The uneven landscape of the tax code means that the optimum conversions may well be an uneven plan you would never have thought of. Our Roth Conversion plan is very lumpy as well.
Some numbers to illustrate:
If you need ACA, then once your income goes above 2-2.5 x Federal poverty level, the premium tax credit phase-out ramps mean your marginal rate can be an additional 12-16% above your ordinary income tax bracket, until you reach 4 X FPL, where it is only 8.5% above your ordinary income bracket. So you can seem to be in the 10-12% bracket, but have marginal tax costs of ~25-27%.
If your conversions take you just above the start of the LTCG tax phase-in, you are paying the 12% ordinary income bracket plus the 15% LTCG, for a 27% marginal rate until all your LTCGs are taxed.
For singles receiving SS, it is possible to simultaneously be in the LTCG phase-in and the SS benefit taxation phase-in. So you can be in the 27% LTCG phase-in zone X 1.85 for SS taxes, for a punishing 49.95% marginal rate for a small range of income (plus state taxes if you have them).
Because paying taxes on Roth Conversions reduces your taxable account, you effectively have put the taxes you paid into your Roth, putting them beyond the reach of the IRS. That reduces future tax drag, the same way getting rid of advisor fees helps. So Pralana will favor doing more conversions early to pull down your taxable account, even if the tax landscape is perfectly flat.
At higher incomes, once you turn 63, IRMAA is like an additional ~5% tax if you go over an IRMAA tier, but that occurs in hard to manage cliffs. If your finances are in the zone where Pralana calculates you will be paying IRMAA someday, it will favor doing bigger conversions prior to age 63 to minimize that.
TCJA expiration will add 3-4% percentage points to the tax rates, so if you tell Pralana that TCJA will expire on schedule, it will favor doing larger conversions in 2024 and 2025 to beat Congress to the punch.
Pralana not only helps you decide on whether to do enough Roth Conversions to "push through" these phase-in zones using your estimated future earnings and returns, I also like to test the Roth plan in bull and bear markets. On the Analysis-Run Analysis tab, click the white box just below the graph that says "Historical Sequence Analysis is inactive, click to enable" and select a Historic Start year to the right of that click box. Then go to Tabular Projections - Summary and look at the final estate value (or any year you choose).
Then go to the Analysis-Roth Conversion tab and make a change to your Roth plan, and go back to Analysis-Run Analysis and click the same button again to enable historical sequence analyses and again return to Tabular Projections - Summary to see the new final estate value. While there are a lot of steps to see the change, this is a unique feature of Pralana that allows you to test how your Roth plan would have worked in bull and bear markets.
Considering the 15 years (60%) between 2002 and 2026 when tax rates were lower than expected I've been of the opinion that pre-paying a future tax liability is a mistake given the capricious nature of Congress.
@kkrupka However, I was still open to conversions until I stumbled upon this article.
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3860359
The Author provides argument exposing these flaws in conventional wisdom:
• Future tax rates need not be higher for a conversion to pay off;
• Nor is it all that helpful to pay the tax on conversion from outside funds;
• Nor are Roth conversions especially beneficial for top bracket taxpayers as compared to middle class taxpayers;
• Rather, the greatest benefit accrues to taxpayers who can make the conversion partly in the zero percent tax bracket.
@kkrupka As others have said, there are lots of variables that can make algorithmic tools suggest conversions that seem counter-intuitive. I've seen it when couples have a big age difference - obviously the thing being avoided is the many years likely in the widow/widower penalty single tax bracket.
PRC is a long-range planning tool, so it's laying out a roadmap that is hazier over the long haul, certain to be corrupted by future legislation and other factors (fate). Do a pin-point Roth analysis toward the end of each year when your tax situation is pretty locked in and you know what you're dealing with. Factor it against other strategies that may have a higher priority for you, such as gain or loss harvesting, etc.
@hecht790 I made a couple manual changes and have been able to smooth things out. So I'm happy with that. But now I'm wondering, should I be comparing effective tax rate, not marginal tax rate, because I could be $10 over in marginal bracket or $50K over in marginal bracket?
@ricke I never tried the historical analysis because I was confused on how to do that. But now that I did, it does give a very interesting perspective. Thank you.
Converting IRA to Roth is taxed by the marginal (the actual tax that you pay for the conversion) regardless of $10 or $50K over. Big conversion may fill the marginal bracket and go into the next bracket.
@ricke can you explain your statement: If your conversions take you just above the start of the LTCG tax phase-in, you are paying the 12% ordinary income bracket plus the 15% LTCG, for a 27% marginal rate until all your LTCGs are taxed.
Wouldn't it be 12% plus 15% divided by 2 for a marginal rate of 13.5%??
Each $ of Roth Conversions above the LTCG phase-in threshold punishes you twice - once as it is taxed as ordinary income at 12% and once as that same $ of Conversions adds to your AGI, so pushes a $ of LTCGs above the AGI threshold for that 15% tax. Since the LTCG AGI phase in start and top of the 12% ordinary income brackets align within almost perfectly, the LTCG phase-in start can be thought of as the true top of the 12% bracket. Then there is a 27% bracket until all the LTCGs are taxed, and then you are into the 22% ordinary income bracket.
If you think about phase-ins, you are going from zero to the normal tax rate, so they have to create zones where the marginal tax is high to catch up. That's why Pralana has the selection for LTCG taxes on the Roth conversion page, it can be a powerful place to stop.
Kitces, among others, has written about it, here is one:
Kitces.com/blog/long-term-capital-gains-bump-zone-higher-marginal-tax-rate-phase-in-0-rate/
To give some examples of the other common phase-outs that create high "pseudo" brackets for folks - the ramps in the ACA premium credit phase out are variable, but over a range are 14-16% AGI taxes on top of the federal 10% rate, so that's a problem for folks on ACA, a $ of Roth Conversions for them creates 25-27% marginal tax costs.
Taxes on SS benefits are phased in at two different speeds 1.5X the ordinary income bracket and then once that is full, it steps up to 1.85 x the ordinary bracket until the maximum of 85% of the SS benefit is taxed. For single folks, they can simultaneously be in the 12% ordinary income bracket + the 15% LTCG bracket and the 1.85X SS phase-in rate. So, over a narrow range, a $ of Roth Conversions for them can cost (12%+15%) x 1.85 = 49.95%! Plus whatever their state demands.
PRC is a long-range planning tool, so it's laying out a roadmap that is hazier over the long haul, certain to be corrupted by future legislation and other factors (fate). Do a pin-point Roth analysis toward the end of each year when your tax situation is pretty locked in and you know what you're dealing with. Factor it against other strategies that may have a higher priority for you, such as gain or loss harvesting, etc.
Pin-point analysis using Pralana or a calculator? One problem is Pralana starts at current year, so taxes for previous year done in March aren't shown. (Or do you set the start of plan a year before?) Also it is rare the interest/ltcp etc for a given year matches the average long term average one enters into Pralana.
A big catalyst to cause aggressive Roth conversions to be recommended are your stated projected returns for stocks and bonds. If those are fairly high, and especially if you're on the younger side, the algorithm is seeing lots of fuel for big tax free growth, above and beyond the disadvantage of paying the taxes now. Just make sure your projected returns are realistic, and take the optimizer results with a grain of salt. Often with my clients, the conversions aren't a make or break thing, just gravy and protection against future tax hikes, losing a spouse early, and ability to stay in lower tax brackets, etc.
Here is a link to a podcast/video by Sensible Money (one of my all-time favorite web sites and person) that talks about all the things to consider when deciding whether or not to do a Roth conversion. Stuart, does PRC take most of these things into consideration?
https://www.sensiblemoney.com/learn/roth-conversions-when-is-enough-enough/
Here is a link to a podcast/video by Sensible Money (one of my all-time favorite web sites and person) that talks about all the things to consider when deciding whether or not to do a Roth conversion. Stuart, does PRC take most of these things into consideration?
https://www.sensiblemoney.com/learn/roth-conversions-when-is-enough-enough/
Nice video pizzaman. My summary of the video and how it applies to Pralana, fwiw:
In this interview Dana Anspach makes some nice points about Roth Conversions that I'd summarize as "the devil is in the details".
The good news is I think Pralana is directly in her spirit of "let's do the math and see", and, in fact, incorporates most of the issues she addresses in the video! Namely, the Roth conversion optimizer takes into account these specific features she highlights:
- RMD effect on capital gains and other taxes
- ACA credits -- (Yay Pralana!)
- social security taxes (Yay Pralana!)
- irmaa and medicare B/D tax (Clap for Pralana-timing if a bit off)
- asset allocation -- Pralana is a "closed system", so any money not put into a Roth will be reinvested into accounts, and any money converted into Roths will grow tax free at that higher expected return.
- hedging against changing tax rates -- user can set explore in Build>Scenario Assumptions.
(Aside: Several of these are hampered, though, by the fact Pralana doesn't withdraw taxes until the following year, preventing precise roth conversion analysis like she does at her firm. When Pralana eventually improves its iterative taxation programming this will not be a problem.)
Two of Anspach's three overall frames for evaluation are incorporated into Pralana:
- Liquidation Value of the IRAs by the user or their heirs. This is addressed in Pralana via the user entering an "effective tax rate" that will be used for this purpose, for self or heirs depending on context. By clicking "effective" the roth conversion will take this final tax rate into account. (I think Pralana's name for this feature--"effective tax"-- is a bit confusing fwiw.)
- Breakeven year -- The graphs show this, and Bill Hines @hines202 has made this point repeatedly: if you see a small benefit or it's too far in the future, is it worth it?
Her third evaluative frame is missing from Pralana,
- 3. Present value of required cash flows-- "The amount of withdrawals a person has to take out over their lifetime including what they to withdraw to pay tax."
The time-value-of-money is not a concept that Pralana utilizes. But upon reflection I don't think the lack of this affects the issue at hand of Roth conversions in Pralana. Why? Because Pralana is a "closed system", so any money not put into a conversion will be reinvested into other accounts, and any money converted into Roths will grow tax free at that higher expected return. In other words, the time-value-of-money is implicitly built into Pralana and the Roth conversion process.
So in short, I think Pralana passes the Anspach test's on all counts.
Good research article on Social Security claiming strategy including Roth conversions:
Group 4
A fourth group consists of married couples when one spouse dies at least one calendar year before the other spouse. In other words, it consists of most married couples. Suppose one spouse (assumed male for clarity) of a married couple dies in 2030. Beginning in 2031, the surviving widow will face tax brackets for singles instead of those for married couples filing jointly. So, her tax bracket (and MTR) will likely be higher than the couple’s tax bracket (and MTR) was when both spouses were alive. In 2033, she will face Medicare premium income threshold levels for singles that are generally half as high as the threshold levels for married couples. So, she may face multiple IRMAA spikes in her Medicare premiums each year beginning in 2033. In fact, beginning in 2033 and continuing for the rest of her life, her annual Medicare premiums due to these IRMAAs could exceed their joint annual premiums when both partners were alive. The lesson is that married couples should consider making Roth conversions while both partners are alive to reduce both the size of their joint lifetime income taxes and the size of their joint lifetime Medicare premiums. See Reichenstein and Meyer (2021b) for additional discussion for this group.
