I've pasted the abstract and conclusion portions below due to it's length, but an interesting read for anyone contemplating conversions. I would also recommend reading the very short section on "the myth of tax diversification" on p.18. Full study attached.
Much has changed since penalty-free Roth conversions were inaugurated in 2010. Tax rates have gone up and down. The re-characterization provision went away. Heirs can no longer stretch out inherited Roth accounts over a lifetime. Medicare surcharges were expanded and began to adjust for inflation. The age to begin Required Minimum Distributions was pushed out to age 72 and the IRS changed the RMD divisor tables to further slow the pace of distribution. Given these developments it seemed worthwhile to re-examine the rationale for Roth conversions. That effort exposed multiple 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, i.e., during a year with no other taxable income.
While the benefits from a Roth conversion are often small and slow to arrive, a Roth conversion will almost always pay off if given enough time, i.e., for life spans that extend past 90 and so long as annual distributions from converted amounts are not taken. Roth conversions work because of compounding, which requires the conversion to be left undisturbed for a long time. The paper elucidates the role played by the mathematics of compounding in underwriting the success of Roth conversions. This paper identified pitfalls and snares that undercut construction of an exact counterfactual to a Roth conversion, including incorrect treatment of the side fund, failure to invest in the same asset across tax structures, neglect of the time value of money, and avoidance of the implications of a progressive tax rate structure that adjusts for inflation.
Conclusion:
A conceptual foundation for Roth conversions was offered in terms of intertemporal tax arbitrage: the potential gain from paying some tax today to pay less tax in future years. The sense from the literature was that under constant tax rates, there could be no gain from intertemporal tax arbitrage. This paper showed that intuition to be mistaken. The payoff from conversion comes slowly, but it did come before age 100 in most of the situations examined. However, although a Roth conversion left untouched is almost certain to pay off if the client lives long enough, the payoff is typically modest, the more so before the 90s, in all but a few of the most favorable scenarios.
Future tax hikes were found to confer an unexpectedly small benefit on a timely conversion. Roth conversions did indeed fare better when future tax rates were made to increase, but not by much; likewise, if client and advisor guessed wrong, and future rates were cut moderately rather than hiked, Roth conversions still paid out, again in almost every case.
Intertemporal tax arbitrage does not require a future tax hike to pay out. Instead, the success of a Roth conversion is guaranteed by the mathematics of compounding. Assets in a totally tax-free structure must compound faster than those same assets in an account subject to even the most minimal tax drag. It may take decades, but given decades, faster compounding always overcomes the initial tax debit.
The current focus in the literature, which drives the Roth conversion decision off the arithmetic difference between present and future tax rates, proved to be misguided. More important than the difference between present and future tax rates is the absolute level of the tax rate applied to the conversion. It is trivial to show that a Roth conversion made at zero percent is likely a good idea; more important was the insight that if the 0%, 10%, and 12% brackets are filled by the conversion, the average rate still remains about 8%, relative to a future rate that might be 3X as large. This scenario produced the greatest and quickest payoff of any examined. Unfortunately, under ordinary circumstances few affluent professional couples, the focus of this paper, will have a year where they can fill the 0%, 10% and 12% brackets with a conversion, and still cover their living expenses without drawing any other taxable income.
The analysis also revealed that top bracket taxpayers do not benefit from Roth conversions to any greater degree than middle bracket taxpayers, even as they are more at risk of slower and proportionately smaller payoffs if they should guess wrong about the level of future rates. The problem is the depth of the hole dug by converting at today’s top rate—the inverse of the happy outcome that issues from converting in today’s bottom tax brackets.
With the mechanism underlying Roth conversion outcomes made clear, advisors can proceed to address individual cases within their practice. The key questions to ask before recommending a Roth conversion include: 1) Is the client on track to accumulate surplus funds in their TDA? 2) Does the client have the patience to wait until after age 90 to see the payoff? 3) Can their affairs be arranged to make a conversion in the 0% bracket? A negative answer to all three questions would indicate the need to examine alternatives, such as backdoor Roth IRAs or life insurance products. A positive answer to the third question gives a green light to proceed.
Updated for 2022:
The conclusions generally hold up in the revised analysis. In particular, the role of compounding tax drag remains central, but is clarified further in the new spreadsheet. However, the precise metrics used in the tables in this paper have been superseded, even as their pattern remains roughly the same across cases. There are two changes of note that emerge from the new spreadsheet.
1. The new analysis found the benefits of paying tax outside the conversion to be stronger than indicated in this paper
2. The new analysis also found an advantage for converting earlier rather than later.
Both restatements can be interpreted in terms of the basic theme of this paper: that compounded tax drag is key to understanding the payoff to be expected from Roth conversions. Because tax drag is exponential, it is more important than the arithmetic difference between conversion tax rate and the future tax rate on RMDs. And again because tax drag is exponential, Roth payoffs are slow to mount but can eventually become quite large.
I bookmarked but didn't read thoroughly yet due to trying to get my book out the door. Does he assume retirement at 65 and starting conversions then? Neither of those is the case for many people. Quite a few of my clients retire earlier and convert away until they start Social Security and retirement distributions. They live on some side-hustle money doing fun things part-time, brokerage accts, older/eligible Roths to create very low income tax brackets.
Another incentive is that a surviving spouse will almost certainly be taxed with much smaller/higher single tax brackets, and trying to protect them, and/or pay taxes to heirs as a gift. Doing Roth conversions when younger and having them grow tax free over long time periods can be a hedge against longevity risk (living longer than you planned). Also against tax rates going up due to TCJA expiring, and to bail us out of a financial apocalypse if we don't stop kicking the can down the road.
Studies are great, but in the end, we all have our own personal high-fidelity "study" when we put our info into PRC and look at the net worth at different stages of life when doing/not doing conversions to see what the effect is on our precise situation.
@hines202 Bill I think most of your questions are answered above. Its more of a quantitative assessment than looking at individuals, although he makes it clear that a Roth can pay off, provided the person is willing to wait until age 90 or beyond based on historical data and changes in the mechanics of Roths over many years since introduced. You're talking more about the reasons WHY someone would want to consider a Roth, whereas he's presenting the metrics pertaining to whether they actually pay off, which is very different. My own experience supports his theory...I don't experience a "break even" on conversions until 4 years after my anticipated end of life (which happens to coincide with age 90, ironically). he also addresses the "benefit myth" regarding tax rates.
In the end, as an academic, I don't think his motivation leans one way or another. He's merely presenting findings that do poke holes in a lot of the popular thinking and assumptions behind conversions. Consequently, it's up to the investor to determine, given these findings, if a Roth still makes sense for them.
This was debated over at bogleheads.org at the time, with the participation of the author, Prof. McQuarrie. The study is applicable in the narrow case of planning for your lifetime only. But much of the driving force for Roth Conversions is that your heirs don't have to pay taxes on the Roth. For simplicity, Prof. McQuarrie did not address that case. Also, he values a $ in tax deferred equal to a $ in Roth. That can be true in some circumstances like QCDs and bequests to charity and nearly true if you need long term care, but far from true if you withdraw the money from tax deferred for a non-deductible purpose.
While he says those things in the paper, the framing throughout leaves a bigger and I think unhelpful message that there is some "payout" period of time below which you shouldn't do them.
Pralana lets you look at Roth Conversions either way. You can consider the eventual taxes likely to be owed on the residual tax deferred accounts or not (effective $ vs. absolute $). So no need to try to decipher the paper that is necessarily full of generalizations, inventing case studies for people that are not you, you can just test it yourself.
If you plan on living past 80 years old as I do, I really don't see a downside to ROTH conversions (assuming you stay in lower tax brackets when you do them). Worst case is you break even in terms of total money at end of life. Roth's do give you flexibility in case opportunities arise such as a big stock market down turn when you can convert at lower values. Or if tax rates really do go up. Or in my case being able to significantly reduce our MAGI so that we get large ACA (Obama Care) tax breaks and $0 deductibles. That last move also means we will not (in 2023) have to pay any State or Federal income taxes because our AGI will be less then the standard deduction for a married couple. Last year (2022) we got an ACA health plan that was health saving account (HSA) compliant which only covered MAGI that were above 250% of the poverty level at a premium of about $750/month plus deductibles and $12,000 total family out of pocket. I was laser focused on getting and maxing out our HSAs. Well, this year the premiums for the same plan went up to about $850/month. Wife said expand your horizons dude 😍 and look at non-HSA plans. Low and behold we got a silver plan for MAGIs between 133% and 150% of the poverty level (we set our MAGI at 145%) and premiums dropped to $95/month with $0 deductibles and $3,000 total family out of pocket. So between the drop in premiums and no Fed/State tax's due, we saved about $11,000 for 2023. Plan to repeat in 2024 & 2025, after which I will be on Medicare. Not bad and thank you very much Roth's 🤑 🤑 🤑
@ricke @pizzaman @ricke @nc-cpl @hines202
I agree that we should use Pralana to make our own decision. For me, Roth conversions are a marginal decision (especially considering all the uncertainties in my assumptions), but I like the idea of having money in a tax free account. This could be part of a future legacy for my kids and they will appreciate not having to pay taxes on it.
I use Pralana to explore trade-offs between each alternative associated with a given decision (like Roth conversions). As reality plays out year by year, I apply my own judgement to how much (zero+) I want to convert in any given year. For example, I used Pralana to help me make a decision on when to take my company pension (now or at some future date, lump sum or monthly benefit). I decided to take my company pension as a monthly benefit this year because the segment interest rates that are used to calculate the monthly benefit have increased significantly. An additional trade-off is now I will not do as large a Roth conversion this year.
Prof. McQuarrie does some interesting calculations that I think are worth understanding if you are into this stuff. Here is a link to an online seminar where he presents his work. Some might find it interesting. I think it includes the 2022 updates.
@golich428 Interesting video. What I got out of it is that Roth conversions can act as an answer to longevity risk.
This was posted as "food for thought," and not to push anyone in one direction or another. Roth's feel like one of those "soft choices" we are presented with...maybe it will pan out as a good move, or maybe it won't. Because the success of a conversion is subject to so many uncontrollable variables, it helps to have a deeper examination of it to pin down whether it's right for one person vs another. Some prioritize paying off their mortgage primarily because it feels good, even if the numbers aren't necessarily supportive of that decision. Some probably choose to convert for the same reason, regardless of data (i.e., wanting to leave tax-free assets to heirs or just have a stash of tax-free money - we'll call them the Qualitative Investor). Others base their decision on hard numbers and research such as this (the Quantitative Investor). Prof. Quarrie's analysis is interesting in that it reexamines some of the beliefs we've had about Roth conversions in an effort to stress test them and look at them within the context of today's financial environment. Regardless of whether you consider yourself a quant or qual type, his analysis is (I think) a way to look at Roths that the large financial companies don't offer.
