high probability of still being around- - if it doesn't give you much benefits then you really shouldn't convert beyond the top of the 12% bracket

remember that future years will likely cause a need for higher medical or other costs, hence keeping a sizable amount in a regular IRA (that hasn't been taxed) allows for that to be withdrawn at low (or no) taxes. For us, I'll be converting my current rollover IRA but not my other 401k (which will just be rolled over.... then withdrawn, even before 72, so as to supplement pension/investment income/SS (which I'm eligible for but have still deferred so as to enable the current Roth conversions)

(** an exception would be a case where one would retire in another state.... either going to a state that taxes at say 9-10% to a no income tax state or going to a state that doesn't tax IRA withdrawals (which would then play into the don't convert or lower level of conversions argument))
User avatar
FiveK
Posts: 15693
Joined: Sun Mar 16, 2014 2:43 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by FiveK »

WoodSpinner wrote: Sat Jul 24, 2021 8:42 pm Metrics Considered:
  • Marginal Tax Rates vs. Future Marginal Rates (Fed + CA State) That's a good indicator. If you sharpen your pencil, Roth can end up better even if the future marginal rates are "somewhat" lower than current. Definition of "somewhat" varies by situation.
  • Total Portfolio Value Not as good as the one below.
  • Total Tax Adjusted Portfolio Value Better than the one above. Does require an estimate of the tax adjustment.
  • Total Taxes (Fed +IRMAA + State) paid No. Even with identical marginal rates, converting later (after growth) will incur more tax. What matters is how much is left after taxes.
  • Return on Investment (Tax Adjusted Portfolio Delta/ Total Taxes Paid Delta (Scenario compared to No Conversions)
  • Internal Rate of Return ( Tax Adjusted Portfolio Delta/ Total Taxes Paid Delta (Scenario compared to No Conversions)Both this one and the one above suffer from the "total taxes" inclusion.
If by "Total Tax Adjusted Portfolio Value" you mean the spendable amount after all taxes, it seems that is the best metric to maximize. If the optimum as measure by a different metric gives less to spend, it seems there would be some explaining needed as to why that is in fact better.
Last edited by FiveK on Mon Jul 26, 2021 11:02 pm, edited 1 time in total.
User avatar
FiveK
Posts: 15693
Joined: Sun Mar 16, 2014 2:43 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by FiveK »

Nestegg_User wrote: Sat Jul 24, 2021 9:24 pm at a 12% bracket, noted in some earlier thread (which I can't remember title), there's really not much benefit for conversions as the payback (resultant tax free growth vs tax paid and its associated growth) takes at least a decade
Are you thinking of Roth Conversions - McQuarrie study? Whether that or another thread, more information is needed to understand what is meant by "payback...takes at least a decade." E.g., if one converts now at 12% but the marginal rate for all future years is 22% or higher, payback is within one year.
User avatar
Nestegg_User
Posts: 2109
Joined: Wed Aug 05, 2009 1:26 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Nestegg_User »

FiveK

the question was if one STAYS at a 12% bracket... then it takes awhile
obviously, if one goes from 12% to 22% it makes sense... as I noted (the difference between the rates-- he was asking which metrics)
virtually all say that it's the length of time that one can enjoy the non-taxed growth of the Roth, before it might have to be withdrawn, that makes sense. The old "don't pay taxes too early... the growth of those taxes paid needs to be taken in account"... and if the Roth is going to be inherited by non spouse there's other considerations (which I really don't care about since we don't have kids... and most of either of our siblings are in a low tax bracket if we pass early and it gets to them via our trust (but the Roth is for us... and our circumstances... anything that passes to them is just their gravy , might be diluted by the taxes paid but still... and there'd still be a traditional IRA or two that would transfer)


(i remember an earlier one than that... but can't remember how long ago that was; I haven't fully looked at the McQ paper or thread...got other things to do in retirement... and i'm already in process with Roth conversions and have my thoughts on how far we need to go, both due to pension/dual SS (although not starting at nearly the same time) and life expectancy (both joint and surviving). Currently, doing to top of 12% (virtually everyone should do at least to that, unless moving from high tax to no tax state (wait until move) or where IRA withdrawals aren't taxed ( where you would convert less under certain circumstances) )
Last edited by Nestegg_User on Sat Jul 24, 2021 11:16 pm, edited 1 time in total.
User avatar
FiveK
Posts: 15693
Joined: Sun Mar 16, 2014 2:43 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by FiveK »

Nestegg_User wrote: Sat Jul 24, 2021 11:04 pm the question was if one STAYS at a 12% bracket... then it takes awhile
Ok, thanks, that helps.

But I still don't understand. If one converts at 12% and pays the tax out of the conversion, then withdraws (after, say, a 10% increase) from the Roth next year, the result is the same as if one waits a year (getting that same 10% increase) and then takes a distribution from the traditional account and pays the tax then. E.g., $1000 * (1 - 12%) * (1 + 10%) = $1000 * (1 + 10%) * (1 - 12%).

There can be a slight advantage to paying the tax out of "cash on hand" in some cases, but again the payback would start immediately.
User avatar
Lee_WSP
Posts: 10346
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Lee_WSP »

it's a tax arbitrage play intended to reduce the taxes paid as a percentage of income.
User avatar
Nestegg_User
Posts: 2109
Joined: Wed Aug 05, 2009 1:26 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Nestegg_User »

In your MCQ paper it still states "Given constant tax rates, in pretax terms it takes a long time to overcome the initial tax debit incurred by the conversion" (page 9) where in figure 4 it takes 18 years for breakeven. {that's the case I was talking about} I did not argue on the case of 12% going to 22% (and I didn't see McQ do so either)
(further in the paper (p10) "It is advantageous to have available Roth funds available from a conversion...unreduced by a tax hit. ....But, prior to the mid-90's, the real dollar advantage gained is small relative to the amount converted")


In our case, we did examine at the time the inclusion of the pension... and saw that there really wasn't any benefits for conversions beyond the 12% bracket (and those largely due to the added flexibility to temporarily up spending if desired (from the converted funds) without higher tax and for funds for surviving spouse in future (potentially higher rate...but now more in fully tax free account for maximal flexibility))
The changes to later initial withdrawals doesn't change the outcome...we'll be in the 22% bracket as MFJ once SS starts.
Last edited by Nestegg_User on Sun Jul 25, 2021 12:17 am, edited 1 time in total.
User avatar
FiveK
Posts: 15693
Joined: Sun Mar 16, 2014 2:43 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by FiveK »

Nestegg_User wrote: Sun Jul 25, 2021 12:02 am In your MCQ paper it still states "Given constant tax rates, in pretax terms it takes a long time to overcome the initial tax debit incurred by the conversion" (page 9) where in figure 4 it takes 18 years for breakeven. I did not argue on the case of 12% going to 22% (and I didn't see McQ do so either)
It wasn't understood there either.

Probably a case of an unstated (or at least misunderstood) difference of assumptions.

Again, if one converts at 12% and pays the tax out of the conversion, then withdraws (after, say, a 10% increase) from the Roth next year, the result is the same as if one waits a year (getting that same 10% increase) and then takes a distribution from the traditional account and pays the tax then. E.g., $1000 * (1 - 12%) * (1 + 10%) = $1000 * (1 + 10%) * (1 - 12%).

In other words, it does not take a long time to overcome the initial tax debit because the tax must be paid whenever money is withdrawn from the traditional account. At least, that's the assumption I'm using....
User avatar
Nestegg_User
Posts: 2109
Joined: Wed Aug 05, 2009 1:26 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Nestegg_User »

see his table 6 and table 7

note that he used base case of 22% bracket in retirement (as he already has described that one would have to have multiple million in tax deferred to bump into higher rates)
User avatar
FiveK
Posts: 15693
Joined: Sun Mar 16, 2014 2:43 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by FiveK »

Nestegg_User wrote: Sun Jul 25, 2021 12:23 am see his table 6 and table 7

note that he used base case of 22% bracket in retirement (as he already has described that one would have to have multiple million in tax deferred to bump into higher rates)
Ignoring the McQ paper for a bit, I'm still trying to understand what you meant by
Nestegg_User wrote: Sat Jul 24, 2021 11:04 pm the question was if one STAYS at a 12% bracket... then it takes awhile
What takes awhile, and how does that differ from the example I gave?
User avatar
Nestegg_User
Posts: 2109
Joined: Wed Aug 05, 2009 1:26 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Nestegg_User »

FiveK wrote: Sun Jul 25, 2021 12:36 am
Nestegg_User wrote: Sun Jul 25, 2021 12:23 am see his table 6 and table 7

note that he used base case of 22% bracket in retirement (as he already has described that one would have to have multiple million in tax deferred to bump into higher rates)
Ignoring the McQ paper for a bit, I'm still trying to understand what you meant by
Nestegg_User wrote: Sat Jul 24, 2021 11:04 pm the question was if one STAYS at a 12% bracket... then it takes awhile
What takes awhile, and how does that differ from the example I gave?

I've already given you the reference... looking at the case of constant tax rates

you might want to further read the paper... he doesn't use the case of taxes paid out of other accounts (as he noted that they would have to be accounted for... both the loss of portfolio value due to the tax payments and their ancillary growth and the need to account for variance between the taxable accounts wherein those funds originated and the tax deferred.... you need to account for all taxes and growth in all accounts
you seem to be arguing for both a 12% and a 22% rate at the same time... (not to mention a ten percent growth in a year... not likely to be persistent)

quoting further probably goes beyond "fair use"... so I'm out
User avatar
FiveK
Posts: 15693
Joined: Sun Mar 16, 2014 2:43 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by FiveK »

Nestegg_User wrote: Sun Jul 25, 2021 1:06 am
FiveK wrote: Sun Jul 25, 2021 12:36 am
Nestegg_User wrote: Sun Jul 25, 2021 12:23 am see his table 6 and table 7

note that he used base case of 22% bracket in retirement (as he already has described that one would have to have multiple million in tax deferred to bump into higher rates)
Ignoring the McQ paper for a bit, I'm still trying to understand what you meant by
Nestegg_User wrote: Sat Jul 24, 2021 11:04 pm the question was if one STAYS at a 12% bracket... then it takes awhile
What takes awhile, and how does that differ from the example I gave?
I've already given you the reference... looking at the case of constant tax rates
The title of Table 6 is "Middle Income Taxpayer Converting Partly in the 0% Bracket" - so that doesn't match "STAYS at a 12% bracket". At least, I'm assuming "STAYS" means both "converts at" and "withdraws at" - did you mean something else?
you seem to be arguing for both a 12% and a 22% rate at the same time
No, just trying to stay with what I thought was your assumption of a constant 12% marginal rate.
... (not to mention a ten percent growth in a year... not likely to be persistent)
Using any growth number doesn't change the result, due to the commutative property of multiplication.
quoting further probably goes beyond "fair use"... so I'm out
Your call - we can ignore the McQ paper completely and work with the "12% bracket" example if you want.
babystep
Posts: 763
Joined: Tue Apr 09, 2019 9:44 am

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by babystep »

WoodSpinner wrote: Sat Jul 24, 2021 8:42 pm
What I am seeing is that the various Metrics simply aren't working as an effective tool to make a selection. There is significant variation based on time period (e.g. 71-80 vs. 81-90. I am also getting conflicting signals from the various metrics as to which is the better scenario.

At this point I need some advice on which metric(s) I should really be prioritizing.
I don't think we can look at different time periods. One key characteristic of Roth conversion is to take advantage of low tax rate years but still being conservative in estimating your TDA growth so that you don't cross the next marginal rate during RMD and SS time periods.

I think it needs to include all the time periods including heirs and charity, if applicable.

Maybe total money in real dollars available to you/heirs/charity post all the taxes are paid?
is50xenough
Posts: 323
Joined: Sat Jul 28, 2018 1:37 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by is50xenough »

WoodSpinner wrote: Sat Jul 24, 2021 8:42 pm I have been doing quite a bit of modeling on Roth Conversions with a goal of identifying the most optimal scenario (given the constraints of predicting the future). A bit puzzled on which metrics to use to compare the various scenarios to provide some insights into which are better than others.

What I am seeing is that the various Metrics simply aren't working as an effective tool to make a selection. There is significant variation based on time period (e.g. 71-80 vs. 81-90. I am also getting conflicting signals from the various metrics as to which is the better scenario.

At this point I need some advice on which metric(s) I should really be prioritizing.

Background:
  • Current info:
Ages: 62 (me), Spouse (61), MFJ
Marginal Brackets (ages 62-71, before RMDs)
Federal 12% (15% after 2026), SS taxed at 85%
State 4%

Scenarios:
1. No Conversions
2. Top of IRMAA Tier2
3. Top of IRMAA Tier3
4. Top of 24% Bracket
5. Balance AGI
6. Balance Effective Tax


Metrics Considered:
  • Marginal Tax Rates vs. Future Marginal Rates (Fed + CA State)
  • Total Portfolio Value
  • Total Tax Adjusted Portfolio Value
  • Total Taxes (Fed +IRMAA + State) paid
  • Return on Investment (Tax Adjusted Portfolio Delta/ Total Taxes Paid Delta (Scenario compared to No Conversions)
  • Internal Rate of Return ( Tax Adjusted Portfolio Delta/ Total Taxes Paid Delta (Scenario compared to No Conversions)
I have looked at these metrics at various time periods (62-70, 71-80, 81-90, 62-90).

Would appreciate any insights or thoughts on how to make a selection!

Thanks in Advance

WoodSpinner
Woodspinner, I’d like to see total portfolio just to better understand total tax adjusted portfolio. Not sure what other metrics get you. Love to see some of your results.

Good luck
basspond
Posts: 1500
Joined: Wed Nov 27, 2013 3:01 am

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by basspond »

We have done calculations based on current rate (12%) and extrapolated growth for both accounts and used 22% and 24% future rates . In the end there is not much difference except that now or then depending on your income might throw you into higher contributions to Medicare, etc. You can also limit your taxes after RMD age by donating from IRA directly to charitable organizations.
Exchme
Posts: 1323
Joined: Sun Sep 06, 2020 3:00 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Exchme »

Lots of factors to think about.

Let's assume you aren't going to run out of money while alive. Then the deciding factor in the metric is what you want to do with your bequests.

If you are giving the money to charity, then the t-IRA, taxable and Roth all pass on equal terms, so there is no heir liquidation tax adjustment. That would tend to push you to small or no Roth Conversions. If you don't know your heirs well enough to care about how much they get vs. the government, then that is equivalent, you only care about net worth without tax adjustment.

If you are doing estate planning to maximize what you pass to your heirs, then you should estimate their tax situation and run the year by year liquidation for the 10 year time frame allowed for your heirs to withdraw your t-IRA money. The No Roth case has higher taxes both on the withdrawal and on the taxable account. I generally assume my heirs will be my financial clone and look at the liquidation in RPM to get the total impact. In my case, I plan on paying for conversions out of taxable and I was surprised to find the change in tax drag in the taxable account was also significant. If you are fortunate enough to owe estate taxes, that would favor doing more Roths than otherwise as the effective marginal bracket for heirs would be high.

If you might run out of money when alive (maybe bad SORR + big LTC needs), that would tend to disfavor the Roth. Since LTC paid out of t-IRA would still get a tax deduction, doing a Roth would have unnecessarily paid conversion taxes and bad SORR might mean your planned future high tax rates never materialized. Of course the opposite is true too, better than planned future financials will make you wish you had done more conversions (my bet is most Bogleheads are so conservative on their projections, they probably fall into this camp, eventually wishing they'd done more!)

The other factor that had a big effect for me was the optimization of asset location. Putting bonds in t-IRA showed its growth and the cut the optimum amount of Roth Conversions, so that's a good idea to build into your tool.
User avatar
cowdogman
Posts: 2054
Joined: Sat Dec 16, 2017 6:44 pm
Location: Washington State

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by cowdogman »

Three metrics:

1. Net worth at various ages (your "portfolio value" metric)

2. Cumulative taxes at various ages

3. Effective tax rate each year (altho 1 and 2 are the real metrics--3 is just interesting to know)

Marginal rates are misleading (because you could have only $1 of taxable income in the top rate).

Total tax adjusted portfolio is interesting, tho I'm not sure how you would do the adjustment--maybe instead taxable/non-taxable ratio? This ratio would be interesting if you were tax planning for your heirs' benefit.

The rate of return calculations could be interesting but they are really just another way of looking at 1 and 2 above--and it's really the $ that matter.

Whatever metrics you find interesting and helpful are great, but a lot of metrics could become confusing.

Regarding your scenarios, you might want to add:

1. Fixed Roth conversion amounts ($50K/year, $100K/year, etc.) just because it'll give a different perspective.

2. Death of one spouse and switch to Single tax bracket. I recently added this to my spreadsheet and found it very interesting--definitely a negative impact but not as bad as I thought. (I just assumed I went first given the effect on SS was the same no matter which of us went first. I set up the spreadsheet to jump from the filing MFJ worksheet to a new filing Single worksheet in the year after death.)

Finally, as I've said in other posts, the biggest impact on whether Roth conversions are worthwhile is when and how you use the converted amounts during retirement--so your spreadsheet should have Roth account liquidation scenarios as well. For example, in my calculations if I use the Roth conversions to reduce RMDs but never use the converted amounts, my net worth never recovers (that is, never exceeds my net worth without conversions).
User avatar
cowdogman
Posts: 2054
Joined: Sat Dec 16, 2017 6:44 pm
Location: Washington State

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by cowdogman »

babystep wrote: Sun Jul 25, 2021 1:27 am
WoodSpinner wrote: Sat Jul 24, 2021 8:42 pm
What I am seeing is that the various Metrics simply aren't working as an effective tool to make a selection. There is significant variation based on time period (e.g. 71-80 vs. 81-90. I am also getting conflicting signals from the various metrics as to which is the better scenario.

At this point I need some advice on which metric(s) I should really be prioritizing.
I don't think we can look at different time periods. One key characteristic of Roth conversion is to take advantage of low tax rate years but still being conservative in estimating your TDA growth so that you don't cross the next marginal rate during RMD and SS time periods.

I think it needs to include all the time periods including heirs and charity, if applicable.

Maybe total money in real dollars available to you/heirs/charity post all the taxes are paid?
I think time periods are very relevant. Whether you look at Roth conversions as an investment or a bet on future tax laws, determining when you have recouped your investment or won the bet is important. And in that context one very important question to ask is whether you are doing Roth conversions for your and your spouse's benefit only or also for your heirs (or somewhere in between--e.g., "primarily for us but also for our heirs").

As to what time period to focus on, look at all of them, but if your Roth conversion strategy is 100% centered on you and your spouse (and not your heirs) focus on your expected mortality date (with the hope that you exceed it). If your Roth conversion calculations tell you will recoup your investment five years after your expected mortality, then it may not be worth making the bet.
babystep
Posts: 763
Joined: Tue Apr 09, 2019 9:44 am

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by babystep »

cowdogman wrote: Sun Jul 25, 2021 10:51 am
babystep wrote: Sun Jul 25, 2021 1:27 am
WoodSpinner wrote: Sat Jul 24, 2021 8:42 pm
What I am seeing is that the various Metrics simply aren't working as an effective tool to make a selection. There is significant variation based on time period (e.g. 71-80 vs. 81-90. I am also getting conflicting signals from the various metrics as to which is the better scenario.

At this point I need some advice on which metric(s) I should really be prioritizing.
I don't think we can look at different time periods. One key characteristic of Roth conversion is to take advantage of low tax rate years but still being conservative in estimating your TDA growth so that you don't cross the next marginal rate during RMD and SS time periods.

I think it needs to include all the time periods including heirs and charity, if applicable.

Maybe total money in real dollars available to you/heirs/charity post all the taxes are paid?
I think time periods are very relevant. Whether you look at Roth conversions as an investment or a bet on future tax laws, determining when you have recouped your investment or won the bet is important. And in that context one very important question to ask is whether you are doing Roth conversions for your and your spouse's benefit only or also for your heirs (or somewhere in between--e.g., "primarily for us but also for our heirs").

As to what time period to focus on, look at all of them, but if your Roth conversion strategy is 100% centered on you and your spouse (and not your heirs) focus on your expected mortality date (with the hope that you exceed it). If your Roth conversion calculations tell you will recoup your investment five years after your expected mortality, then it may not be worth making the bet.
The time periods are indeed relevant but OP is trying to figure out metrics for comparison. The problem is the variation in metrics for different time periods (see the bold part in OP) which is somewhat by design for Roth conversions hence the holistic comparison needs to have full time duration.
User avatar
Lee_WSP
Posts: 10346
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Lee_WSP »

Marginal is the correct term, but bogleheads as a forum use it too narrowly to describe the last dollar. Instead the margin amount is supposed to be the amount of money being converted where you take the marginal rate on the sum and compare it to doing nothing.
diy60
Posts: 967
Joined: Wed Sep 07, 2016 6:54 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by diy60 »

Lee_WSP wrote: Sun Jul 25, 2021 11:04 am Marginal is the correct term, but bogleheads as a forum use it too narrowly to describe the last dollar. Instead the margin amount is supposed to be the amount of money being converted where you take the marginal rate on the sum and compare it to doing nothing.
+1
Excellent summary. I would add that one should include any major "costs" (e.g. taxes, IRMAA, NIIT, etc) in the numerator. However it will be situational dependent. For my situation I only include taxes and IRMAA, everything else is insignificant in the analysis.
User avatar
Lee_WSP
Posts: 10346
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Lee_WSP »

diy60 wrote: Sun Jul 25, 2021 11:19 am
Lee_WSP wrote: Sun Jul 25, 2021 11:04 am Marginal is the correct term, but bogleheads as a forum use it too narrowly to describe the last dollar. Instead the margin amount is supposed to be the amount of money being converted where you take the marginal rate on the sum and compare it to doing nothing.
+1
Excellent summary. I would add that one should include any major "costs" (e.g. taxes, IRMAA, NIIT, etc) in the numerator. However it will be situational dependent. For my situation I only include taxes and IRMAA, everything else is insignificant in the analysis.
Yes, there's a lot of small costs that are income dependent, they are included in the term marginal cost, but people sometimes need reminding of their existence.
User avatar
cowdogman
Posts: 2054
Joined: Sat Dec 16, 2017 6:44 pm
Location: Washington State

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by cowdogman »

babystep wrote: Sun Jul 25, 2021 10:56 am
cowdogman wrote: Sun Jul 25, 2021 10:51 am
babystep wrote: Sun Jul 25, 2021 1:27 am
WoodSpinner wrote: Sat Jul 24, 2021 8:42 pm
What I am seeing is that the various Metrics simply aren't working as an effective tool to make a selection. There is significant variation based on time period (e.g. 71-80 vs. 81-90. I am also getting conflicting signals from the various metrics as to which is the better scenario.

At this point I need some advice on which metric(s) I should really be prioritizing.
I don't think we can look at different time periods. One key characteristic of Roth conversion is to take advantage of low tax rate years but still being conservative in estimating your TDA growth so that you don't cross the next marginal rate during RMD and SS time periods.

I think it needs to include all the time periods including heirs and charity, if applicable.

Maybe total money in real dollars available to you/heirs/charity post all the taxes are paid?
I think time periods are very relevant. Whether you look at Roth conversions as an investment or a bet on future tax laws, determining when you have recouped your investment or won the bet is important. And in that context one very important question to ask is whether you are doing Roth conversions for your and your spouse's benefit only or also for your heirs (or somewhere in between--e.g., "primarily for us but also for our heirs").

As to what time period to focus on, look at all of them, but if your Roth conversion strategy is 100% centered on you and your spouse (and not your heirs) focus on your expected mortality date (with the hope that you exceed it). If your Roth conversion calculations tell you will recoup your investment five years after your expected mortality, then it may not be worth making the bet.
The time periods are indeed relevant but OP is trying to figure out metrics for comparison. The problem is the variation in metrics for different time periods (see the bold part in OP) which is somewhat by design for Roth conversions hence the holistic comparison needs to have full time duration.
Understand your point (I think) but almost nobody lives the "full time duration"--to the end of the RMD table--115 years old. The metrics do vary by age a lot--I've made that point in various posts. And as I said above we should look at various ages in the analysis but I would focus on the expected mortality.

Edit: Plus I think the fact that the metrics vary by age is very important to understand when doing a Roth analysis. There is a long pay back period to the initial investment in Roth conversions--that is, the investment of the incremental tax paid for the conversions.
User avatar
Topic Author
WoodSpinner
Posts: 3499
Joined: Mon Feb 27, 2017 12:15 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by WoodSpinner »

FiveK wrote: Sat Jul 24, 2021 10:36 pm
WoodSpinner wrote: Sat Jul 24, 2021 8:42 pm Metrics Considered:
  • Marginal Tax Rates vs. Future Marginal Rates (Fed + CA State) That's a good indicator. If you sharpen your pencil, Roth can end up better even if the future marginal rates are "somewhat" lower than current. Definition of "somewhat" varies by situation.
  • Total Portfolio Value Not as good as the one below.
  • Total Tax Adjusted Portfolio Value Better than the one above. Does require an estimate of the tax adjustment.
  • Total Taxes (Fed +IRMAA + State) paid No. Even with identical marginal rates, converting later (after growth) will incur more tax. What matters is how much is left after taxes.
  • Return on Investment (Tax Adjusted Portfolio Delta/ Total Taxes Paid Delta (Scenario compared to No Conversions)
  • Internal Rate of Return ( Tax Adjusted Portfolio Delta/ Total Taxes Paid Delta (Scenario compared to No Conversions)Both this one and the one above suffer from the "total taxes" inclusion.
If by "Total Tax Adjusted Portfolio Value" you mean the spendable amount after all taxes, it seems that is the best metric to maximize. If the optimum as measure by a different metric gives less to spend, it seems there would be some explaining needed as to why that is in fact better.
So let's put some real numbers and see if it helps. The numbers reflect the deltas of the Total Tax Adjusted Portfolio Value which is the same as "spendable after taxes" at 90.

Image

I would certainly eliminate No Conversions and Balance Effective Taxes based on your suggestions..

This leaves the other scenarios in-play and the variance between them is fairly small (under 2%). I don't think this metric is enough to make a decision on.

If you then look at Return on Investment (ROI) it helps highlights the tax costs of getting the Tax adjusted Portfolio Value and can provide an insight into the efficiency of the tax payments versus spendable wealth.

Image.

RoI does provide some meaningful variance and if I stopped here, I have a clear answer to convert to the Top of the 24% Marginal bracket.

However, I think (and certainly could be wrong) that the Internal Rate of Return (IRR) is a better financial metric for evaluating expenses versus returns.

Image.

My confusion is this 2 metrics do not lead to the same conclusions.

FWIW, I am leaning towards converting to the Top of IRMAA Tier2.

Any thoughts on how to best decide?

WoodSpinner
WoodSpinner
MileKing
Posts: 74
Joined: Thu Mar 28, 2019 12:14 pm
Location: Reno, NV

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by MileKing »

Nestegg_User wrote: Sun Jul 25, 2021 12:02 am In your MCQ paper it still states "Given constant tax rates, in pretax terms it takes a long time to overcome the initial tax debit incurred by the conversion" (page 9) where in figure 4 it takes 18 years for breakeven. {that's the case I was talking about} I did not argue on the case of 12% going to 22% (and I didn't see McQ do so either)
(further in the paper (p10) "It is advantageous to have available Roth funds available from a conversion...unreduced by a tax hit. ....But, prior to the mid-90's, the real dollar advantage gained is small relative to the amount converted")
As I understand it, Figure 4 in the MCQ paper is a pre-tax analysis. That is, any tax deferred dollars such as those in a traditional IRA are not being adjusted for taxes. Those unadjusted tax deferred dollars are being compared against the Roth conversion scenario where taxes have already been paid. This doesn't seem like an apples-to-apples comparison to me. $1M in a traditional IRA is not the same as $1M in a Roth IRA. The MCQ paper also discusses "liquidation analysis", which is what is reported in part B of Table 4. Curiously enough, MCQ only reports the after-tax analysis (positive for a Roth conversion) starting at age 85 leaving the reader to wonder what the study results are pre-85 years old. As others in this thread and other Roth conversion threads have noted, at the same tax rate Roth conversions are breakeven from day 1.

WoodSpinner, I believe establishing one's goals with regards to Roth conversions are needed before identifying suitable metrics to compare scenarios. Two goals that are frequently discussed by Bogleheads when assessing retirement needs are:
1. Maximizing the size of one's portfolio, preferably tax-adjusted, for benefit of heirs (spouse, other beneficiaries, charities)
2. Maximizing one's level of spending while alive
Goal #1 seems to be addressed by your metric Total Tax-Adjusted Portfolio Value. I don't see anything to address Goal #2. All of the tax related metrics seem to play into both of these goals in that an increase in taxes paid will likely reduce your portfolio value and your level of spending. If your goal is to pay the least taxes over your lifetime (or retirement) or have your heirs pay the least taxes, then the tax related metrics may have value. Otherwise, sticking with Total Tax-Adjusted Portfolio Value and/or something like Annual Spending Level Supported, seem more relevant.
User avatar
Topic Author
WoodSpinner
Posts: 3499
Joined: Mon Feb 27, 2017 12:15 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by WoodSpinner »

is50xenough wrote: Sun Jul 25, 2021 3:00 am
Woodspinner, I’d like to see total portfolio just to better understand total tax adjusted portfolio. Not sure what other metrics get you. Love to see some of your results.

Good luck
Hypothetical …

At age 90:
Taxable $700,000
Roth $3,000,000
IRA $900,000
Total $4,600,000

Marginal rate at 90, 28% Fed, 9.3% State (37.3% Marginal)
Tax Adjusted Portfolio
Taxable $700,000
Roth $3,000,000
IRA $564,300
Total $4,264,000

Does this help?

WoodSpinner
WoodSpinner
User avatar
Topic Author
WoodSpinner
Posts: 3499
Joined: Mon Feb 27, 2017 12:15 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by WoodSpinner »

Exchme wrote: Sun Jul 25, 2021 9:49 am Lots of factors to think about.

Let's assume you aren't going to run out of money while alive. Then the deciding factor in the metric is what you want to do with your bequests.

If you are giving the money to charity, then the t-IRA, taxable and Roth all pass on equal terms, so there is no heir liquidation tax adjustment. That would tend to push you to small or no Roth Conversions. If you don't know your heirs well enough to care about how much they get vs. the government, then that is equivalent, you only care about net worth without tax adjustment.

If you are doing estate planning to maximize what you pass to your heirs, then you should estimate their tax situation and run the year by year liquidation for the 10 year time frame allowed for your heirs to withdraw your t-IRA money. The No Roth case has higher taxes both on the withdrawal and on the taxable account. I generally assume my heirs will be my financial clone and look at the liquidation in RPM to get the total impact. In my case, I plan on paying for conversions out of taxable and I was surprised to find the change in tax drag in the taxable account was also significant. If you are fortunate enough to owe estate taxes, that would favor doing more Roths than otherwise as the effective marginal bracket for heirs would be high.

If you might run out of money when alive (maybe bad SORR + big LTC needs), that would tend to disfavor the Roth. Since LTC paid out of t-IRA would still get a tax deduction, doing a Roth would have unnecessarily paid conversion taxes and bad SORR might mean your planned future high tax rates never materialized. Of course the opposite is true too, better than planned future financials will make you wish you had done more conversions (my bet is most Bogleheads are so conservative on their projections, they probably fall into this camp, eventually wishing they'd done more!)

The other factor that had a big effect for me was the optimization of asset location. Putting bonds in t-IRA showed its growth and the cut the optimum amount of Roth Conversions, so that's a good idea to build into your tool.
Exchme,

I doubt we will run out of money. Roth will likely go to my wife and daughter to spend. IRA should have funds to cover her LTC (I have LTC insurance) plus our charitable bequests which will be significant). Roth might also be used for some un-budgeted expenses — 30 years is a long time and I doubt my estimates anticipate everything.

I have also found significant benefit to the optimization of asset location and plan to post in this in a separate thread. Not only is the the IRA growth slowed but the equities moved to Roth are now growing tax free. I am still trying to figure out how to quantify the benefit— did you do any work along those lines?

WoodSpinner
WoodSpinner
User avatar
Topic Author
WoodSpinner
Posts: 3499
Joined: Mon Feb 27, 2017 12:15 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by WoodSpinner »

cowdogman wrote: Sun Jul 25, 2021 10:18 am Three metrics:

1. Net worth at various ages (your "portfolio value" metric)

2. Cumulative taxes at various ages

3. Effective tax rate each year (altho 1 and 2 are the real metrics--3 is just interesting to know)

Marginal rates are misleading (because you could have only $1 of taxable income in the top rate).

Total tax adjusted portfolio is interesting, tho I'm not sure how you would do the adjustment--maybe instead taxable/non-taxable ratio? This ratio would be interesting if you were tax planning for your heirs' benefit.

The rate of return calculations could be interesting but they are really just another way of looking at 1 and 2 above--and it's really the $ that matter.

Whatever metrics you find interesting and helpful are great, but a lot of metrics could become confusing.

Regarding your scenarios, you might want to add:

1. Fixed Roth conversion amounts ($50K/year, $100K/year, etc.) just because it'll give a different perspective.

2. Death of one spouse and switch to Single tax bracket. I recently added this to my spreadsheet and found it very interesting--definitely a negative impact but not as bad as I thought. (I just assumed I went first given the effect on SS was the same no matter which of us went first. I set up the spreadsheet to jump from the filing MFJ worksheet to a new filing Single worksheet in the year after death.)
I am also seeing the same thing. Actual taxes are increased but not enough to impact her retirement. Marginal rates do grow at least 1 bracket. More on that in a separate post.
Finally, as I've said in other posts, the biggest impact on whether Roth conversions are worthwhile is when and how you use the converted amounts during retirement--so your spreadsheet should have Roth account liquidation scenarios as well. For example, in my calculations if I use the Roth conversions to reduce RMDs but never use the converted amounts, my net worth never recovers (that is, never exceeds my net worth without conversions).
I am not seeing this effect! Taxes are paid from the IRA since I have no taxable. Are you working in Real or Nominal $s? I am modeling in Nominal FWIW (it’s the way I started in 2017). I guess I could do a NPV and discount back by the inflation rates to double check…..

WoodSpinner
WoodSpinner
User avatar
Topic Author
WoodSpinner
Posts: 3499
Joined: Mon Feb 27, 2017 12:15 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by WoodSpinner »

cowdogman wrote: Sun Jul 25, 2021 11:52 am
babystep wrote: Sun Jul 25, 2021 10:56 am
cowdogman wrote: Sun Jul 25, 2021 10:51 am
babystep wrote: Sun Jul 25, 2021 1:27 am
WoodSpinner wrote: Sat Jul 24, 2021 8:42 pm
What I am seeing is that the various Metrics simply aren't working as an effective tool to make a selection. There is significant variation based on time period (e.g. 71-80 vs. 81-90. I am also getting conflicting signals from the various metrics as to which is the better scenario.

At this point I need some advice on which metric(s) I should really be prioritizing.
I don't think we can look at different time periods. One key characteristic of Roth conversion is to take advantage of low tax rate years but still being conservative in estimating your TDA growth so that you don't cross the next marginal rate during RMD and SS time periods.

I think it needs to include all the time periods including heirs and charity, if applicable.

Maybe total money in real dollars available to you/heirs/charity post all the taxes are paid?
I think time periods are very relevant. Whether you look at Roth conversions as an investment or a bet on future tax laws, determining when you have recouped your investment or won the bet is important. And in that context one very important question to ask is whether you are doing Roth conversions for your and your spouse's benefit only or also for your heirs (or somewhere in between--e.g., "primarily for us but also for our heirs").

As to what time period to focus on, look at all of them, but if your Roth conversion strategy is 100% centered on you and your spouse (and not your heirs) focus on your expected mortality date (with the hope that you exceed it). If your Roth conversion calculations tell you will recoup your investment five years after your expected mortality, then it may not be worth making the bet.
The time periods are indeed relevant but OP is trying to figure out metrics for comparison. The problem is the variation in metrics for different time periods (see the bold part in OP) which is somewhat by design for Roth conversions hence the holistic comparison needs to have full time duration.
Understand your point (I think) but almost nobody lives the "full time duration"--to the end of the RMD table--115 years old. The metrics do vary by age a lot--I've made that point in various posts. And as I said above we should look at various ages in the analysis but I would focus on the expected mortality.

Edit: Plus I think the fact that the metrics vary by age is very important to understand when doing a Roth analysis. There is a long pay back period to the initial investment in Roth conversions--that is, the investment of the incremental tax paid for the conversions.
It’s clear that payback will occur in my 80’s (assuming I am first to die). In some ways, Roth Conversions are a bet on your longevity since the return increases over time.

WoodSpinner
WoodSpinner
User avatar
Topic Author
WoodSpinner
Posts: 3499
Joined: Mon Feb 27, 2017 12:15 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by WoodSpinner »

MileKing wrote: Sun Jul 25, 2021 3:12 pm
Nestegg_User wrote: Sun Jul 25, 2021 12:02 am In your MCQ paper it still states "Given constant tax rates, in pretax terms it takes a long time to overcome the initial tax debit incurred by the conversion" (page 9) where in figure 4 it takes 18 years for breakeven. {that's the case I was talking about} I did not argue on the case of 12% going to 22% (and I didn't see McQ do so either)
(further in the paper (p10) "It is advantageous to have available Roth funds available from a conversion...unreduced by a tax hit. ....But, prior to the mid-90's, the real dollar advantage gained is small relative to the amount converted")
As I understand it, Figure 4 in the MCQ paper is a pre-tax analysis. That is, any tax deferred dollars such as those in a traditional IRA are not being adjusted for taxes. Those unadjusted tax deferred dollars are being compared against the Roth conversion scenario where taxes have already been paid. This doesn't seem like an apples-to-apples comparison to me. $1M in a traditional IRA is not the same as $1M in a Roth IRA. The MCQ paper also discusses "liquidation analysis", which is what is reported in part B of Table 4. Curiously enough, MCQ only reports the after-tax analysis (positive for a Roth conversion) starting at age 85 leaving the reader to wonder what the study results are pre-85 years old. As others in this thread and other Roth conversion threads have noted, at the same tax rate Roth conversions are breakeven from day 1.

WoodSpinner, I believe establishing one's goals with regards to Roth conversions are needed before identifying suitable metrics to compare scenarios. Two goals that are frequently discussed by Bogleheads when assessing retirement needs are:
1. Maximizing the size of one's portfolio, preferably tax-adjusted, for benefit of heirs (spouse, other beneficiaries, charities)
This is probably the closest to our goal.
2. Maximizing one's level of spending while alive
Goal #1 seems to be addressed by your metric Total Tax-Adjusted Portfolio Value. I don't see anything to address Goal #2. All of the tax related metrics seem to play into both of these goals in that an increase in taxes paid will likely reduce your portfolio value and your level of spending. If your goal is to pay the least taxes over your lifetime (or retirement) or have your heirs pay the least taxes, then the tax related metrics may have value. Otherwise, sticking with Total Tax-Adjusted Portfolio Value and/or something like Annual Spending Level Supported, seem more relevant.
This goal doesn’t resonate since we seem to have enough to meet our needs + wants. That said, one of the next steps is to do a Monte Carlo Analysis to see the probabilistic effect of our Roth Conversion strategy on our Retirement. It’s easy to mistake an Excel model for reality….

WoodSpinner
WoodSpinner
User avatar
cowdogman
Posts: 2054
Joined: Sat Dec 16, 2017 6:44 pm
Location: Washington State

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by cowdogman »

WoodSpinner wrote: Sun Jul 25, 2021 2:29 pm
My confusion is this 2 metrics do not lead to the same conclusions.
Different metrics providing different results is a feature, not a bug.

More importantly, would you prefer a high IRR/ROI or more money? In other words, why would you look beyond the dollar results at various ages?

I am curious how you did the IRR calculation on a Roth conversion. What were the variables?
User avatar
cowdogman
Posts: 2054
Joined: Sat Dec 16, 2017 6:44 pm
Location: Washington State

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by cowdogman »

MileKing wrote: Sun Jul 25, 2021 3:12 pm As others in this thread and other Roth conversion threads have noted, at the same tax rate Roth conversions are breakeven from day 1.
Bolding above is mine.

This is a key point buried in a very helpful post and a point that I have never quite agreed with. The argument for this point is that a Roth conversion amount + the tax thereon = the resulting Roth account balance. In other words, the tax isn't an investment, it's the price for buying an investment (the Roth) worth (on an after tax basis) the same as the conversion amount + the tax. Breakeven is on day one--conditioned on the bolded language above.

I do see Roth conversions as an investment (or a bet) and I won't know whether it was a good investment (or bet) until I know what the tax laws will be in the future and until I use (if ever) the money in the Roth account in lieu of a taxable withdrawal. From a planning perspective, I think it is fair and appropriate to treat the Roth conversion tax as an investment with a future breakeven date, although I'm now re-thinking how to calculate that.
User avatar
Topic Author
WoodSpinner
Posts: 3499
Joined: Mon Feb 27, 2017 12:15 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by WoodSpinner »

cowdogman wrote: Sun Jul 25, 2021 3:59 pm
WoodSpinner wrote: Sun Jul 25, 2021 2:29 pm
My confusion is this 2 metrics do not lead to the same conclusions.
Different metrics providing different results is a feature, not a bug.

More importantly, would you prefer a high IRR/ROI or more money? In other words, why would you look beyond the dollar results at various ages?

I am curious how you did the IRR calculation on a Roth conversion. What were the variables?
Well, since most of the ending portfolio amounts are within a few % of each (see post above) other it seems that looking at the costs from taxes to achieve them is worthwhile.

IRR of return is calculated using:
Delta Taxes paid (Fed, IRMAA, State) (e.g. delta of expected taxes each yearConversionScenario - NoConversions) with the Delta Tax Adjusted Portfolio Value as the last value in the array. Used the Excel IRR function to calculate.

WoodSpinner
WoodSpinner
User avatar
Lee_WSP
Posts: 10346
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Lee_WSP »

cowdogman wrote: Sun Jul 25, 2021 4:24 pm
MileKing wrote: Sun Jul 25, 2021 3:12 pm As others in this thread and other Roth conversion threads have noted, at the same tax rate Roth conversions are breakeven from day 1.
Bolding above is mine.

This is a key point buried in a very helpful post and a point that I have never quite agreed with. The argument for this point is that a Roth conversion amount + the tax thereon = the resulting Roth account balance. In other words, the tax isn't an investment, it's the price for buying an investment (the Roth) worth (on an after tax basis) the same as the conversion amount + the tax. Breakeven is on day one--conditioned on the bolded language above.

I do see Roth conversions as an investment (or a bet) and I won't know whether it was a good investment (or bet) until I know what the tax laws will be in the future and until I use (if ever) the money in the Roth account in lieu of a taxable withdrawal. From a planning perspective, I think it is fair and appropriate to treat the Roth conversion tax as an investment with a future breakeven date, although I'm now re-thinking how to calculate that.
I'm not sure we can know whether we broke even on the conversion or not until we die and even then, we can't have a full accounting until our beneficiaries finish withdrawing the money.

On the one hand, if you're going to end life with zero or you need to leave your heirs some non tax deferred dollars, doing a little more aggressive Roth conversions makes sense. But if on the other hand, you're going to die with too much and you don't care about leaving your heirs tax deferred dollars, putting off taxes as long as possible may make more sense especially if you have other income adjusted payments to consider.

I'd say most people fall somewhere in between those two. The further extreme with a strong preference for Roth are people who are going to be leaving a multi generational legacy locked up in an irrevocable trust whose own tax rates are less than the highest bracket.
User avatar
cowdogman
Posts: 2054
Joined: Sat Dec 16, 2017 6:44 pm
Location: Washington State

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by cowdogman »

WoodSpinner wrote: Sun Jul 25, 2021 5:04 pm
Well, since most of the ending portfolio amounts are within a few % of each (see post above) other it seems that looking at the costs from taxes to achieve them is worthwhile.

IRR of return is calculated using:
Delta Taxes paid (Fed, IRMAA, State) (e.g. delta of expected taxes each yearConversionScenario - NoConversions) with the Delta Tax Adjusted Portfolio Value as the last value in the array. Used the Excel IRR function to calculate.

WoodSpinner
Sorry, maybe I wasn't clear. What value do the return percentage (ROI and IRR) add to the various dollar amounts--portfolio value, cumulative taxes, whatever? If you find them helpful, that's really all that matters, but the dollar amounts would seem to trump any return calculation.

Regarding the IRR, thanks. So "=IRR([array of years, with the first years being Roth conversion taxes (expressed as negatives) and then a number of years at $0],[portfolio value delta])"?

Have you thought of running the same calculation using [delta cumulative taxes] in place of [portfolio value delta]? Ditto for the ROI calculation.

I'm trying to figure out why your numbers are so different from mine. I'm guessing CA state taxes (I'm in WA with no state income tax) and I plan to rely solely on our taxable account for any spending gap above SS and RMDs. So now I'm wondering whether I should add a potential future WA state income tax to my spreadsheet.
User avatar
cowdogman
Posts: 2054
Joined: Sat Dec 16, 2017 6:44 pm
Location: Washington State

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by cowdogman »

Lee_WSP wrote: Sun Jul 25, 2021 5:16 pm On the one hand, if you're going to end life with zero or you need to leave your heirs some non tax deferred dollars, doing a little more aggressive Roth conversions makes sense. But if on the other hand, you're going to die with too much and you don't care about leaving your heirs tax deferred dollars, putting off taxes as long as possible may make more sense especially if you have other income adjusted payments to consider.
Well put. I'm definitely in the second category. If I could do something that cost nothing and was risk free (or close to them) to reduce the tax burden on heirs I would.

I also find these Roth conversions similar to the Social Security at 62 or 70 discussions. The choice often reflects personality as much as logical argument. (I'm in the 70 camp.) I would just have a very hard time voluntarily paying taxes early to do a Roth--at least above 12/15%.
User avatar
Lee_WSP
Posts: 10346
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Lee_WSP »

cowdogman wrote: Sun Jul 25, 2021 6:02 pm
Lee_WSP wrote: Sun Jul 25, 2021 5:16 pm On the one hand, if you're going to end life with zero or you need to leave your heirs some non tax deferred dollars, doing a little more aggressive Roth conversions makes sense. But if on the other hand, you're going to die with too much and you don't care about leaving your heirs tax deferred dollars, putting off taxes as long as possible may make more sense especially if you have other income adjusted payments to consider.
Well put. I'm definitely in the second category. If I could do something that cost nothing and was risk free (or close to them) to reduce the tax burden on heirs I would.

I also find these Roth conversions similar to the Social Security at 62 or 70 discussions. The choice often reflects personality as much as logical argument. (I'm in the 70 camp.) I would just have a very hard time voluntarily paying taxes early to do a Roth--at least above 12/15%.
Adding to the above. I've been saying it for the past few days. It's a tax arbitration play. But... we may not live long enough to see it play out. Hopefully we will, it shouldn't take too long if you convert in the 12%, but avoid the 22% bracket. Converting in the 22% to avoid the 22% is.... a wash unless 22 turns into 25. Converting in the 22% to save you from the 24% is not very much of a difference, but if you believe that the 24% is going to turn into 28%, then you get a better arbitrage rate.
User avatar
Topic Author
WoodSpinner
Posts: 3499
Joined: Mon Feb 27, 2017 12:15 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by WoodSpinner »

cowdogman wrote: Sun Jul 25, 2021 5:51 pm
WoodSpinner wrote: Sun Jul 25, 2021 5:04 pm
Well, since most of the ending portfolio amounts are within a few % of each (see post above) other it seems that looking at the costs from taxes to achieve them is worthwhile.

IRR of return is calculated using:
Delta Taxes paid (Fed, IRMAA, State) (e.g. delta of expected taxes each yearConversionScenario - NoConversions) with the Delta Tax Adjusted Portfolio Value as the last value in the array. Used the Excel IRR function to calculate.

WoodSpinner
Sorry, maybe I wasn't clear. What value do the return percentage (ROI and IRR) add to the various dollar amounts--portfolio value, cumulative taxes, whatever? If you find them helpful, that's really all that matters, but the dollar amounts would seem to trump any return calculation.
Not sure of the confusion….

Did you see the numbers posted above (to FiveK)?

There isn’t much difference between many of them. So if I want to look a bit deeper and understand how much each scenario cost in taxes, This lead to using RoI and IRR calculations.
Regarding the IRR, thanks. So "=IRR([array of years, with the first years being Roth conversion taxes (expressed as negatives) and then a number of years at $0],[portfolio value delta])"?
Not quite, I take the yearly delta of taxes No Conversions - Conversion Scenario and then the final delta Conversion Scenario Tax Adjusted Portfolio Value - No Conversion Tax Adjusted Portfolio Value for the input to IRR.

This represents the rate of a return on the Delta from No Conversions.

Have you thought of running the same calculation using [delta cumulative taxes] in place of [portfolio value delta]? Ditto for the ROI calculation.
Easy enough to do but don’t see the value….

The IRR is calculating how much the investment of the taxes paid early for the Roth conversions is delivering. I think I am using it appropriately but want to make sure,
I'm trying to figure out why your numbers are so different from mine. I'm guessing CA state taxes (I'm in WA with no state income tax) and I plan to rely solely on our taxable account for any spending gap above SS and RMDs. So now I'm wondering whether I should add a potential future WA state income tax to my spreadsheet.
We might have to share our models and look under the hood at our assumptions for Portfolio Value, Income, Taxes, IRMAA, Cashflow and Asset Locations to better understand.

Will work on an anonymizes copy that can be shared, perhaps you could do the same? We could exchange and try to better understand.

Thanks for your insights…

WoodSpinner
WoodSpinner
Exchme
Posts: 1323
Joined: Sun Sep 06, 2020 3:00 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Exchme »

WoodSpinner wrote: Sun Jul 25, 2021 3:35 pm I have also found significant benefit to the optimization of asset location and plan to post in this in a separate thread. Not only is the the IRA growth slowed but the equities moved to Roth are now growing tax free. I am still trying to figure out how to quantify the benefit— did you do any work along those lines?

I found the same thing in my case, combining asset location optimization with doing Roth Conversions provided more benefit and cut the optimum amount of Roth Conversions, in my case the optimum amount of conversions was cut in half.

The wiki suggests that asset location optimization looks good due to a stealth increase in stock allocation since tax deferred is being filled with bonds (so the government's share becomes bonds not stocks, leaving your share more stocks and less bonds). I checked that effect between my 1st and 2nd place contenders and the stealth asset allocation effect was real, but much smaller than the differences between cases, so I think I found the right optimum (given the many life, tax and return assumptions!).

I do all my evaluations from the perspective of maximizing my estate if my heirs' were my financial clone, so evaluate after heir liquidation. That at least sets a metric that I can use. For comparison purposes, I converted those future values back to present values and compare to the total of the Roth Conversions.
User avatar
Lee_WSP
Posts: 10346
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Lee_WSP »

It's a question of paying today what may be put off until tomorrow or even ten years or more after death.

If you're able to defer it until after death, then you only care if you care about your heirs. If you do care, it is still a tax rate today vs tax rate for them comparison.

I do not believe there is any rate of return on such a decision as it is a tax arbitrage decision and the only "return" is in paying a lower tax rate.
User avatar
cowdogman
Posts: 2054
Joined: Sat Dec 16, 2017 6:44 pm
Location: Washington State

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by cowdogman »

WoodSpinner wrote: Sun Jul 25, 2021 6:35 pm
Regarding the IRR, thanks. So "=IRR([array of years, with the first years being Roth conversion taxes (expressed as negatives) and then a number of years at $0],[portfolio value delta])"?
Not quite, I take the yearly delta of taxes No Conversions - Conversion Scenario and then the final delta Conversion Scenario Tax Adjusted Portfolio Value - No Conversion Tax Adjusted Portfolio Value for the input to IRR.

This represents the rate of a return on the Delta from No Conversions.
I think we're saying the same thing. But I'll try the calculation.
WoodSpinner wrote: Sun Jul 25, 2021 6:35 pm
Have you thought of running the same calculation using [delta cumulative taxes] in place of [portfolio value delta]? Ditto for the ROI calculation.
Easy enough to do but don’t see the value….

The IRR is calculating how much the investment of the taxes paid early for the Roth conversions is delivering. I think I am using it appropriately but want to make sure,
I had a similar discussion with someone else last week.

Doing investment return calculations on Roth conversions takes a little shoehorning. You have to decide what the investment is and how you calculate profit. I'm not sure there is only one correct way of doing a return calculation for Roth conversions, but that doesn't matter if the calculation you use is helpful. I do think using incremental taxes on the Roth conversions as the "investment" makes sense but there are at least a couple amounts you can use as profit (loss): the delta in portfolio value and the delta in cumulative taxes. I keep track of both (as dollar amounts, not as ROI/IRR) and I find it interesting that in a lot of scenarios using various ages one will be negative and one will be positive. For example, if I do Roth conversions but then never withdraw/use the money in the Roth account, at some age (usually around 92) my Roth conversion will go into the black on cumulative taxes (that is, my cumulative tax bill will be lower with conversions after 92) (because of the benefit of lower RMDs) tho my portfolio value never recovers and the negative delta continues to increase until the end of my spreadsheet.

May I ask how you come up with the tax-adjusted portfolio value?
MathWizard
Posts: 6542
Joined: Tue Jul 26, 2011 1:35 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by MathWizard »

My spouse and I are exactly one year older.
Also at the top of the 12% bracket.

We have 1.5 million on tax deferred and another 600K in Roth and taxable.

There is no easy rule of thumb.

1. The correct way to model this is to make a spreadsheet which computes taxes both state and federal taking into consideration tax brackets, health care , and includes RMDs. This is a bit complicated and is dependent on the state you live in, and how long each of you expects to live.

2. I plan using age 85 for my demise, after which my spouse will file single for another 15 years. Both are estimated based on family history, with slightly longer lifespans because we are wealthier than our families were growing up, with better healthcare.

3. The big benefit comes when my spouse is a widow, filing as a single. Without these, taxes go up dramatically, pushing her well into the 25% bracket, with 85% of SS as taxable income.
As I have planned it, RMDs have a mild effect pushing her to a maximum marginal rate of 18.5% combined tax rate , eventually becoming 15% in tax deferred withdrawals. At that point, almost all money is in Roth.

This allows about 1500 extra in discretionary spending, and leaves our kids Roth IRAs rather than tax deferred , so that it has no impact on their taxes.

It does require Roth conversions up to the top of the 12% or 15% bracket to age 78 to get this done.
babystep
Posts: 763
Joined: Tue Apr 09, 2019 9:44 am

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by babystep »

cowdogman wrote: Sun Jul 25, 2021 11:52 am
babystep wrote: Sun Jul 25, 2021 10:56 am
cowdogman wrote: Sun Jul 25, 2021 10:51 am
babystep wrote: Sun Jul 25, 2021 1:27 am
WoodSpinner wrote: Sat Jul 24, 2021 8:42 pm
What I am seeing is that the various Metrics simply aren't working as an effective tool to make a selection. There is significant variation based on time period (e.g. 71-80 vs. 81-90. I am also getting conflicting signals from the various metrics as to which is the better scenario.

At this point I need some advice on which metric(s) I should really be prioritizing.
I don't think we can look at different time periods. One key characteristic of Roth conversion is to take advantage of low tax rate years but still being conservative in estimating your TDA growth so that you don't cross the next marginal rate during RMD and SS time periods.

I think it needs to include all the time periods including heirs and charity, if applicable.

Maybe total money in real dollars available to you/heirs/charity post all the taxes are paid?
I think time periods are very relevant. Whether you look at Roth conversions as an investment or a bet on future tax laws, determining when you have recouped your investment or won the bet is important. And in that context one very important question to ask is whether you are doing Roth conversions for your and your spouse's benefit only or also for your heirs (or somewhere in between--e.g., "primarily for us but also for our heirs").

As to what time period to focus on, look at all of them, but if your Roth conversion strategy is 100% centered on you and your spouse (and not your heirs) focus on your expected mortality date (with the hope that you exceed it). If your Roth conversion calculations tell you will recoup your investment five years after your expected mortality, then it may not be worth making the bet.
The time periods are indeed relevant but OP is trying to figure out metrics for comparison. The problem is the variation in metrics for different time periods (see the bold part in OP) which is somewhat by design for Roth conversions hence the holistic comparison needs to have full time duration.
Understand your point (I think) but almost nobody lives the "full time duration"--to the end of the RMD table--115 years old. The metrics do vary by age a lot--I've made that point in various posts. And as I said above we should look at various ages in the analysis but I would focus on the expected mortality.

Edit: Plus I think the fact that the metrics vary by age is very important to understand when doing a Roth analysis. There is a long pay back period to the initial investment in Roth conversions--that is, the investment of the incremental tax paid for the conversions.
Bold emphasis added. That is what I meant by full time duration.
User avatar
Topic Author
WoodSpinner
Posts: 3499
Joined: Mon Feb 27, 2017 12:15 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by WoodSpinner »

Exchme wrote: Sun Jul 25, 2021 7:37 pm
WoodSpinner wrote: Sun Jul 25, 2021 3:35 pm I have also found significant benefit to the optimization of asset location and plan to post in this in a separate thread. Not only is the the IRA growth slowed but the equities moved to Roth are now growing tax free. I am still trying to figure out how to quantify the benefit— did you do any work along those lines?

I found the same thing in my case, combining asset location optimization with doing Roth Conversions provided more benefit and cut the optimum amount of Roth Conversions, in my case the optimum amount of conversions was cut in half.

The wiki suggests that asset location optimization looks good due to a stealth increase in stock allocation since tax deferred is being filled with bonds (so the government's share becomes bonds not stocks, leaving your share more stocks and less bonds). I checked that effect between my 1st and 2nd place contenders and the stealth asset allocation effect was real, but much smaller than the differences between cases, so I think I found the right optimum (given the many life, tax and return assumptions!).

I do all my evaluations from the perspective of maximizing my estate if my heirs' were my financial clone, so evaluate after heir liquidation. That at least sets a metric that I can use. For comparison purposes, I converted those future values back to present values and compare to the total of the Roth Conversions.
Could you tell me more? Or provide some examples? I want to make sure I understand your approach!

Thanks

WoodSpinner
WoodSpinner
User avatar
Topic Author
WoodSpinner
Posts: 3499
Joined: Mon Feb 27, 2017 12:15 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by WoodSpinner »

Lee_WSP wrote: Sun Jul 25, 2021 8:01 pm It's a question of paying today what may be put off until tomorrow or even ten years or more after death.

If you're able to defer it until after death, then you only care if you care about your heirs. If you do care, it is still a tax rate today vs tax rate for them comparison.

I do not believe there is any rate of return on such a decision as it is a tax arbitrage decision and the only "return" is in paying a lower tax rate.
By paying taxes early and moving assets to a more optimal location (at least in my situation) it is more than a tax arbitrage. I am reducing the overall tax bill and increasing the tax adjusted portfolio value.

The tax arbitrage aspect is important but it isn’t a metric that will help me select among the alternatives I am considering. They all have positive tax arbitrage.

WoodSpinner
WoodSpinner
User avatar
Lee_WSP
Posts: 10346
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Lee_WSP »

WoodSpinner wrote: Sun Jul 25, 2021 8:58 pm
Lee_WSP wrote: Sun Jul 25, 2021 8:01 pm It's a question of paying today what may be put off until tomorrow or even ten years or more after death.

If you're able to defer it until after death, then you only care if you care about your heirs. If you do care, it is still a tax rate today vs tax rate for them comparison.

I do not believe there is any rate of return on such a decision as it is a tax arbitrage decision and the only "return" is in paying a lower tax rate.
By paying taxes early and moving assets to a more optimal location (at least in my situation) it is more than a tax arbitrage. I am reducing the overall tax bill and increasing the tax adjusted portfolio value.

The tax arbitrage aspect is important but it isn’t a metric that will help me select among the alternatives I am considering. They all have positive tax arbitrage.

WoodSpinner
How so? How are you increasing the tax adjusted portfolio value? If the marginal rate on the amount converted and the tax deferred amount are the same, there is no difference.

If otoh, you paid a lower rate, then you paid less taxes.
User avatar
Topic Author
WoodSpinner
Posts: 3499
Joined: Mon Feb 27, 2017 12:15 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by WoodSpinner »

Lee_WSP wrote: Sun Jul 25, 2021 9:04 pm
WoodSpinner wrote: Sun Jul 25, 2021 8:58 pm
Lee_WSP wrote: Sun Jul 25, 2021 8:01 pm It's a question of paying today what may be put off until tomorrow or even ten years or more after death.

If you're able to defer it until after death, then you only care if you care about your heirs. If you do care, it is still a tax rate today vs tax rate for them comparison.

I do not believe there is any rate of return on such a decision as it is a tax arbitrage decision and the only "return" is in paying a lower tax rate.
By paying taxes early and moving assets to a more optimal location (at least in my situation) it is more than a tax arbitrage. I am reducing the overall tax bill and increasing the tax adjusted portfolio value.

The tax arbitrage aspect is important but it isn’t a metric that will help me select among the alternatives I am considering. They all have positive tax arbitrage.

WoodSpinner
How so? How are you increasing the tax adjusted portfolio value? If the marginal rate on the amount converted and the tax deferred amount are the same, there is no difference.

If otoh, you paid a lower rate, then you paid less taxes.
Take a look at the metrics posted in response to FiveK. Yes the Tax Adjusted Portfolio Value is increased and the overall tax bill is lowered.

WoodSpinner
WoodSpinner
User avatar
Lee_WSP
Posts: 10346
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Lee_WSP »

WoodSpinner wrote: Sun Jul 25, 2021 9:08 pm
Take a look at the metrics posted in response to FiveK. Yes the Tax Adjusted Portfolio Value is increased and the overall tax bill is lowered.

WoodSpinner
It's your retirement and accounts, I'm not going to tell you how you should think of them. However, I do not agree with your point of view as far as ROI goes. I think of it as a cost reduction.
User avatar
Nestegg_User
Posts: 2109
Joined: Wed Aug 05, 2009 1:26 pm

Re: Metrics to Compare Roth Conversion Scenarios ??

Post by Nestegg_User »

(deleted)
Last edited by Nestegg_User on Sun Jul 25, 2021 9:43 pm, edited 1 time in total.
Post Reply