Roth Conversions

[NW-Bound]

Well it goes without saying that tIRA withdrawals in retirement will be taxed at some TBD average rate. Roth conversions however are taxed at your (perhaps higher) current marginal tax rate.

I would not do Roth conversion when I was still working.

But now, I am retired and have been living off my after-tax accounts, and can pay no tax if that's what I like.

By stuffing the Roth with retirement money, I am paying some tax now rather than more later. When I exhaust my after-tax money and start to draw from IRA/401k, it looks like I will be in 25% bracket at least, so I've got nothing to lose by doing at least the 15% bracket now to lock in that lower rate.

I-ORP tells me to put pedal to the metal to the top of 25% bracket, but the tax bill hurts too much, compared to paying nothing now.

PS. I already missed out on some Roth conversions the last couple of years, and did not go as far as I should. I enjoyed the low-tax status too much to realize it will come back and hurt me more later.

 

[jebmke]

I stop just short of the pesky 30% marginal rate, which doesn't appear in any tables.

 

[kaneohe]

Well it goes without saying that tIRA withdrawals in retirement will be taxed at some TBD average rate. Roth conversions however are taxed at your (perhaps higher) current marginal tax rate.

Even with future tax rate increases, the future average rate just may be less than your current marginal rate.

- Your mileage may vary though

Just to expand on your wise words, I bolded the key ones. If you have a tax-efficient portfolio w/ a low base income from investments/no pension, your TIRA withdrawals could conceivably be taxed at a combination of 0,10,15% brackets (an average rate) . However if you have a tax-inefficient portfolio with lots of taxable interest, active funds w/ lots of ST/LTCGs/, pension, SS,etc., your TIRA withdrawals would be taxed at your retirement marginal tax rate (similar to how your Roth conversions are taxed at your conversion marginal tax rate).

 

[NW-Bound]

... However if you have a tax-inefficient portfolio with lots of taxable interest, active funds w/ lots of ST/LTCGs/, pension, SS,etc., your TIRA withdrawals would be taxed at your retirement marginal tax rate (similar to how your Roth conversions are taxed at your conversion marginal tax rate).

Ever since they tax SS, which can be up to $84K/yr for a couple of high earnings who also delay till 70, it leaves only $11.5K for IRA withdrawal to stay under the 25% bracket (starts at $95.5K). And at 70, RMD also hits, leaving you no choice.

Perhaps Congress will just limit SS for these high-income people, and take away their tax agony. No more SS for them, no more tax problem.

PS. Correction: only a max of 85% of SS is currently taxed. So, the $84K max SS is only taxed to $71.4K, leaving $24.1K for IRA/401k withdrawal.

 

[N02L84ER]

Our retirement is based solely on taxable investments, IRAs, Roths, and hopefully SS in 6 - 14 years. We have been doing Roth conversions since my DH retired 10 years ago. So far we have converted $182K+ (an average of $18K/yr) and have paid $619 in tax or 0.34%. We only convert enough to cancel out our exemptions, deductions, and foreign tax credits. When our guess is wrong about the amount of dividends that are qualified or the size of the foreign tax credit, we just let the conversion ride which is why the tax is greater than $0. We now keep our AGI below $62K because of ACA so I am sure our federal taxes will be $0 until Medicare age. With no pensions or annuities, I doubt our age 70+ taxes will get much above the current 10% bracket.

 

[sengsational]

What PPACA limit do you target (if any)?

I'm trying to figure out my own plan for next year, and I'm thinking 249% of FPL to get both subsidies and cost-sharing.

2016 will be the first year where I get to choose my AGI and it's freaking me out a little.

For us, a couple that doesn't expect to spend much on medical, the extra premiums of a Silver plan don't make sense, so Bronze for us. On the exchange, I targeted about 250% FPL. Any lower and Medicaid seemed to kick-in. I re-ran 2014 taxes with various Roth conversion amounts (stepping 5K at a time) and I found that converting above 250% was generating a marginal tax rate of 29%. And if I hit the cliff, the marginal rate went over 400%. So I'm going to keep my Roth conversions small, but I discovered that going lower than 250% didn't cut my tax bill, so might as well convert up to that level.

 

[SecondCor521]

For us, a couple that doesn't expect to spend much on medical, the extra premiums of a Silver plan don't make sense, so Bronze for us. On the exchange, I targeted about 250% FPL. Any lower and Medicaid seemed to kick-in. I re-ran 2014 taxes with various Roth conversion amounts (stepping 5K at a time) and I found that converting above 250% was generating a marginal tax rate of 29%. And if I hit the cliff, the marginal rate went over 400%. So I'm going to keep my Roth conversions small, but I discovered that going lower than 250% didn't cut my tax bill, so might as well convert up to that level.

Thanks for the reply!

Makes sense. I'm planning a similar approach in terms of filling out a pro-forma return at 1K intervals and looking at the marginal rate and trying to pick a good inflection point.

I have low expected medical expenses, but I have minor kids and my ex-wife prefers more insurance to less, so I'm waffling between Bronze and Silver.

I also have a list of AGI cutoffs as to where credits and deductions disappear, so I'll be looking at that also. As HOH, I notice that my 15% bracket ends before 250% of FPL, so I may aim for that.

It's hard to keep track of so many moving parts!

 

[NW-Bound]

ACA premium subsidy drops off very rapidly with income. To see how it is so, I looked into how it works.

The subsidy is the difference between the 2nd lowest Silver plan in your location and what you have to contribute. What you have to contribute is a percentage of your income. As your income goes up, that percentage also goes up. So, your contribution goes up parabolically.

Here's an example. At 150% FPL, you have to contribute 4% of your income towards the premium. At 300% FPL, you have to contribute 9.5%. Therefore, in this case doubling your income causes your premium to go up 4.75x ( 2*9.5/4 = 4.75)

At 200% FPL, the percentage is 6.3%. The percentage flattens out at 9.5% past 300%. So, doubling your income from 200% FPL to 400% FPL causes your premium to go up only 3x ( 2*9.5/6.3 = 3.02x) and not 4.75x as the previous example.

And of course at 400% FPL + $1, you fall off the cliff.

Isn't it wonderful? How did they come up with the above? When you couple the above with the tax codes that we already have, you can run spreadsheet galore.

 

[pb4uski]

We fill up the 15% federal bracket.

Assuming current tax law in the future (not true, but what are you going to do...), we'll be up in the 25% or more when RMDs kick in in 15 years so convert as much as we can.

+1. Top of 15% tax bracket for the same reason... 15% is better than 20%. However, the reality is we pay about 8% because some of our converted amount is covered by deductions and exemptions, some is at 10% and some at 15% but much of the 15% is filled by LTCG and qualified dividends.

 

[Looking4Ward]

+1. Top of 15% tax bracket for the same reason... 15% is better than 20%. However, the reality is we pay about 8% because some of our converted amount is covered by deductions and exemptions, some is at 10% and some at 15% but much of the 15% is filled by LTCG and qualified dividends.

This is the first full year that I will not have any earned income so I've been eagerly anticipating the start of my own Roth conversions for years to come.

Per Taxcaster, after total estimated dividends and capital gain distributions this year I can convert $16K from tIRA to Roth before hitting the top of the 15% bracket. Taxes owed would be $623 or an effective rate of 3.9%.

If I bump that up to $30K it increases tax due to $4500 or an effective tax rate of 15%, a move I may consider because that might be better now than 15 years from now when I have to start RMD's.

Question: Do you prefer to wait until December when you have a more complete picture of the year's dividends and capital gains distributions to decide how much to convert to Roth and then pay any taxes due when you file your return, or do you convert throughout the year and pay estimated taxes?

 

[NW-Bound]

Having done Roth conversion the last 2 years, I always waited till the end after MF dividends were declared.

 

[samclem]

If I bump that up to $30K it increases tax due to $4500 or an effective tax rate of 15%, a move I may consider because that might be better now than 15 years from now when I have to start RMD's.

While the "tax torpedo" caused by RMDs and SS can be scary, if you've got a fairly normal situation I'd be strongly biased to avoiding paying taxes above the 15% rate right now if your motivation is just to avoid paying at tat rate, or even more, later. Rationale:
Your projections for the size of your taxable accounts and your IRAs (and thus RMDs) are likely to be highly dependent on how equities and bonds do for the next 15 years. Nobody knows how they'll do, and being off by a few percent per year can make a big difference in the size of your future taxes. Maybe you won't have a tax problem at all--maybe equities and bonds will do very poorly and you'll be on one of those FIRECalc lines we all dread. If so, you'll regret having given money to the government now that could be used to support needed spending/allow your portfolio to recover in future years. Or, equities might take off and you'll have doubled your money in 15 years, and owe a lot of taxes--that's no problem at all, you're still well ahead of where you thought you'd be, and 15 years farther down the line.
Unless you've got an unusual situation, I'd recommend keeping that "fuel" in your tank as long as possible. You don't know how long the trip will be, or what MPG you'll be getting.

Question: Do you prefer to wait until December when you have a more complete picture of the year's dividends and capital gains distributions to decide how much to convert to Roth and then pay any taxes due when you file your return, . . .?

Yes. And I even leave a bit of cushion in case something changes--an amended 1099, etc.

 

[kaneohe]

......................................

Per Taxcaster, after total estimated dividends and capital gain distributions this year I can convert $16K from tIRA to Roth before hitting the top of the 15% bracket. Taxes owed would be $623 or an effective rate of 3.9%.

If I bump that up to $30K it increases tax due to $4500 or an effective tax rate of 15%, a move I may consider because that might be better now than 15 years from now when I have to start RMD's.
......................................

Another way to look at this: converting in 2 steps.......
Conversion 1: convert 16K , pay taxes of $623
Conversion 2: convert additional 14K, pay taxes of $3877; pay marginal tax rate of 27.7%
You will have to evaluate whether the effective or marginal tax rate is what you want to look at and also the consequences 15 yrs later of doing what you choose to do.

 

[Looking4Ward]

Excellent points - thank you!

 

[Aeowyn]

Since it is so hard to predict the future... what will my portfolio returns be between now and RMDs? What will happen to tax laws... What will my earned income look like in the future before full retirement (DH self employed and I'm currently doing part time contract work and starting a cattle ranch).... Will we have some low income years between now and when we start collecting pension/SS that would provide a good opportunity for conversions?... Will one of us pass away causing the other to be in a higher or lower tax bracket?

I've decided to work with what I know today in determining conversion strategy. I look at what tax bracket we would be in if we were 70 today based on estimated income from pension (in today's dollars) + RMDs (based on current value of tax deferred accounts) + 85% of estimated SS (in today's dollars and assuming DH and I both wait until 70 to start collecting) and subtracting standard deduction and personal exemptions (based on current tax law).

Once I have the estimated tax bracket at age 70, my goal then is to be at the top of that bracket this year and then reevaluate each year. So for example if the market soars up I might be more aggressive in future years on conversions and if the market tanks I might get less aggressive. Also I'm 20 years from RMDs. As I get closer to age 70, I might base my estimated future bracket at an older age than 70.

I may have been too aggressive earlier in my Roth leanings. We currently have 70% of our retirement savings in Roth $$ vs. 30% tax deferred. Based on my current estimates at age 70, we will be in the 15% bracket. With our income this year I think we may be just into the 25% tax bracket so I might make some Traditional IRA contributions to lower us to the top of the 15% bracket.

 

[sengsational]

I've decided to work with what I know today in determining conversion strategy.

Sounds like a good (the only?) strategy. What makes it easier to do is that you're going to run the analysis periodically; no need to execute the entire plan you ran this year. The next analysis will take into account changes to laws and markets. In other words, you put a plan together that covers a lifetime, but you only execute, say, one year of it before re-evaluating. Could 'something big' come along that invalidates the entire long-term plan? Could that something big make what you did this year non-optimal. Sure. But if it was in the ballpark before, it's probably still in the ballpark after the markets and laws gyrate.

If you haven't done some conversion 'what-if' runs on i-orp, I'd suggest that.

 

[GTFan]

The i-orp tool now has the capability to run an optimization with PPACA "on" (taking advantage of the premium tax credit and limiting income) as well as with or without Roth conversions. Everyone's situation is different; some people get very little benefit from conversions and/or PPACA, others more. I'm a bird in the hand kind of guy, and PPACA currently comes out on top and I see the benefit each year, which is why I'm going with the PPACA over larger Roth conversions.

Same here - not doing conversions and keeping MAGI below 150% of FPL for a family of two to max out ACA Silver plan savings (we have lots of ways to live frugally while still enjoying vacations). I can't ignore the almost $900 a month that I'm getting out of this starting next year, not to mention the WAY lower out of pocket HC costs if something does happen to us during the year. I'm talking plans with $500 individual deductibles and $1500 max family OOP vs. a Bronze with $6450 and $13k.

Add that to lost growth on investments due to taxes on conversions and it tilts a lot more to ACA subsidies if you can manage income below 250% FPL, not to mention that ACA plans are getting worse every year so you *could* be better off getting a much better cost-shared Silver plan with that income.

My plan when the tax torpedo hits at 70 1/2 is to give as much as I can to charities to negate the tax hit. I'm not interested in a leaving a large legacy to my children.

Obviously, YMMV.

 

[kaneohe]

...................................

Add that to lost growth on investments due to taxes on conversions and it tilts a lot more to ACA subsidies.......................................

Your conclusion about not doing conversions due to ACA considerations may very well be true but I am not sure the above is a valid booster argument:
1)If you have 1000 in TIRA and leave it there and TIRA doubles, you will have 2000 in TIRA
2)If you convert 1000 in TIRA to Roth w/15% tax rate, you have 850 in Roth. If Roth doubles you have 1700. Lost growth is 150 because you put 150 less in Roth at the start.

Is the TIRA 2000 better than the Roth 1700? seems to be comparing Roth apples(AT) and TIRA bananas (BT) , no?

 

[NW-Bound]

Your conclusion about not doing conversions due to ACA considerations may very well be true but I am not sure the above is a valid booster argument:
1)If you have 1000 in TIRA and leave it there and TIRA doubles, you will have 2000 in TIRA
2)If you convert 1000 in TIRA to Roth w/15% tax rate, you have 850 in Roth. If Roth doubles you have 1700. Lost growth is 150 because you put 150 less in Roth at the start.

Is the TIRA 2000 better than the Roth 1700? seems to be comparing Roth apples(AT) and TIRA bananas (BT) , no?

If you stay in the 15% bracket, the tIRA 2000 will be worth only 1700 when you cash it out. In this case, banana = apple.

If withdrawing the tIRA 2000 causes you to enter the 25% tax bracket, then banana < apple.

If the market stays as lousy as it has been recently, I guess I will not have to worry much about how much to do Roth conversions. I can have either banana or apple, and it makes no difference.

 

[ejman]

If you stay in the 15% bracket, the tIRA 2000 will be worth only 1700 when you cash it out. In this case, banana = apple.

If withdrawing the tIRA 2000 causes you to enter the 25% tax bracket, then banana < apple.

If the market stays as lousy as it has been recently, I guess I will not have to worry much about how much to do Roth conversions. I can have either banana or apple, and it makes no difference.

Yes, if Mr. Bogle is correct in his 10 year forecast I suspect that the Roth conversion discussion will join the pantheon of how many angels can dance on the head of a pin... i.e. take SS at 62/66/70? or pay off the mortgage?

 

[Options]

I played a bit with I-ORP optimizer. Its recommendation for Roth conversion depends on what I enter as the projection for stock gain. At the default of 6% nominal gain over 2.5% inflation, it tells me to go max out to fill up the 25% bracket. When I cut the gain down to about 4%, it cuts back to the 15% bracket...

...Your projections for the size of your taxable accounts and your IRAs (and thus RMDs) are likely to be highly dependent on how equities and bonds do for the next 15 years. Nobody knows how they'll do, and being off by a few percent per year can make a big difference in the size of your future taxes. Maybe you won't have a tax problem at all--maybe equities and bonds will do very poorly and you'll be on one of those FIRECalc lines we all dread....

Emphasis added

This has always been my issue with i-orp. All outputs are based on guesstimates of future PF growth. As samclem wisely points out, if your guesstimate (inputs) is off by a few percent, it will affect your output. In my view, this makes i-orp's recommendations at best ball park estimates, if not unreliable. As they say, garbage in, garbage out.

 

[Whisper66]

Emphasis added



This has always been my issue with i-orp. All outputs are based on guesstimates of future PF growth. As samclem wisely points out, if your guesstimate (inputs) is off by a few percent, it will affect your output. In my view, this makes i-orp's recommendations at best ball park estimates, if not unreliable. As they say, garbage in, garbage out.

True of most calculations we do for estimating. I don't see I-orp as any worse than any other tool we use in that regard. But one has to make some basis for decisions. And that will include assumptions that may prove wrong over time.


Sent from my iPhone using Early Retirement Forum

 

[samclem]

True of most calculations we do for estimating. I don't see I-orp as any worse than any other tool we use in that regard. But one has to make some basis for decisions. And that will include assumptions that may prove wrong over time.

The next step is to examine the impact of being incorrect in each direction. If we overestimated the rate of portfolio growth, then we should have converted up to a lower level and the impact on QOL is possibly significant (because our investments have done poorly and we gave money away that is now needed). If we underestimated the rate of growth, there are no dire consequence (because our portfolio is fat) and, while we pay more taxes than optimum (in retrospect), we're still money ahead vs our planning.
So, I'll be very conservative with these conversions.

 

[hnzw_rui]

The next step is to examine the impact of being incorrect in each direction. If we overestimated the rate of portfolio growth, then we should have converted up to a lower level and the impact on QOL is possibly significant (because our investments have done poorly and we gave money away that is now needed). If we underestimated the rate of growth, there are no dire consequence (because our portfolio is fat) and, while we pay more taxes than optimum (in retrospect), we're still money ahead vs our planning.
So, I'll be very conservative with these conversions.

+1. It's not just the Roth conversions but accumulation, too.

My bad scenario favors maxing out tax deferred accounts with excess savings going towards Roth IRA then taxable versus putting everything in Roth 401k/403b/457b. The risk of prioritizing Roth is a much smaller portfolio.

My good scenario just puts me in a higher tax bracket come RMD time.

Between the two, I'd rather pay the higher taxes versus not have enough savings to live on.

 

[Whisper66]

Guess I'm not too concerned about the variance in my I-orp and other calculations when run using various reasonable assumptions. I'll use whatever I consider the most likely scenario to guide financial moves for that year. Then rerun the calcs each year using any new data and adjust plans as appropriate based on the results. So I may be doing aggressive Roth conversions for a couple years and then back off later on (for example).


Sent from my iPhone using Early Retirement Forum