Here's a statement of the obvious: The opinions expressed here are those of the participants, not those of the Mutual Fund Observer. We cannot vouch for the accuracy or appropriateness of any of it, though we do encourage civility and good humor.
I'm not quite sure about 'time' but I converted mine last year when Mr. Market took a dump all over everything. Lucky probably but it saved me a ton when those positions quickly re-doubled/tripled in the ensuing months. What I'm saying is that I waited for a correction and used it to my advantage.
The best time to convert to a Roth is when income is low, taxes are anticipated to be higher in the future, and when the value of the asset is low such as a market correction.
Gentlemen, I really appreciate for your inputs. Last spring was a great opportunity to do a large conversion. Unfortunately many of us did not plan for it. Now we are evaluating which how much and which funds to convert per year. Great reminder that the tax table is for the adjusted gross income.
To reduce tax in retirement, we did few small changes. We switched our 401(K) to Roth 401(K) several years ago. Still we have rollover and traditional IRAs to consider for higher future tax and RMD. Higher retirement income incurs additional cost on Medicare premium and taxation on most of the social security benefit.
The question can be split into two parts: 1. Best time during a given year 2. Best year
1. In theory, the best time within a year to convert is when your T-IRA portfolio is at its nadir. Which is great if you can see into the future.
Pragmatically, waiting until there's a market correction (10% dip or more) can leave you converting at a higher value than converting now. If the market rises 15% while you're waiting and then drops 10% (to 103.5% of your starting value), what's the point? Or if you're planning on converting a bit each year, you can easily wind up stuck at the end of the year with a higher market still waiting for that correction.
If you're trying to keep MAGI under a certain "cliff" threshold (e.g. for ACA subsidies or to avoid IRMAA) then you should likely be conservative and top off your conversion near year end once you have a better handle on MAGI YTD. If your concern is about edging into another tax bracket, it's not a big deal if you wind up a few dollars over or under, so don't wait until year end for that reason.
2. If you're converting and paying taxes with non-IRA money, sooner can be better than later even if your expected tax rates in the future will be a little higher. That's because you're effectively adding money to your IRA by prepaying your taxes. (A dollar in a Roth is worth more than a dollar in a T-IRA.)
OTOH, many states give tax breaks for retirement income (including Roth conversions) once one reaches a certain age. So it can be advantageous to wait until then. You might convert between the time you reach that age and the time you retire and actually start drawing money from your T-IRA and/or pension (and possibly using up that tax break).
Several states give some sort of help on Medicare Part D prescriptions, where your income needs to be low enough to qualify. Doing Roth conversions earlier (even at somewhat higher tax rates) may help qualify for these or other income-related programs.
The largest such program is EPIC in NYS. Its income cutoff is decidedly middle class: $75K single, $100K married.
The rule of thumb short answer is the sooner the better (both within year and across years); just don't do too much in a single year and keep various income thresholds in mind. And if you're thinking of making charitable contributions, keep some money in the T-IRA so you have it available for QCDs.
Actually, any time is pretty good. I did 3 - all between the ages of 63 and 70.
But (from my inner speculative nature), the more depressed the asset at the time, the better the eventual outcome. There’s a big “IF” here. This only works if you are correct in your assessment that a given asset (fund, stock, etc.) will turn around within your expectant time horizon and you have the patience to allow that to occur. The tax-free winnings are hard to beat - if your gambit works.
It’s easier in this environment, I think, to consider what not to convert. I wouldn’t do anything related to bonds or other fixed income. If my own time horizon were just a bit longer, I’d look at EM or international funds. And, I’d probably try to time the evolving disaster in AARK-type funds at a future point. But my general reading says it’s still too early to buy those. (If they keep falling at this rate, however, “It shan’t be long”.)
If you do succeed in grabbing off a rising star with a Roth conversion, consider scaling out a bit at some future point to partially lock in the gains.
Edit / Add:
@Sven I thought it would be helpful to add that you can be selective as to which fund(s) you convert. Once you’ve converted a fund (or other asset) your fiduciary (ie Fido) sets this up as a separate account. With my Fido accounts, both the Roth and Traditional IRA appear on one easy to access page. One reason for the separate accounts is to allow you to move money around (buy and sell) inside that account. Obviously, transfers back and forth between a Roth and Traditional IRA would not be allowed.
Does it matters what you convert, if you can always buy or sell anything in either account tax free? Once you have the capital in a Roth, I agree it should be probably devoted to more risky, long term investments.
Converting before you are on Medicare and after are two different calculations.
It is important to remember that the IRMAA surcharges go from zero dollars ( B Medicare Premium $1776 /year in 2021) to $700 for ONE DOLLAR of income ( AGI before the standard deduction) above $176,000 for a couple and $111,000 single. A couple would therefore pay $1400 extra.
IRMAA is based on two years before 1040, so 2022 will look at 2020 AGI. Once you file 2021 taxes in 2022 you can ask for redo, if your 2021 income is lower. It takes a month or so, so you will end up paying more for a few months.
One strategy I am considering is to do a large conversion, all in one year, before I take SS at 70. Then the impact of the conversion on the IRMAA will be limited to one year.
Other things to consider. IRS lets you put up to $135,000 in a QLAC which will reduce you RMD proportionally.
Lawrence Kotlikoff from BU has a great article in Barrons
He also runs a web site with a neat financial planning program for $100, Maxifi. It may be overkill for a lot of folks, but it will allow you to run all sorts of projections and scenarios about Roth conversions pretty easily
I feel that longevity insurance is one of the few useful products that the financial industry has created in the past several years. That said, if it's not a product that you are interested in independent of tax considerations, then buying a QLAC is a poor way to reduce RMDs.
Getting back to timing of conversions ... Though recharacterizations (undo's) of conversions are no longer allowed, there are a couple of other tactics that provide some or much of the same effect.
One is to use a recharacterization that is still permitted. A contribution as opposed to a conversion can still be recharacterized. So if you contributed $6000 to a Roth IRA in 2021 but in doing your taxes you discover that you'd have been better off taking the deduction, you can recharacterize the $6K as a contribution to a traditional IRA. (You can also do the opposite: contribute to a T-IRA and then recharacterize it to a Roth.) https://www.irs.gov/retirement-plans/retirement-plans-faqs-regarding-iras#Recharacterization of IRA Contributions
Another it to take advantage of the 60 day rollover rule (sometimes limited to one per year). One could withdraw money from a T-IRA near the end of the year, and then early the next year when the tax situation is clearer (and within 60 days of the withdrawal) either put the money back into the T-IRA (60 day rollover), put the money into a Roth (indirect rollover conversion), or split the money between a T-IRA and a Roth.
If any money goes back into a T-IRA, you can't do this again for another 365 calendar days. The restriction is according to days, not fiscal years.
“Does it matters what you convert, if you can always buy or sell anything in either account tax free?”
That’s correct @sma3. I suspect it may perhaps matter from a purely psychological standpoint as one contemplates doing a Roth conversion. I was driven to do my first by the depths of the ‘07-‘09 market rout. After I ran out of tax-deferred cash to keep “buying down” inside my IRA in early ‘09 the Roth conversion occurred to me as the next logical step - another way to “wager” on the longer term value of heavily depressed assets.
But if, unlike me, you can walk and chew gum at the same time it doesn’t matter. You can always move in and out of opportunistic investments.
Thank you everyone for your contribution to this topic.
@msf, great stuff to consider as I am working on the $ amount we want to convert. Most likely it will be done over several years before RMD begins while coordinating with my spouse’s RMD which starts several years later.
@hank, we did exactly what you suggested several years ago when we consolidated IRAs in the same brokerage. We have laid our a plan on what goes in the Roth versus Traditional IRA.
@sma3, I have been reviewing the details on income level and their impact on Medicare premium. Great suggestion on Maxifi for evaluate various scenarios - will check it out.
What's up ? From above post, '' It is important to remember that the IRMAA surcharges go from zero dollars ( B Medicare Premium $1776 /year in 2021) to $700 for ONE DOLLAR of income ( AGI before the standard deduction) above $176,000 for a couple and $111,000 single. A couple would therefore pay $1400 extra." $176,000 for a couple = $88,000 for a single MWOT OR $111000 for a single = $222.000 for a couple MWOT
Most Medicare Part B participants pay a premium that covers 25% of the actual cost of the coverage. (The government picks up the rest, using general tax revenue.) In 2021, that 25% comes to $148.50/mo, or $1782/year.
If one's MAGI¹ (AGI from line 7 of form 1040 plus tax-exempt interest from line 2a) is above $88K (single or married filing separately) or $176K (married filing jointly), then one's share of the actual Part B cost increases.
If one's MAGI is just over this threshold, then instead of paying 25% of the cost of Part B, one will pay 35%. That's a 40% increase (35%/25% = 1.4x). So in 2021, the first level of IRMAA comes to 40% x $148.50 = $59.40/mo, or $712.80/year. https://www.medicare.gov/your-medicare-costs/part-b-costs
There are also higher brackets, where one pays 50% of the actual part B cost (IRMAA = 100% of Part B premium), 65% of Part B cost (IRMAA = 160% of premium), 80% of part B cost (IRMAA = 220% of premium), or 85% of part B cost (240% of premium). No matter what one's income level, the government is still picking up some of the tab. https://secure.ssa.gov/poms.nsf/lnx/0601101031
In 2022, the first threshold is at $91K ($182 married filing jointly), and IRMAA is $68/mo. All the thresholds for 2021 and 2022 along with the total (part B premium plus IRMAA) amounts are on the Medicare.gov site given above.
¹MAGI is calculated using the last income tax return filed. For 2022 IRMAA, that is the income tax form for 2020, filed as recently as just two months ago, including extensions. Your Jan 2022 IRMAA can't be based on 2021 income since you haven't reported it yet.
While it is old (2015) and uses the RMD date of 70 1/2 and an anticipated return of 8% yearly, it is still probably correct that you have to live a long while past 85 to come out ahead.
I am not sure MaxiFI would model this for you, but it might. The calculations are not difficult. What you want to know other than how long you will live is the expected return ( or lack thereof) on your IRA assets required to beat the QLAC. I doubt tax rates will go down any.
@ sma3 "Does it matters what you convert, if you can always buy or sell anything in either account tax free?" I've run into a reason to move shares in a conversion: a closed fund. I own FDGRX Fidelity Growth in my IRA. It's closed except to those who already shares. By moving some shares to my Roth IRA, I'll be able to buy more shares in that Roth account. Same reason holds for moving shares directly to an account which is not tax-sheltered in any way. David
Comments
I covered a Roth Conversion in:
https://seekingalpha.com/article/4454723-best-tax-efficient-funds
To reduce tax in retirement, we did few small changes. We switched our 401(K) to Roth 401(K) several years ago. Still we have rollover and traditional IRAs to consider for higher future tax and RMD. Higher retirement income incurs additional cost on Medicare premium and taxation on most of the social security benefit.
1. Best time during a given year
2. Best year
1. In theory, the best time within a year to convert is when your T-IRA portfolio is at its nadir. Which is great if you can see into the future.
Pragmatically, waiting until there's a market correction (10% dip or more) can leave you converting at a higher value than converting now. If the market rises 15% while you're waiting and then drops 10% (to 103.5% of your starting value), what's the point? Or if you're planning on converting a bit each year, you can easily wind up stuck at the end of the year with a higher market still waiting for that correction.
If you're trying to keep MAGI under a certain "cliff" threshold (e.g. for ACA subsidies or to avoid IRMAA) then you should likely be conservative and top off your conversion near year end once you have a better handle on MAGI YTD. If your concern is about edging into another tax bracket, it's not a big deal if you wind up a few dollars over or under, so don't wait until year end for that reason.
2. If you're converting and paying taxes with non-IRA money, sooner can be better than later even if your expected tax rates in the future will be a little higher. That's because you're effectively adding money to your IRA by prepaying your taxes. (A dollar in a Roth is worth more than a dollar in a T-IRA.)
OTOH, many states give tax breaks for retirement income (including Roth conversions) once one reaches a certain age. So it can be advantageous to wait until then. You might convert between the time you reach that age and the time you retire and actually start drawing money from your T-IRA and/or pension (and possibly using up that tax break).
Several states give some sort of help on Medicare Part D prescriptions, where your income needs to be low enough to qualify. Doing Roth conversions earlier (even at somewhat higher tax rates) may help qualify for these or other income-related programs.
The largest such program is EPIC in NYS. Its income cutoff is decidedly middle class: $75K single, $100K married.
The rule of thumb short answer is the sooner the better (both within year and across years); just don't do too much in a single year and keep various income thresholds in mind. And if you're thinking of making charitable contributions, keep some money in the T-IRA so you have it available for QCDs.
But (from my inner speculative nature), the more depressed the asset at the time, the better the eventual outcome. There’s a big “IF” here. This only works if you are correct in your assessment that a given asset (fund, stock, etc.) will turn around within your expectant time horizon and you have the patience to allow that to occur. The tax-free winnings are hard to beat - if your gambit works.
It’s easier in this environment, I think, to consider what not to convert. I wouldn’t do anything related to bonds or other fixed income. If my own time horizon were just a bit longer, I’d look at EM or international funds. And, I’d probably try to time the evolving disaster in AARK-type funds at a future point. But my general reading says it’s still too early to buy those. (If they keep falling at this rate, however, “It shan’t be long”.)
If you do succeed in grabbing off a rising star with a Roth conversion, consider scaling out a bit at some future point to partially lock in the gains.
Edit / Add:
@Sven I thought it would be helpful to add that you can be selective as to which fund(s) you convert. Once you’ve converted a fund (or other asset) your fiduciary (ie Fido) sets this up as a separate account. With my Fido accounts, both the Roth and Traditional IRA appear on one easy to access page. One reason for the separate accounts is to allow you to move money around (buy and sell) inside that account. Obviously, transfers back and forth between a Roth and Traditional IRA would not be allowed.
Does it matters what you convert, if you can always buy or sell anything in either account tax free?
Once you have the capital in a Roth, I agree it should be probably devoted to more risky, long term investments.
Converting before you are on Medicare and after are two different calculations.
It is important to remember that the IRMAA surcharges go from zero dollars ( B Medicare Premium $1776 /year in 2021) to $700 for ONE DOLLAR of income ( AGI before the standard deduction) above $176,000 for a couple and $111,000 single. A couple would therefore pay $1400 extra.
IRMAA is based on two years before 1040, so 2022 will look at 2020 AGI. Once you file 2021 taxes in 2022 you can ask for redo, if your 2021 income is lower. It takes a month or so, so you will end up paying more for a few months.
One strategy I am considering is to do a large conversion, all in one year, before I take SS at 70. Then the impact of the conversion on the IRMAA will be limited to one year.
Other things to consider. IRS lets you put up to $135,000 in a QLAC which will reduce you RMD proportionally.
Lawrence Kotlikoff from BU has a great article in Barrons
https://www.barrons.com/articles/most-retirement-planning-is-wrong-laurence-kotlikoff-51631207476
He also runs a web site with a neat financial planning program for $100, Maxifi. It may be overkill for a lot of folks, but it will allow you to run all sorts of projections and scenarios about Roth conversions pretty easily
Kitces, Why A QLAC In An IRA Is A Terrible Way To Defer The Required Minimum Distribution (RMD) Obligation
https://www.kitces.com/blog/why-a-qlac-in-an-ira-is-a-terrible-way-to-defer-the-required-minimum-distribution-rmd-obligation/
Getting back to timing of conversions ... Though recharacterizations (undo's) of conversions are no longer allowed, there are a couple of other tactics that provide some or much of the same effect.
One is to use a recharacterization that is still permitted. A contribution as opposed to a conversion can still be recharacterized. So if you contributed $6000 to a Roth IRA in 2021 but in doing your taxes you discover that you'd have been better off taking the deduction, you can recharacterize the $6K as a contribution to a traditional IRA. (You can also do the opposite: contribute to a T-IRA and then recharacterize it to a Roth.)
https://www.irs.gov/retirement-plans/retirement-plans-faqs-regarding-iras#Recharacterization of IRA Contributions
Another it to take advantage of the 60 day rollover rule (sometimes limited to one per year). One could withdraw money from a T-IRA near the end of the year, and then early the next year when the tax situation is clearer (and within 60 days of the withdrawal) either put the money back into the T-IRA (60 day rollover), put the money into a Roth (indirect rollover conversion), or split the money between a T-IRA and a Roth.
If any money goes back into a T-IRA, you can't do this again for another 365 calendar days. The restriction is according to days, not fiscal years.
That’s correct @sma3. I suspect it may perhaps matter from a purely psychological standpoint as one contemplates doing a Roth conversion. I was driven to do my first by the depths of the ‘07-‘09 market rout. After I ran out of tax-deferred cash to keep “buying down” inside my IRA in early ‘09 the Roth conversion occurred to me as the next logical step - another way to “wager” on the longer term value of heavily depressed assets.
But if, unlike me, you can walk and chew gum at the same time it doesn’t matter. You can always move in and out of opportunistic investments.
@msf, great stuff to consider as I am working on the $ amount we want to convert. Most likely it will be done over several years before RMD begins while coordinating with my spouse’s RMD which starts several years later.
@hank, we did exactly what you suggested several years ago when we consolidated IRAs in the same brokerage. We have laid our a plan on what goes in the Roth versus Traditional IRA.
@sma3, I have been reviewing the details on income level and their impact on Medicare premium. Great suggestion on Maxifi for evaluate various scenarios - will check it out.
It is important to remember that the IRMAA surcharges go from zero dollars ( B Medicare Premium $1776 /year in 2021) to $700 for ONE DOLLAR of income ( AGI before the standard deduction) above $176,000 for a couple and $111,000 single. A couple would therefore pay $1400 extra."
$176,000 for a couple = $88,000 for a single MWOT
OR
$111000 for a single = $222.000 for a couple MWOT
zero IRMAA under $176,000 $88,000
$59 /mo/person $176,000 to $222,000 $88,000 to $111,000
$148.50/mo/person $222,000 to $276,000 $111,000 to $138,000
There is a great $4 book ( Amazon) with all the data you need
James McGlynn Retirement Planning Tips for Baby Boomers
If one's MAGI¹ (AGI from line 7 of form 1040 plus tax-exempt interest from line 2a) is above $88K (single or married filing separately) or $176K (married filing jointly), then one's share of the actual Part B cost increases.
If one's MAGI is just over this threshold, then instead of paying 25% of the cost of Part B, one will pay 35%. That's a 40% increase (35%/25% = 1.4x). So in 2021, the first level of IRMAA comes to 40% x $148.50 = $59.40/mo, or $712.80/year.
https://www.medicare.gov/your-medicare-costs/part-b-costs
There are also higher brackets, where one pays 50% of the actual part B cost (IRMAA = 100% of Part B premium), 65% of Part B cost (IRMAA = 160% of premium), 80% of part B cost (IRMAA = 220% of premium), or 85% of part B cost (240% of premium). No matter what one's income level, the government is still picking up some of the tab.
https://secure.ssa.gov/poms.nsf/lnx/0601101031
In 2022, the first threshold is at $91K ($182 married filing jointly), and IRMAA is $68/mo. All the thresholds for 2021 and 2022 along with the total (part B premium plus IRMAA) amounts are on the Medicare.gov site given above.
¹MAGI is calculated using the last income tax return filed. For 2022 IRMAA, that is the income tax form for 2020, filed as recently as just two months ago, including extensions. Your Jan 2022 IRMAA can't be based on 2021 income since you haven't reported it yet.
Here is link to 2022 data
https://www.cms.gov/newsroom/fact-sheets/2022-medicare-parts-b-premiums-and-deductibles2022-medicare-part-d-income-related-monthly-adjustment
2022 basic B is $170/mo First step up at same AGI is now $68/mo
@msf
Thanks for the link re QLAC
While it is old (2015) and uses the RMD date of 70 1/2 and an anticipated return of 8% yearly, it is still probably correct that you have to live a long while past 85 to come out ahead.
I am not sure MaxiFI would model this for you, but it might. The calculations are not difficult. What you want to know other than how long you will live is the expected return ( or lack thereof) on your IRA assets required to beat the QLAC. I doubt tax rates will go down any.
"Does it matters what you convert, if you can always buy or sell anything in either account tax free?"
I've run into a reason to move shares in a conversion: a closed fund.
I own FDGRX Fidelity Growth in my IRA. It's closed except to those who already shares. By moving some shares to my Roth IRA, I'll be able to buy more shares in that Roth account.
Same reason holds for moving shares directly to an account which is not tax-sheltered in any way.
David