Capture Zero Tax LTCG Instead of Doing Roth Conversions?

[Huston55]

OK, I need some constructively critical analysis here (not usually difficult to come by in these parts �� ); and I consider that a good thing BTW.

Question: Using the ‘prototypical E-R.org member, is it better to capture LTCG in the lower tax brackets of ER (ages 57-70ish) or to do Roth conversions?

Profile: Using info from recent polls & threads (Thx MidPack) plus some interpolation, we’ll use a prototypical “ERorg” member (we’ll call him “Yorg”):
Age=57, married
NW=$2.9M
AA=50/50 (earns 4%/yr real)
Asset Location=40% taxable/40% tax deferred/20% tax free
Annual Spend=$84k
WDR=3.4% ($99k)
Pension Income= $25k
SS=$64k/yr (assume maxed out, deferred to 70 & 1.5 times max earner)

Stipulations: (I think these are reasonable, and we need to prevent this from getting too complicated)
1. Nobody can tell the future so, we assume tax brackets will remain the same (posted at the bottom for easy reference)
2. Mr. & Mrs. Yorg will live to their actuarial ages because, again, nobody can tell the future [Note: this means that MFJ tax brackets apply for almost all years.]
3. Market returns will remain within historical ranges (ie: nothing crazy stupid outside typical secular bull/bear markets) because,...well, you know.
4. This analysis is relevant only for what we can choose to do in the ER years (57 - 70.5) because, that’s the only time one has the capacity to influence it with the methods being evaluated (how to best fill the various marginal tax buckets).
5. Mr. & Mrs. Yorg have no overriding legacy goals. [Note: Assuming otherwise could complicate this analysis significantly. So, for this analysis, please abide by this stipulation.]
6. Managing to ACA (or other income based) subsidies is not considered.

Discussion: (only the beginning I’m sure �� )

I’ve read many (most?) of the E-R.org threads on Roth conversions and, the consensus seems to be that it’s wise (ie: minimizes lifetime tax bite) to do Roth conversions up to the XX% bracket during the early retirement, low income, pre-RMD years. I’ve spent less time @ BH but, the consensus there seems to be the same. That makes intuitive sense to me but, when analyzing the sample situation, I just can’t get “Roth Conversions” to always be the best choice for for the “Yorgs.” I keep coming back to capturing zero tax LTCGs first.

To simplify all the ‘puts & takes’ of the various options, I chose a simple model (Complicated ones make my head hurt): In my model, I accept that the Yorgs have an existing tax liability and, if they do nothing, they will pay exactly those taxes for the remainder of their time, this is “neutral” (neither positive nor negative); I view this liability as a line from left to right with time on the horizontal axis (the line can be straight, rising or variable; whatever works for you but, it actually doesn’t make any difference). The Yorg’s goal is to be below the line, defined as avoiding their existing tax liability (saving spendable NW). So, if they save $5k in taxes in year one, they’re $5k ahead (“below the line”); do it again in year two & they’re $10k ahead, and so on.

So, I said I concluded that capturing zero tax LTCG is best for the Yorgs. Here’s how I concluded that. They can capture “X” LTCG (on top of any other income) up to ~$77k and avoid 15% of “X”; they’re now X(15%) below the line of their baseline tax liability. They could continue to do this for all the RE years until RMD time, and build up a substantial amount of tax avoidance (increased NW). What if they did a Roth Conversion instead? Using the new tax brackets below, I see they’d pay 11.5% on the conversion, meaning they’re (temporarily) X(11.5%) above the line. [Note: We know from Kitces & others that if marginal tax rates remain the same, this turns out to be a wash come RMD time. Remembering my stipulation #1 above, we’ll call Roth Conversion “a wash.”].

But, what about the fact that the Yorg’s marginal tax rate during RMD time will be higher than it is now? And, if they convert to Roths now, they’d avoid those taxes right? OK, let’s look at this. We now have a numerical target to determine if Roth Conversions are better; it’s X(15%) that we got from capturing zero tax LTCG. We’ll examine several tax bracket scenarios.

1. Up through 12% Bracket, Age 57-70:
Income = $25k (pension) + $25k (DVD + Int) - $24k (2 Deductions) = $26k

LTCG to fill 12% bucket:
Available headroom (rounding)= $77k - $26k = $51k
Savings = $51k(15%)=$8k

Roth Conversion to fill 12% bucket: Savings is zero (See above, it’s a wash)

Conclusion: filling bucket with LTCG is best choice

2. Up through 22% Bracket, Age 71:
Income = $25k (pension) + $25k (DVD + Int) + $64k (SS) + $73k (RMD) - $24k (2 Deducions) = $163k

Conclusion: Doing conversions (age 57-70) into the 22% bracket would save the Yorg’s nothing because they remain in the 22% bracket even with RMDs. Plus, they would have foregone the opportunity to capture zero tax LTCGs.

3. Up through 24% Bracket, Age 93 (Max RMD):
Income = $25k + $25k + $64k + $126k (RMD) - $24k = $216k

Conclusion: It begins to get fuzzier here, since ~$50k of income is being pushed into the 24% bracket. But, that’s as compared to the 22% bracket (2% delta). I suppose in a scenario compared to the 12% bracket, the delta would be 10% but, that doesn’t seem reasonable, since we’re comparing it to the bracket we’d be doing conversions into - the 22% bracket. Since even this exaggerated scenario of $50k being pushed into a 10% higher bracket (22-12) still falls short of the $50k X 15% savings in #1 above, Roth conversions still don’t seem to be a winner.


OK, I know much of this rides on the return assumed. In today’s environment, I think 4% real for a 50/50 AA is very realistic. But, if we increase that from the 4% used above to 6%, I get the following results (big changes are RMDs so that’s what’s adjusted):

1. Same

2. 22% Bracket, Age 71:
Income = $25k + $25k + $64k + $96k (RMD) - $24k = $186k

Conclusion: Pushes $21k into a 2% higher bracket (24% vs 22%), which doesn’t equal the 15% savings on $50k for 13 years.

3. 24% Bracket, Age 96 (Max RMD):
Income = $25k + $25k + $64k + $261k (RMD) - $24k = $351k

Conclusion: Now we’re into brackets significantly above the 22% bracket we’d be doing early retirement Roth conversions in. The additional tax would be ($315k-$165k)(.24-.22) + ($351k-$315k)(.32-.22) < $4k/yr. Again, still not the $8k/yr benefit of capturing zero tax LTCG and, I hate the be the bearer of bad news but, with max RMDs at age 96, most of us will not have to worry about this trade off. But, it’s a good ‘worst case’ marker.


OVERALL CONCLUSION (and my advice to the Yorgs): Mr. & Mrs. Yorg should use their low income, pre-RMD early retirement years to capture LTCG before considering any Roth Conversions.

What would your advice be?

 

[zedd]

Did a quick scan and apologize if you mentioned it but are you concerned about managing income to maximize ACA subsidy? If so, all of your analysis gets a lot more complicated in that ACA takes priority and everything else take a backseat.

 

[Huston55]

Did a quick scan and apologize if you mentioned it but are you concerned about managing income to maximize ACA subsidy? If so, all of your analysis gets a lot more complicated in that ACA takes priority and everything else take a backseat.

Good question.

No, should not be considered. I’ve added that clarification.

 

[pb4uski]

But the conversion is not a wash. With the conversion the Yorgs pay 11.5% now and avoid paying 22% or more later, saving 10.5% or more. Admittedly, with your example, they save 15% (0% now compared to 15% later) by prioritizing LTCG, so LTCG is the better decision in the scenario that you laid out.

How much in unrealized LTCG do the Yorgs have?

In our case, we'll do both LTCG and Roth conversions. Since we are living on taxable accounts and those generate mostly qualified dividends and LTCG, our rebalancing (selling stocks for cash for spending) naturally results in some LTCG each year and those LTCG will decay over time as we use our taxable funds (meanwhile our tax-deferred and tax-free funds grow). My projections are that our taxable funds will be exhausted right about the time that we start SS depending on whether we go with FRA or with FRA for DW and age 70 for me.

Over the past 5 years, we have converted $260k and paid $19k in tax (7.4%).

From now until FRA we hope to convert another $300k and pay about 12% on average... so about 9.8% on the total vs 22% or more had we not done the conversions and waited until later. We will also do over $250k in LTCG at 0% between now and SS.

Since we expect our taxable account LTCG to be pretty much gone about the time we start SS, doing Roth conversions to the top of the 12% bracket is a slam dunk for us.

I guess I could do LTCG and then Roth conversions and it will be about the same, but I'm not sure that I would gain anything from doing it that way.

 

[Svensk Anga]

I think it is not either this or that, but some of both. If they have been investing with tax efficiency in mind, their taxable is mostly stocks. They are going to have to sell to meet expenses. Since they made it to ER, their stocks mostly show gains. With $50k of room each year, maybe they can sell $100k or more of stock to fill up the 0% LTCG bracket. One year's stock sale may carry them through 2 or 3 years. Might as well do Roth conversions in the off years.

You stipulated that both live to actuarial life expectancy, but that is unknowable. I think it is prudent to hedge the likelihood that one will be widowed and therefore filing single at some point.

I think it is also prudent to have a chunk of Roth money available for large, unpredictable expenses, so that meeting the need does not also generate a big high marginal rate tax bill.

 

[Huston55]

But the conversion is not a wash. With the conversion the Yorgs pay 11.5% now and avoid paying 22% or more later, saving 10.5% or more. Admittedly, with your example, they save 15% (0% now compared to 15% later) by prioritizing LTCG, so LTCG is the better decision in the scenario that you laid out.

PB-thx for the details on how you’re approaching this.

I don’t think I added enough detail to answer all questions, sorry for that (I was afraid I was already getting wordy in the post.) For example, how much unrealized LTCG to the Yorgs have? I didn’t specify but, given that 40% of their $2.9M portfolio is taxable and they retired @ 57, it’s likely quite a bit; enough for them to consider whether to take LTCG vs making Roth conversions.

Regarding whether the conversion (for the Yorgs) is a wash or not, the answer depends on “when” in their income hierarchy they recognize the Roth Conversion income. You assumed it’s from the first dollar (to get your 11.5% tax rate). I assumed it’s from the top of the 12% bracket so they could achieve maximum savings capturing LTCGs at Zero tax, resulting in Roth conversions being taxed at 22%; making it a wash.

I suppose we can’t say one is right or wrong but, I’d like hear your rationale for filling up the 12% bracket with Roth Conversions versus untaxed LTCG.

 

[Texas Proud]

It was a long OP, so I did not read it all... but...

Here is my take... why take CG if you do not need to? IOW, let the gains grow as much as possible and wait until a death and get step up in basis...


As others have mentioned, I would probably do a little of both since I need to take CG for living expenses right now... but, I have a large ACA credit that trumps the whole thing...

Also, I do not have pension and will not get close to the max SS.... I think I will be in the 12% bracket the rest of my life... well, or 0%....

 

[Huston55]

It was a long OP, so I did not read it all... but...

Here is my take... why take CG if you do not need to? IOW, let the gains grow as much as possible and wait until a death and get step up in basis...


As others have mentioned, I would probably do a little of both since I need to take CG for living expenses right now... but, I have a large ACA credit that trumps the whole thing...

Also, I do not have pension and will not get close to the max SS.... I think I will be in the 12% bracket the rest of my life... well, or 0%....

Thx for the input.

I chose the ‘Yorgs’ because I thought they’d come closest to representing the prototypical E-R.org member; sounds like it doesn’t do a good job of describing your situation. But, I’ll try to address your questions anyway.

- I believe heirs get a stepped-up basis for taxable, tax deferred & tax free accounts.
- I purposely did not address ACA subsidies because doing so would essentially negate the exercise. Although, I know that’s an overriding factor for many.
-With relative certainty you’ll remain in the 12% bracket, it seems to me that if you needed to tap your portfolio, capturing LTCGs @ Zero tax would be optimal. And, even if you don’t need all the LTCGs up to the top of the 12% bracket to cover current expenses, you’d benefit by filling up that bucket with them, then immediately repurchasing the same asset, thus raising your cost basis & lowering your future tax liability.

Food for thought.

 

[Montecfo]

I do not really think there is a rule of thumb as it depends on factors that vary so much even among "typical" folks.

But one observation I would make is that recognizing a LTCG provides funds for spending. Roth conversions do not.

Secondly, when you take a LTCG, the funds continue to generate future taxes. Roths do not.

Bottom line is you have to run the numbers in each case. I tend to think Roth conversions would be favored in many cases because they eliminate future taxes.

 

[Huston55]

I do not really think there is a rule of thumb as it depends on factors that vary so much even among "typical" folks.

But one observation I would make is that recognizing a LTCG provides funds for spending. Roth conversions do not.

Secondly, when you take a LTCG, the funds continue to generate future taxes. Roths do not.

Bottom line is you have to run the numbers in each case. I tend to think Roth conversions would be favored in many cases because they eliminate future taxes.

Yep, all good observations. Also some of the reasons I said in the OP that ‘intuitively’ Roth conversions made sense to me.

The only additon I’d make is that LTCG capture does not have to provide for spending; it can also be used to increase the cost basis of the investment with no taxes due. But, I agree that in most cases the funds are likely spent by a couple like the Yorgs.

 

[pb4uski]

PB-thx for the details on how you’re approaching this.

I don’t think I added enough detail to answer all questions, sorry for that (I was afraid I was already getting wordy in the post.) For example, how much unrealized LTCG to the Yorgs have? I didn’t specify but, given that 40% of their $2.9M portfolio is taxable and they retired @ 57, it’s likely quite a bit; enough for them to consider whether to take LTCG vs making Roth conversions.

Regarding whether the conversion (for the Yorgs) is a wash or not, the answer depends on “when” in their income hierarchy they recognize the Roth Conversion income. You assumed it’s from the first dollar (to get your 11.5% tax rate). I assumed it’s from the top of the 12% bracket so they could achieve maximum savings capturing LTCGs at Zero tax, resulting in Roth conversions being taxed at 22%; making it a wash.

I suppose we can’t say one is right or wrong but, I’d like hear your rationale for filling up the 12% bracket with Roth Conversions versus untaxed LTCG.

There is really no question of when in the income hierarchy it is... since both LTCG and Roth conversions are discretionary, they get the marginal rates... the pension income, Qdivs and interest come first because they are non-discretionary. In my case I focus on the top of the 12% tax bracket because once you are in 22%land, there isn't much benefit since the next higher bracket is 24% and the 2% difference is hardly worth fretting over and the brackets are broad. At that point the answer is clear because 15% is a lot better than 22%.

Our choice would have been to do both the way we are doing or do LTCG first and then Roth conversions.... but no matter which we did by the time we were SS age we would have ended up in roughly the same spot since we would have just run out of capacity to do LTCG for the entire time... which is why how much the Yorgs have in unrealized LTCG could have a bearing. IF their $1,160k taxable is 50/50 AA that would be $580k in stock... and if that is 50% unrealized gains then that would be $290k so at $51k a year the capacity to do LTCG would be gone in 6-8 years so they would then turn to Roth conversions.

 

[jj]

Thx for the input.

I chose the ‘Yorgs’ because I thought they’d come closest to representing the prototypical E-R.org member; sounds like it doesn’t do a good job of describing your situation. But, I’ll try to address your questions anyway.

- I believe heirs get a stepped-up basis for taxable, tax deferred & tax free accounts.
- I purposely did not address ACA subsidies because doing so would essentially negate the exercise. Although, I know that’s an overriding factor for many.
-With relative certainty you’ll remain in the 12% bracket, it seems to me that if you needed to tap your portfolio, capturing LTCGs @ Zero tax would be optimal. And, even if you don’t need all the LTCGs up to the top of the 12% bracket to cover current expenses, you’d benefit by filling up that bucket with them, then immediately repurchasing the same asset, thus raising your cost basis & lowering your future tax liability.

Food for thought.

There is no step-up in basis for heirs with tax-deferred accounts, that is only on taxable accounts. Tax-free (Roth) accounts have RMDs for non-spouse heirs, but no tax to pay on those mandatory withdrawals.

 

[karen1972]

Well with the tax bracket going from 15 to 12 it does change things slightly in favor of LTCG, else it would have been a wash.

For me I'm much younger, 45, so it changes those equations tipping it towards ROTH. One because RMDs will be significantly higher and two because there are years where it suggests I dip into the ROTH to pay expenses to prevent pushing other income into the higher brackets.

 

[Huston55]

Well with the tax bracket going from 15 to 12 it does change things slightly in favor of LTCG, else it would have been a wash.

For me I'm much younger, 45, so it changes those equations tipping it towards ROTH. One because RMDs will be significantly higher and two because there are years where it suggests I dip into the ROTH to pay expenses to prevent pushing other income into the higher brackets.

Good point about being younger than the Yorgs changing your tipping point/strategy. When you say, “...where it suggests...”, do you mean i-ORP or some other tool?

 

[karen1972]

Good point about being younger than the Yorgs changing your tipping point/strategy. When you say, “...where it suggests...”, do you mean i-ORP or some other tool?

Yes i-ORP... I'm not as aggressive with conversions as i-ORP suggests because of ACA tax credits.

I also sold some high dividend stocks because I'm thinking dividends are costing me because they fill up too much of my "freebie" space. Maybe I"m thinking it wrong, but with a 60% cost basis, I'd prefer to take $12K cap gains than $5K dividends for the same $5K hit to AGI.

 

[Huston55]

Yes i-ORP... I'm not as aggressive with conversions as i-ORP suggests because of ACA tax credits.

I also sold some high dividend stocks because I'm thinking dividends are costing me because they fill up too much of my "freebie" space. Maybe I"m thinking it wrong, but with a 60% cost basis, I'd prefer to take $12K cap gains than $5K dividends for the same $5K hit to AGI.

I think you mean “sell” $12k of stocks (including $4.8k in cap gains), instead of “take $12k cap gains” which would add $12k to your AGI. But, yes, I see why you’d do that.

 

[audreyh1]

I’ve been pondering this issue. With the new higher AMT levels we could actually realize up to $109K in ordinary income with Roth conversions without paying AMT. The cap gains income could be much higher without pushing ordinary income into AMT rates. This is a big change from prior years.

We don’t necessarily realize gains - well maybe occasionally due to rebalancing. But we do get a lot of qualified divs and cap gains distributions. So that already fills up our 0% cap gains slot.

So then it’s a trade off between pushing ALL cap gains out of the 0% cap gains rate due to ordinary income moving into the 22% tax bracket.

For the cap gains chunk pushed out of the 0% bracket, it would mean 12%+15%=27% at least. And maybe 12%+15%+3.8% = 30.8% for some of it.

Once we start receiving SS and RMDs we’ll probably no longer be within the 12% tax bracket, so we’ll be paying 22% on some of it and no longer get 0% cap gains. Kind of the same situation.

We can enjoy say 7 years of lower tax rates since we will no longer pay AMT, enjoy the 12% tax bracket, and pay 0%, 15% and maybe some at 18.8% on our capital gains rate income.

Or we can pay 12% and 22% on Roth conversions but lose our 0% cap gains bracket which effectively means 27% to 30.8% on the amount pushed out, and 15% and 18.8% on the rest.

Our tax-deferred funds are about 15% of our retirement assets, so they don’t represent a huge jump in income, but it’s still a meaningful jump with higher taxes on the new income which would include SS as well. If our ordinary income exceeds $109K with the RMDs plus SS we’ll pay 26% on some of it due to AMT. Well, until 2026 when the old AMT levels reappear.

I’m not sure we can draw down our IRAs enough to make that much of a difference. Say both drawing SS already knocks us out of the 0% cap gains slot. Then the RMDs don’t do much additional damage.

 

[audreyh1]

What does capture LTCGs mean? Does it mean that they are spending down their equities and thus reducing their equity exposure in taxable accounts?

Equities in general tend to be the more tax-efficient investment, and if held forever pass to heirs with a step up in basis. It’s not like you have to realize gains in your lifetime except maybe for rebalancing and then it can be done with selecting highest cost basis shares to reduce the realized gains. There are also occasional opportunities for tax loss harvesting.

 

[walkinwood]

For an ER'd couple with no earned income, $24,000 of non-qualified divs + int is tax free.

Assume you've got your fixed income in a T-IRA.

Is there an advantage to
1) Converting funds from T-IRA to ROTH (say to the top of the 12% bracket)
2) Investing ROTH funds into Equity assets
3) Selling equities in your taxable account and buying bonds there. @ 4% yield, you can have $600K in bonds in your taxable account and not pay any income tax.

Selling equities would generate cap gains of course. But you can do some selling & some ROTH conversions. And then there are the cap gains distributions to consider. I gets complex.

There may be merit to this, but I haven't been able to focus enough to create a model for it.

 

[pb4uski]

What does capture LTCGs mean? Does it mean that they are spending down their equities and thus reducing their equity exposure in taxable accounts?

Equities in general tend to be the more tax-efficient investment, and if held forever pass to heirs with a step up in basis. It’s not like you have to realize gains in your lifetime except maybe for rebalancing and then it can be done with selecting highest cost basis shares to reduce the realized gains. There are also occasional opportunities for tax loss harvesting.

It means selling equities at a gain and rebuying the same equities that same day... or the next day... if 0% then you just accomplish a stepped-up basis at no cost.

 

[harley]

- I believe heirs get a stepped-up basis for taxable, tax deferred & tax free accounts.

Heirs get a stepped up basis for taxable, and there are no taxes on tax free accounts. But there is no step-up for tax deferred accounts. They have to pay tax at their bracket for whatever RMDs they have to take with those accounts. It doesn't matter what the accounts are invested in.

 

[Huston55]

Heirs get a stepped up basis for taxable, and there are no taxes on tax free accounts. But there is no step-up for tax deferred accounts. They have to pay tax at their bracket for whatever RMDs they have to take with those accounts. It doesn't matter what the accounts are invested in.

Yep, correction noted earlier above. Thx.

 

[pj.mask]

My question is (and better calculated closer to realization) would alternating converting capital gains and Roth conversion early to allow ACA or other type subsidies late 50s - considering ill probably be in the 24% bracket throughout retirement and I don’t want to run into oversized RMD as I intend to leave an inheritance. Not quite the question asked but what I have been musing lately.

 

[pb4uski]

The impression that I have is that the hurdle for ACA subsidies is so low that you can do a little, but generally not enough to move the needle very much. IIRC subsidies phased out at around $65k of AGI (so $41k for TI in 2018) for a couple and the subsidies are better if your income is lower.

 

[Finance Dave]

The impression that I have is that the hurdle for ACA subsidies is so low that you can do a little, but generally not enough to move the needle very much. IIRC subsidies phased out at around $65k of AGI (so $41k for TI in 2018) for a couple and the subsidies are better if your income is lower.

Can you clarify your statement please? What do you mean the hurdles are so low that you can't do enough to move the needle?

400% of FPL is the MAGI point at which ALL subsidy disappears. For 2018 that number is $64,960.

And how did you come up with $41k? I assume you're taking the $64,960 and subtracting the $24k exemption for a married couple?

Thanks in advance for clarifying