By Dr. Leif Dahleen, Columnist
The prevailing thought was that there would be a one-year window in which high earners could legally do a Roth conversion. From one of the many articles on the subject from back then:
“2010 is the only year that this exception is allowed. Starting in 2011 (unless congress changes it), it reverts back to the original way where all the tax is owed in that year.”
My Mega Roth Conversion: A $212,000 Mistake?
Guess what? Our federal government liked the tax revenue from all the 2010 conversions so much that the AGI limit was lifted permanently (for conversions only).
Of course, I didn’t know that was going to happen. I will also admit to being a personal finance novice at the time. I knew that Roth money was good money and I wanted more Roth money. So I filled out some paperwork and converted our traditional IRAs and my SEP IRA. I wasn’t going to let that one year window pass me by!
Let’s talk numbers. My wife and I had traditional IRAs worth a combined $36,000 ($30,000 of which was tax deferred, with $6,000 in non-deductible contributions). The SEP IRA, after 5 years of tax-deferred contributions of $25,000, $45,000, $45,000, $49,000, and $49,000 had grown to $242,000.
To convert the whole lot to Roth, I had to pay income taxes on $272,000. There was a provision (which no longer exists) to spread the tax pain over the following 2 tax years, so we took advantage of those interest-free loans. The conversion was made in 2010, but the taxes were paid in 2012 (2011 return) and 2013 (2012 return).
In 2012, with the addition of $136,000 of “income” from half of the conversion, we paid an extra $53,557 in federal and state income tax. If that $53,557 had remained invested in an S&P 500 fund, we would have seen a 114.2% return from April 2012 to January 2018 according to the S&P 500 Dividends Reinvested Price Calculator. That money with four years of 12.2% annualized returns would be worth $113,000.
In 2013, we similarly added $136,000 to our taxable income on the 1040. The $53,577 we paid in 2013 would have grown 85.1% in three years, a 14.1% annualized return. Had we not paid the tax, we would now have $99,000. Thank you, bull market!
In summary, our calculated net worth would be $113,000 + $99,000 = $212,000 higher today if I hadn’t made the Roth conversion in 2010.
Have I Made a Huge Mistake?
Gob and his huge mistakes.
This is not a simple question to answer. Yes, our net worth would be a larger number if I hadn’t made the conversion. But now I’ve got a lot more of that good Roth money that will (presumably) never be taxed again. More than five years have passed, which means I can withdraw the entire $278,000 penalty-free and tax-free. Not that I would, but I could.
What Would It Cost to Convert Today?
In the 7-plus years since making the conversion in October of 2010, the S&P 500 is up 161%. The $272,000 invested in the S&P 500 would have grown to $709,000 (not precisely because my Roth holds different asset classes including REIT, small cap value, and emerging markets).
Let’s call it $700,000 today. If it were still in the tax-deferred IRAs, to convert it all, I’d be paying the tax bill in one year rather than two. Converting $700,000 at once with a high-income in a high tax state would create a tax bill well over $300,000. Ouch!
Is There a Better Way to Make Roth Conversions?
Yes! Wait until you’re retired (particularly if you plan to retire early). You can convert smaller chunks one year at a time when you are in lower income brackets. A good strategy is to fill the 12% bracket, and perhaps the 24% bracket with Roth conversions. Play around with tax software and see what works in your situation.
So, Did I Make a Huge Mistake?
We make the best decisions we can with the knowledge we have at the moment. When I was 35, the thought of an early retirement hadn’t crossed my mind. I thought 2010 would be my only opportunity to get my hands on more Roth money, before going back to many years of adding $50,000 a year in tax-deferred retirement savings.
Knowing what I know now, having my eye on an early retirement, if I could fire up the flux capacitor and go back in time, I would advise my younger self not to make those conversions. “There will be time to make those conversions in a lower tax bracket when you’re early retired”, I would say. While in hindsight, it was probably not the best decision, I did the best I could with the information I had at the time.
Doc, could I borrow your ride?
The good news is I’ve since become an employee, and have been building up some tax-deferred retirement savings. The 457(b) will be part of our drawdown strategy to help fund our monthly expenses in early retirement. The 401(k) will be rolled over to a traditional IRA, and year by year, slowly converted to Roth. I’ll still be able to take advantage of those Roth conversions in much lower tax brackets down the line. Unless the rules change. Again.
Did you take advantage of the “one-time opportunity” in 2010? What would you have done in my situation? Would you say I made a huge mistake?
Great article. I think this concept is way underappreciated among financial professionals. That is, the opportunity cost of paying taxes and not having the money invested in index funds. Jim Lange addresses this in his books. What I struggle with is that they could change the rules on the Roth, but I am confident that I will get a satisfactory return in an index fund over 10 plus years with money that would have otherwise gone to the IRS. Also, all the money is not taken out at 70.5, only the RMD. The RMD is taxed, but the other money continues to grow tax deferred in the index fund. The different assumptions about losing out on the growth of the money you pay to Uncle Sam versus the advantages of the Roth can vary the result. I guess my question to PoF would be what assumptions and calculations have you made about the partial conversion technique, and how confident are you in those assumptions versus just letting the money grow instead of cutting that tax check to the IRS. Even the bible agrees, a bird in the hand is worth more than two in the bush. That is, the extra money in tax savings today is worth more than a possible (not guaranteed) benefit in the future. In your Sunday news summary you might include Howard Marks memo to clients (if not already done). There is also a great article challenging the 4 percent withdrawal rate in today’s WSJ.
Yes I read that WSJ article. Wade Pfau is always so gloomy.
This question has nothing to do with the magnitude of tax you pay now vs how much you pay later. It is all about the tax rates on the front end compared to those when the money is spent. Frankly James Lange misses the point by muddying the water with taxable accounts as a way to pay the taxes and he gets the math WRONG. He does not follow the amount of money earned vs the amount of after-tax money spent. He also makes two serious mistakes by basing much of his theory on the fact the Bush tax cuts would not be continued – which they were and the tax rates would go up – which they didn’t. Many people converting their retirement money at 15% mostly towards a Roth are now most likely going to be spending less money in the 12% bracket as it is ALL about the tax bracket differential between the front-end and the back-end tax rates.
I like the article as well, but not sure about the conclusion. There is no way to be certain about the returns or the future tax brackets. In addition, many of us perform the backdoor Roth maneuver each year after we have maxed out our 401(k)/403(b), 457, etc. I do it for myself as well as my spouse ($11,000,00/year). My question is: If the backdoor Roth is a good idea, how is the Mega backdoor Roth not a good idea?
I dont think POF was saying it was a bad idea to convert to Roth, I think he was saying it was a bad idea to convert that much when he was in the top bracket. As he steps back now at 0.5 FTE he could convert money in 2018 and pay less tax. If he did it once he fully retires he could convert for even less.
No doubt that’s true. We can always look at these decisions in hindsight and make the calculations and see if it was the right decision. Also, if there is knowledge at the time of pulling back to .5 FTE, of course that changes the calculation. Or if we know, for instance, that tax laws are changing in 2018, that might change our decision. However, at the time we make the decision, is it still a good idea to make the backdoor Roth contribution (even if we are in the top bracket and anticipate continuing to be in the top bracket and have no knowledge of future tax changes or pulling back to part time). Let’s assume we are 45 years old. My main point is: If it is a good idea to make the backdoor Roth contribution, is it not also a good idea to make the Mega backdoor Roth contribution? It seems to me that the logic should remain the same, regardless of the amount of the conversion.
As I say below (sorry, meant it to be a reply to DCD’s first comment) the mega conversion pushes one into higher tax brackets and therefore is likely to end up being an error if the alternative is backdoor plus nonmega; relatively spread out over the years; conversion. Paying your current (top) tax rate on $11K/year is not much, paying two or three brackets up on $400K is a bad idea compared to paying one or none bracket up on $100K over 5 years (PROBABLY- and what if one’s Trad IRA grows from $400K to $1M over the 5 years… well in that case I might wait for a market correction with all the risks of ‘hoping’ for that).
Jenn, your point is well taken. My point assumed already being in the highest bracket. It definitely changes the calculus if it pushes you to a higher bracket.
WCI- I do understand higher bracket is only for amount above bracket. But that + 5% tax on the extra conversion amount above the 28% or 35% rate might be saved by splitting it up over a few years, especially right before retirement when paying 33% this year on the extra instead of 25% on a similar amount the next year. Not applicable though if there is no likelihood of lower than max top tax rate for several years.
DCD- us retired FPs forget there are working neuro and plastic surgeons reading these articles also.
Agreed. Just making sure you knew that.
It’s not clear to me from your comments whether you understand that when you get “pushed into a higher bracket” that the higher bracket only applies to the income in that bracket, not all of your income.
Yes, I understand how our taxes are applied.
From POF’s comments about the website traffic, it’s not obvious that he’s earning any less money doing 0.5 FTE. Maybe he’ll always make more money over the next twenty years than the preceding five and will conclude the Roth conversion was an excellent choice while flying his private jet car to the Cayman Islands.
I hope so, but I don’t think that is his plan.
You understand there is a difference between Backdoor Roth IRAs (and Mega Backdoor Roth IRAs) and Roth conversions, right? The first has no tax bill, the second you have to pay taxes on. So the Backdoor Roth IRA and Mega Backdoor Roth IRA are pretty much always good ideas. A Roth conversion (or a mega roth conversion if it is particularly big and POF is describing it) is not always a good idea. In fact, this is a good example of a time when it may not have been a good idea in retrospect.
I do understand the difference. I should have been more specific. If it doesn’t push you to another bracket, and you are in the top bracket, why would you not do the Roth conversion? Let’s assume at retirement, your bracket remains the same (your top bracket).
If you won’t be in the top bracket in retirement it wouldn’t be worth doing a conversion now at the top bracket. You might also not have the taxable money to pay the tax bill. But otherwise, sure, if you’ll be in the top bracket forever Roth makes a lot of sense, unless the brackets drop, but I don’t think I’d bet on that.
To WCI: sorry, I’m getting confused: can you please explain the difference between these two (Backdoor Roth IRA vs Roth conversion), besides the different taxation regimen? Based on your tutorial on the blog (https://www.whitecoatinvestor.com/backdoor-roth-ira-tutorial/), it seems like the Backdoor Roth IRA is a 2-step process, and its second step is a Roth conversion.
Is the difference that: (a) a Backdoor Roth IRA is a process to allow high earners to do indirect contributions (eg, from a taxable account) to a Roth IRA, and (b) a Roth conversion is a process to convert/transfer pre-taxmoney from a traditional IRA into a Roth IRA (post-tax $$)?
Clarification appreciated, many thanks in advance.
You sound to me like you understand it well. Most Roth conversions are taxable. You’re just moving pre-tax money into a Roth IRA or 401(k). A Roth conversion is one of the steps in the Backdoor Roth IRA process.
The general principal is to convert during low bracket years, and do NOT convert during high bracket years. He could not have predicted lower brackets to come via TCJA reform.
Breaking into a higher tax bracket because of a mega Roth conversion is the reason it is not so good an idea as the back door Roth.
We’re in early retirement and I was avoiding much in Roth conversions expecting continued Level income. However when I considered social security (looks like it’ll still be there for us in some amount 10-15 years down the road) and on top of that RMD from our traditional IRAs I saw my tax bracket will rise a lot or a little at 70 if not earlier with SS. So I’m doing bigger Roth conversions now with the goal of decreasing RMD income later on and not being in a much higher bracket when we start SS and RMDs. A gamble given that who knows if tax structure will remain the same or even if tax treat ment of Roths will remain the same so I mostly max out my 25% bracket. As we get closer and our Traditional IRAs grow we might go on and max out the next bracket.
Forget being in a low tax bracket at 70. Taxarmeggedon. Social Security, plus RMD, plus dividends , plus muni bond interest all add up. I can’t do backdoor roths because of a sizable SEP/IRA. I am making small Roth conversions now knowing I will not convert the entire IRA. I figure that some is better than none. The math is tricky here.
Muni bond interest is at least federal tax free and sometimes state tax free too.
Taxarmageddon isn’t a bad thing to have, but if you can do something earlier that gives you more money after-tax, why not?
I did a “Mega-Roth” conversion at the earliest opportunity. The success level depends on your assumptions and what the future holds (who really knows?). Based on my best guesses that I plugged into a Fidelity calculator, it saved me 250K over the course of my lifetime. No regrets here.
That was my point. I did it as well. It’s all based on assumptions. I prefer the advantages of the Roth.
wealthydoc,
I seriously doubt the validity of any calculator that would suggest you can save $250k by moving your money to a Roth. In the first place most people will experience a lower tax rate in retirement, which does NOT favor a Roth account.
If fact the only way you even save a penney is to prove you will spend the money that you converted to a Roth in a higher tax bracket than at the time of the conversion. Nothing else really matters.
The absolute worst thing you can do is convert all your money into a Roth as this will seal your future withdrawal rate at a much lower level, thus giving you much less after tax money in retirement, because too much of it was spent on taxes up front.
I have written extensively on this subject:
https://seekingalpha.com/article/4140837-risk-roth-ira-revolution
Dave
Could have, would have, should have, maybe, and if only…all belong in the same waste bucket. Still, it’s instructive to look at prior decisions in order to reduce future mistakes.
Example: I took a $10,000 loan against my 457 to pay off my daughters college tuition as her 529 ran short. This money was then NOT in the market for the second half of 2017 (10% run up and 5% more in January). Not smart. But…much of this gain has evaporated in the last two weeks. Now I pay back the loan balance ($8000) and buy in after the >10% correction. Mitigates some of the lost opportunity in the market.
I have no Roth money. Since I already bought my retirement cabin and land and am paying on it, I don’t have enough money left over after maxing out my employers 401K and my own Solo 401 (prior to 2018, this was a SEP). It has gone up in value (on paper) by $55,000 (15%).
I’m going to downsize in 2019. My expenses will drop by over half, I will take half a year off (except for locums) and be in a much lower bracket (could be in the new 12% bracket for a year). As you said, that might be a year to “create some Roth money”.
The government changes the rules all the time. They have ripped me off before (raising my retirement age to 67.5, and failing to pay my medical student loan interest during residency as they agreed to in 1986). You can’t count on the government to do anything they promise, including not taxing the Roth money of “rich people”.
I expect them to reduce my social security benefits or tax more of it based on my retirement income, raise the early retirement age, charge me more for Medicare for doing well, and reduce the money I can protect from current taxes with 401K/SEP (they considered this recently).
I hope it works out for you in the long run on the Roth deal. WCI and other folks on this blog talk about Roth money being highly desirable.
Whoa…you’ve been a physician since sometime around 1986, had to take a loan from a retirement account to pay for your kids college, have no Roth money, and have managed to save 360k in your 401ks after a 30ish year career in medicine (55k growth = 15% per your post)…and you’re questioning the wisdom of people putting money into Roth accounts? Sure, the government will always do stupid stuff and abscond with our money, but it’s a near certainty that the very real missed benefit of not taking advantage of available Roth space now is far greater than the hypothetical, low probability risk of the game being changed and Roth money being taxed again.
Your money, your life — but not taking advantage of at least a backdoor Roth, which effectively costs you nothing to shelter already post-tax dollars because the gubbmit could potentially change the rules doesn’t seem like a great plan.
Well, that is not exactly correct…I’m not sure I was clear. I was in medical school from 1986 to 1990 borrowing $67,000 that went up to $83,000 during residency (as the government did not honor their agreement to pay the interest during residency as the notes stipulated). I started making a “doctor’s wage” in 1994 and have maxed out my retirement accounts since.
My retirement accounts have about a million in them (not $350K). That is not bad for a 54 year old psychiatrist who had to pay back $83,000 in student loans. The 15% I refer to above is the appreciation in value of the retirement home I already purchased. I bought it in 2016 and IT has gone up in value by 15% (not my retirement accounts).
I have four children and did not save enough for their college educations. That is true. I would say it was on of my top few financial mistakes. The eldest daughter’s four year degree cost $92,000. I saved only $30,000, paid another $25,000 from moonlighting weekends, and borrowed $10,000 from my 457 for the last semester. The rest she got via scholarships. My $10,000 loan will have been out of the market for a 10% run up…and a 10% correction. Lucked out on that move. Essentially, it cost me nothing.
As far as the Roth thing, and having “no Roth money” I explained that, and WCI has posted here that someone like me with a current marginal rate of 37% might NOT want to convert any monies at present and might NOT do better than current tax deferral, unless you have “extra” money after all tax deferral accounts. Any extra money I get goes on the mortgage of the land adjacent to my cabin as it is at 6.25% (a guaranteed 6.25% return).
And, yes, I am questioning the wisdom of Roth conversions for many, which was the point of this blog post. In other words the author questioned it…
Yes, in the 37% bracket now and with only a $1M retirement account, I think doing a Roth 401(k) or a Roth conversion this year would be a colossal mistake for you.
I misspoke a bit. My marginal rate for 2018 is 35% ($400K to $600K), not 37% (the highest bracket starts at $600K).
“Only a million”…now I feel bad.
I will endeavor to persevere.
It’s not that a million isn’t a lot of money. It’s just that with “only a million” in a tax-deferred account, you can get it out at a very low tax rate. The RMD at 70 on a million bucks is only $36K. Even with another $30K in SS payments, your marginal tax rate is only 12% and your effective tax rate is even lower. For MFJ- your first $24K comes out tax free and the rest is at 12%. So if you’re saving at 35% now and pulling it out at 12% later, that tax-deferred account is providing you a huge boost in return on investment.
Yes. Well I have picked my own strategy and it is to live on a lot less than I do now. I’m not going to keep plugging away at this pace. As I said, I’m making up, to the degree I can, for years of inadequate saving. I only found this blog and book about 18 months ago?
Minimize current taxes, max out the tax deferred accounts,
downsize and go to half time at 56. Retire at 59.5, possibly working a Locum or two per year to augment my savings.
Thanks for your answers as always. I find them helpful.
Making annual conversions to Roth will build a so-called Roth ladder, which is a great strategy to access your 401(k) / IRA money prior to age 59.5. Converted assets can be accessed tax and penalty-free after it has seasoned in the Roth for 5 years (technically January 1st of the year 5 years later, so closer to four years if you convert late in the year).
Question : am I correct that you cannot access only the “converted”assets, in other words if you have $100K in a Roth and you convert $10K, it’s only that $10K that cannot be accessed until it has seasoned?
… I would advise my younger self not to make those conversions. “There will be time to make those conversions in a lower tax bracket when you’re early retired”, I would say. While in hindsight, it was probably not the best decision, I did the best I could with the information I had at the time….
Consider also that early retirement is not the only situation where you may be in a lower tax bracket; for example consider someone maybe with not a very large income but still in a marginal tax bracket of 25% who one day becomes disabled or partially disabled who can still generate some income (with or without disability insurance, unsure if disability benefits are even taxed?), who contributed to a tax-deferred account while in the 25% bracket but unfortunately finds himself in a lower bracket from a disability but still with the wherewithal to convert some into the Roth but is now in a lower bracket; same principle of “converting in a lower tax bracket” but for a different reason.
Is it true that another situation where Roth conversions don’t look so good is if you retire to a tax-free state but make the conversions during your working career in a taxable especially a high-tax state, your state income tax goes up during the conversion while you are converting but it was unnecessary because you retired in a tax-free state?
Another thing to consider is that I can imagine it just so happened that a person was only able to have a tax-deferred account for the first many years of his career, and only later can he also do a Roth; the more options you have in retirement the better, so if that person with a tiny Roth mid-career suddenly is able to get his Roth to a much higher value by way of conversions, then he will have a more options because he has a larger Roth rather than a very small Roth.
Similarly along the lines of more options, for you who may in hindsight “may have not made the best decision by converting”, something really catastrophic happens and you are oh so glad that you can tap your now-much-larger Roth tax-free.
I’m surprised this Roth conversion thing used by “rich folks” has not been erased by tax law.
“Economists have warned about exploding future revenue losses associated with Roth IRAs. With these accounts, the government is “bringing in more now, but giving up much more in the future,” said economist and Forbes contributor Leonard Burman. In a study for The Tax Policy Center, Burman calculated that from 2014 to 2046, the Treasury would lose a total of $14 billion as a result of IRA-related provisions in the 2006 tax law. The losses stem from both Roth conversions and the ability to make nondeductible IRA contributions and then immediately convert them to Roths.”
That quote suggests Roth money is a good thing… so the government should get rid of it’s availability…
“If the Roth IRA owner expects that the tax rate applicable to withdrawals from a traditional IRA in retirement will be higher than the tax rate applicable to the funds earned to make the Roth IRA contributions before retirement, then there may be a tax advantage to making contributions to a Roth IRA over a traditional IRA or similar vehicle while working. There is no current tax deduction, but money going into the Roth IRA is taxed at the taxpayer’s current marginal tax rate, and will not be taxed at the expected higher future effective tax rate when it comes out of the Roth IRA. There is always risk, however, that retirement savings will be less than anticipated, which would produce a lower tax rate for distributions in retirement. Assuming substantially equivalent tax rates, this is largely a question of age. For example, at the age of 20, one is likely to be in a low tax bracket, and if one is already saving for retirement at that age, the income in retirement is quite likely to qualify for a higher rate, but at the age of 55, one may be in peak earning years and likely to be taxed at a higher tax rate, so retirement income would tend to be lower than income at this age and therefore taxed at a lower rate.”
Closing the loophole was explicitly rejected by Congress in the most recent round of tax law updates. Of course, no one’s “home is safe while Congress is in session” and all that, so the law could change at a later date. All the more reason to do the conversions now if they make sense otherwise.
If you are in the highest tax bracket should you fund the roth part of the 401k solo and group or should you put all the money in tax deferred.
It’s hard to give up putting what you can in the roth 401k but at the same time you are paying taxes on that money that you would not have to if you just put in in the non roth portion of the 401k .
In general it varies, but most would be better off putting it all in tax-deferred. See graph showing tax-deferred ends up much more profitable vs Roth even over decades:
Argh, link got scrubbed. Google “mad fientist how to access retirement funds early”.
As to the question above, doesn’t it depend on age and time away from retirement?
My marginal tax rate is 37%. Making Roth contributions now just doesn’t add up. As I’ve said before, I need lower taxes NOW, not in a few years. For me, that’s because I am dropping my income markedly from “work extra now to retire early” to “working half-time for a year or two” on purpose.
Doesn’t the current Roth law incentivize an older worker in their peak earning years to markedly drop their income for a year (or two) to do a large Roth conversion at the beginning of retirement?
Imagine going from making $400,000 to $500,000 down to $100,000 by taking a few months off, then dropping to half-time on the other side of the break. There would be an opportunity to recharacterize my entire old SEP (worth $100,000 now) to Roth money while in a much lower bracket. Then in early semi-retirement putting all of the 1099 monies into the Roth portion of a Solo 401.
How and when to create Roth money is VERY individual.
Tax law always incentivizes us to work and earn less.
Yes. Look at the tax jump for your first dollar above $315,000…it goes up 8% to a nasty 32%. Not a bad problem to have I guess, but the jump up is almost enough to make one shoot for $314,000 or less. Looking them over, looks like under $77,400 is pretty good too (12%). That’s it, I’m quitting psychiatry and becoming a bartender. Just kidding.
Of course, these are just federal tax rates. Michigan wants their 4.25%, and social security and a Medicare want theirs. (6.25% up to $127,200 for social security and 1.45% for Medicare, unless you make > $200,000, then add in another Medicare surtax of 0.9%).
That’s quite a chunk for all those “rich folks “ on dollar number $315,001: 32%+4.25%+2.35% = 38.6%. Add another 5% for dollar number $400,001 (43.6%).
I think, for me, they are so onerous, I’m actually looking to make less. That’s sort of sad.
That’s your choice. I actually wouldn’t mind paying $1M in taxes. 🙂 That would be a gross income of $3M+. I could do a lot of good with that, even after tax.
If you’re working more than you would like to if you didn’t need the money, I would suggest either cutting back on work or figuring out a way to get to the point where you don’t need the money ASAP.
I am doing just that. I am working 26 weekends and 3 holidays a year on the side socking away as much as I can. Hence a psychiatrist in the 37% bracket. That is an anomaly. My current wages are exactly double the average wage for psychiatry listed in the most recent physician compensation survey.
https://www.medscape.com/slideshow/compensation-2017-overview-6008547#1
Can’t wait to retire. As I have said before, I am downsizing my home. I’m in the final stretch to my dream: a mountain cabin (bought it), smaller expenses (they drop in half), and being able to work half time (for a year or two) until I’m 59.5 and then quitting.
All I have to do is work another year full time, sell my house, pay off the cabin and move.
I enjoyed the article because I did exactly the same thing with the same thought process, although only about half as much so my tax cost was lower. But I’m not feeling the same angst about potential lower net worth because making the conversion allows the backdoor Roth IRA to be done without being subject to the pro-rata rule. No one knows for how long the government will continue to allow this, but as long as they do do I want to be able to take advantage of this. If I wait until my tax bracket is significantly lower, I would have lost the opportunity to increase the Roth portion of my portfolio. I view the Roth accounts as a way to increase your financial flexibilty in how you use assets in retirement, especially for those whose pre-tax retirement savings lead to RMDs that may exceed what they actually need.
Great point. That was part of my reasoning as well.
That’s true, although there are other ways to “hide” SEP-IRA money. After doing this, I had a couple more years of 1099 income, and I started contributing to a SEP-IRA again in 2011 and 2012. When I became employed, I rolled the SEP over into my employer’s 401(k) and started making backdoor Roth contributions.
Best,
-PoF
Thank you for all the great feedback.
With the information I had back in 2010, I think I made a sound decision, even if I wouldn’t make it again knowing everything I know now.
In hindsight, if I had known that I would be strongly considering a very early retirement by 2020, and that there would be a 24% tax bracket that goes up to an income of $315,000 starting in 2018, I would have held off and eventually have rolled that money over into a 401(k).
If I had not converted to Roth back in 2010, I’d be a 401(k) mllionaire today (or very close to it), and would be focusing on a more aggressive Roth conversion strategy when my earned income stops. WIth that expanded 24% bracket, it wouldn’t be difficult to convert all of that money over 5 to 10 years.
That being said, the past cannot be changed and I’m not sad about having more than twice as much of our retirement in Roth space as I do tax-deferred.
Thanks again!
-PoF
Apologies if I’m not understanding the math but the discussion on whether to do a Roth now (while in a higher tax bracket) or later (when in a lower tax bracket) has me confused. Doesn’t one have to factor in the tax free aspect of earnings inside the Roth when doing the calculation? For example, assume I convert $10K into a Roth at a combined tax rate of 40%. If after 10 years my balance doubles, isn’t that an effective rate of 20% (due to earnings being tax free)? If after 20 years my balance doubles again, isn’t that an effective rate of 10%?
Since the entirety of a traditional IRA (original contribution plus earnings) is taxed upon withdrawal, wouldn’t my later tax bracket have to be significantly lower to offset the tax free benefits on earnings that occurs if I convert to a Roth immediately? Using my example above, my tax rate in 20 years would need to be less than 10% to beat a conversion today, even at a 40% rate. Seems to me, if you’ve got time in your favor and so long as you don’t expect a retirement tax rate of less than 10%, it would almost always be better to do the Roth conversion now. Please be kind if I’m completely off base.
No. That’s not the way the math works. It isn’t about the total amount of taxes paid, but about the rate at which the taxes are paid.
For example, let’s say your tax rate is 40% and you have $100K now, $200K in a decade, and $400K in two decades. All that money is in a tax-deferred account. If you convert it now, it becomes $60K, which then becomes $120K and $240K. If you convert it in a decade, the $200K becomes $120K and then $240K after another decade. If you convert it in two decades, it costs you $160K to pay for the conversion and, guess what, you’re left with $240K. Same, same, same as long as your tax rate never changes. So your decisions of when to convert come down to trying to arbitrage tax rates- i.e. pay for a conversion at a lower rate than you saved when you put the money in a tax-deferred account to start with.
Unfortunately, life isn’t quite that simple. This is because the money used to pay the taxes on the conversion has to come from somewhere. And if that somewhere is a taxable account, the sooner you convert the better because a conversion essentially allows you to move taxable money and tax-deferred money into a tax-free account, where it now grows without tax drag. So if your tax rates truly will remain the same, and you’re maxing out your accounts, you’re better off converting sooner than later.
Hpe that helps.
Most helpful. Your second paragraph would suggest that it’s always?/usually? better to pay the tax due on a conversion out of a taxable account, rather than from the IRA balance, so that the entire/higher amount receives the favorable tax treatment.
That’s correct.
WCI,
I agree with you that if you have a taxable account, then the sooner you get it into a Roth the better, EXCEPT for the most normal case where tax rates are lower in retirement and it doesn’t make sense to do the conversion at all.
I realize that at your income level you are probably going to have a taxable account at some point in time, but using the taxable account to pay the taxes on a conversion does NOT make the Roth funds any more valuable – it just creates more of them, and reduces your taxable account. The issue is that even the 2% tax differential created by the recent tax cuts is enough to give an advantage to the IRA + taxable account over the Roth conversion of the money, due to the fact that a taxable account is already about half of a Roth account to begin with, given the fact you paid the taxes on the front end and on the back end you are only paying about half your tax rate on the long term capital gains. See my recent article on Seeking Alpha linked to above called the “Risk of the Roth IRA revolution.”
The tax code changes could have gone the other way as well. But I agree with your general premise- that deferring taxes during your peak earnings years is usually the right move.
PoF,
Converting all of your IRA money to Roth is the absolute worst thing you can do at any point in your life. What that leaves you with, if you don’t have a pension is a tax rate of ZERO with much less spendable after-tax income and not even being able to fill up your standard deduction bucket. In my volunteer tax work I have seen this many times with retirees. They leave thousands of tax-free dollars on the table that could have been taken from an IRA tax-free, if they had one.
Good point, Dave. It would be a waste to have no taxable income to use up the standard deduction and very low bracket(s). No sense in converting in the 24% bracket to later have zero taxable income.
In my situation, I expect to draw down a 457(b) for ten years or so, will have dividends (currently about $30,000 a year from low-dividend passive index funds), and can always do some capital gain harvesting to fill up the 0% capital gains bracket.
Cheers!
-PoF
WCI,
Finally in answer to your original question, while the Mega Roth conversion was “expensive” in hindsight and there was probably a better way to do it, it is not all bad.
We cannot predict what the future tax rates will be, though in reality I can’t see any Senators or Representatives running on a platform to raise taxes. The truth is sometimes in retirement we would like to buy something which would create a large tax burden if we withdrew it from a IRA all in one year. I myself paid off a $180k mortgage over two years using a combination of Taxable, Roth, & IRA funds, in my first two years of retirement.
Dave
I made the big conversion in 2010 of $160,000 and spread the tax over two years. My taxable income in 2011 and 2012 was effectively $80,000 higher than it would have been. I paid the tax out of cash I had on hand, rather than foolishly whittling down the IRA. Mistake? Goodness no! I had a legal Arizona residence and worked in California a few years and filed NR on California income taxes. Then in 2015 became a California resident. I also started a Roth 401k with the new company I worked for since 2013. Now I am 59 and more than 46% of my 401ks and IRAs are Roth accounts. The Roth accounts grew rather fast from 2010 to 2018. No regrets at all. I can retire in coastal California with lower income tax than most states. In three or four years the ratio will be 50/50 and also my municipal bonds are not taxable either. I did not even count them.