401k's, IRA's and RMD's - tax consequences

[Chuckanut]

Finally, one doesn't use sequential withdrawals (all taxable first, all tax-deferred next, all Roth last), but one makes withdrawals from a mixture of account types in order to have the lowest taxes.

+1

That's why I am converting some of my tIRA money into my smaller Roth IRA's. I want that flexibility later in life.

 

[karluk]

Basically, I'm thinking that capital gain harvesting (CGH) at 0% if possible is priority number one prior to roth conversion.
The ability to CGH is limited by existing income (pension, dividends, filing status)
If a person fails to CGH they have less opportunity once SS starts, and once RMD's start, they will pay 15% on capital gains, assuming a large IRA/401K savings.
Comments ?

I'm seeing an edge for Roth conversions, but it's not very big and I suppose is susceptible to the assumptions one makes. I used as an example a person who will be $10,000 below the top of the 15% tax bracket at the end of the year and in order to fill up the tax bracket has a choice between converting $10,000 tax deferred to a Roth and realizing $10,000 in LTCG from a $20,000 taxable investment with a $10,000 cost basis. To pay taxes, I assumed $1,500 in a money market account, which either is used to pay taxes or is added to the taxable account if no taxes are due. He wants to know which choice will give him more after tax money ten years from now, assuming 5.4% growth no matter where the money is located, 15% tax now on Roth conversions, 0% now on LTCG, 25% in ten years on withdrawals from the tax deferred account, and 15% in ten years on LTCG.

Case 1: Leave the tax deferred money alone and realize $10,000 LTCG. No federal taxes due, so the taxable account starts with $21,500 and a $21,500 cost basis. In ten years, the tax deferred account will have grown to $36,378.48. If the cost basis hasn't changed, he will owe $2,231.77 LTCG tax and be left with $34,146.71 from the taxable account. The tax deferred account has grown to $16,920.22 and provides an additional $12,690.17 after paying 25% in taxes. The total is $34,146.71 + $12,690.17 = $46,836.88.

Case 2: Convert $10,000 tax deferred to a Roth IRA, leaving all of the capital gain unrealized. The taxable account starts with a $20,000 balance and a $10,000 cost basis. In ten years the Roth will have grown to $16,920.22 and can be withdrawn tax free. The taxable account has grown to $33,840.45. If the cost basis hasn't changed, it will generate LTCG taxes of $3576.07, leaving $30,264.38. The total is $16,920.22 + $30,264.38 = $47,184.60. This is $347.72 more than the after tax value of Case 1.

If find this result counterintuitive, but can't find anything wrong with the math. Perhaps someone else will weigh in with another perspective.

 

[samclem]

I'm seeing an edge for Roth conversions, but it's not very big and I suppose is susceptible to the assumptions one makes.
. . .
If find this result counterintuitive, but can't find anything wrong with the math. Perhaps someone else will weigh in with another perspective.

Here's one wrinkle/thing to consider. From a post I wrote a previous time we went down this path:

- Survivor/Filing Single Taxes . . . .: Assuming you are in the upper portion of the 15% bracket when MFJ, a surviving spouse would probably have marginal rates well into the 25% range. Under the above conditions:
-- Cap Gains harvesting: Every dollar of basis that is reset higher now saves the survivor 15 cents (for others following along, that's the cap gains tax rate for those in the 25% income tax bracket)
-- Roth Conversion: Every dollar converted to a Roth now costs 15 cents in present taxes, but saves the surviving spouse from paying 25 cents in taxes. Net benefit: 10 cents.

So, looked at from this angle (minimizing taxes to a surviving spouse), the CG loss harvesting has a 5% edge (or, "provides 50% greater benefit!"). Plus, it keeps more money in your account (because you didn't pay those 15% taxes this year), and that's "pad" against uncertainties of the future. Market returns might stink for 20 years: in that case you'll be glad you kept this money in hand AND you'll avoid high taxes after all

And here's another recent thread that Midpack started on the "Roth convert or don't" issue, with lots of other contributors.

 

[karluk]

- Survivor/Filing Single Taxes . . . .: Assuming you are in the upper portion of the 15% bracket when MFJ, a surviving spouse would probably have marginal rates well into the 25% range. Under the above conditions:
-- Cap Gains harvesting: Every dollar of basis that is reset higher now saves the survivor 15 cents (for others following along, that's the cap gains tax rate for those in the 25% income tax bracket)
-- Roth Conversion: Every dollar converted to a Roth now costs 15 cents in present taxes, but saves the surviving spouse from paying 25 cents in taxes. Net benefit: 10 cents.

So, looked at from this angle (minimizing taxes to a surviving spouse), the CG loss harvesting has a 5% edge

This quote captures the intuitive feel that LTCG harvesting should beat Roth conversions, but is a very poor way to frame the issue. I'm almost certain that the surviving spouse gets a step up in basis anyway, so in the case the spouse dies, it would have been a VERY poor choice to pick LTCG harvesting over Roth conversions. The surviving spouse gets no benefit at all from the LTCG harvesting and misses out on the benefits of having money in a Roth IRA, now that he or she is in a higher tax bracket.

 

[karluk]

If find this result counterintuitive, but can't find anything wrong with the math.

Looking at it a little more, I see that with the assumptions I made, the after tax winner depends on how long the holding period is before the assets are sold. LTCG harvesting wins until year seven, when Roth conversions gain the edge. I guess years of tax free growth are the deciding factor for the longer holding periods.

 

[samclem]

This quote captures the intuitive feel that LTCG harvesting should beat Roth conversions, but is a very poor way to frame the issue. I'm almost certain that the surviving spouse gets a step up in basis anyway, so in the case the spouse dies, it would have been a VERY poor choice to pick LTCG harvesting over Roth conversions. The surviving spouse gets no benefit at all from the LTCG harvesting and misses out on the benefits of having money in a Roth IRA, now that he or she is in a higher tax bracket.

The step-up in basis depends on whether the assets are owned jointly or not. If owned jointly, the surviving spouse gets the benefit of 50% of value being stepped up in basis. If the assets were owned solely by the deceased spouse, then the survivor gets a full step-up in basis.

Thanks for bringing that up. That's why we have these exchanges.

If TaxCut or TurboTax wants a competitive advantage, this is the kind of stuff they should include in their software. And, throw in a mdule that optimizes (present and long-term) for ACA subsidies..

 

[karluk]

The step-up in basis depends on whether the assets are owned jointly or not. If owned jointly, the surviving spouse gets the benefit of 50% of value being stepped up in basis. If the assets were owned solely by the deceased spouse, then the survivor gets a full step-up in basis.

Thanks for bringing that up.

Ok, thanks. I didn't know that wrinkle about a 50% step up for joint assets.

 

[Cat-tirement]

Basically, I'm thinking that capital gain harvesting (CGH) at 0% if possible is priority number one prior to roth conversion.
The ability to CGH is limited by existing income (pension, dividends, filing status)
If a person fails to CGH they have less opportunity once SS starts, and once RMD's start, they will pay 15% on capital gains, assuming a large IRA/401K savings.
Comments ?

I am thinking the same thing. Next year will be my first full year of retirement (though DW is still working 3 days/week), so I plan to take the max LTCG I can while paying 0% on it. This is even more of a priority for us, before moving from no state income tax where we are now to a high state income tax where LTCG are taxed as ordinary income.

 

[PatrickA5]

I am thinking the same thing. Next year will be my first full year of retirement (though DW is still working 3 days/week), so I plan to take the max LTCG I can while paying 0% on it. This is even more of a priority for us, before moving from no state income tax where we are now to a high state income tax where LTCG are taxed as ordinary income.

That's what I'm doing this first year of semi-retirement. My first goal is to get as much LTCG and Qualified Dividends as possible taxed at 0%. I should have about $19K combined. I still have enough of the 15% bracket left to convert $10K each for my DW and my Roths this year. My state doesn't tax the first $10K (per spouse) of retirement income, so my entire Roth conversion should have 15% Federal taxes and 0% State taxes. I think that's about as good I'm going to get it this year. Next year won't have nearly as much LTCG, so I should be able to convert more IRA money.

 

[Sunset]

Fantastic discussion !
At supper I did think of the roth conversion having the advantage due to 10 yr time, and that it would extend more every year as it eats up the 5% difference over time.
I never would have thought of the step up factor due to death, (its not hopefully near on the horizon).
Possibly collecting the LTCG for short term issues, like buying a car/boat has value vs IRA withdrawl.

 

[Sunset]

I am thinking the same thing. Next year will be my first full year of retirement (though DW is still working 3 days/week), so I plan to take the max LTCG I can while paying 0% on it. This is even more of a priority for us, before moving from no state income tax where we are now to a high state income tax where LTCG are taxed as ordinary income.

We should trade houses as I want to move to a tax free State especially one that is warmer

 

[Sunset]

That's what I'm doing this first year of semi-retirement. My first goal is to get as much LTCG and Qualified Dividends as possible taxed at 0%. I should have about $19K combined. I still have enough of the 15% bracket left to convert $10K each for my DW and my Roths this year. My state doesn't tax the first $10K (per spouse) of retirement income, so my entire Roth conversion should have 15% Federal taxes and 0% State taxes. I think that's about as good I'm going to get it this year. Next year won't have nearly as much LTCG, so I should be able to convert more IRA money.

Wow, your post made me think could my State be like that, and it is , so now I have to crunch some numbers to see if we can squeeze some in this year.

 

[Likes_to_Lurk]

As I was doing my own analysis, I realized that New York state excludes the first $20k of retirement income (including IRA withdrawals).

 

[pb4uski]

I prioritized LTGC harvesting in my first year of retirement and then decided to prioritize Roth conversions. For us, harvesting will occur"naturally" in ER since we are living off taxable accounts.

 

[Cat-tirement]

We should trade houses as I want to move to a tax free State especially one that is warmer

Sorry I have to disappoint you, but my house isn't in a particularly warm place (unless you're comparing it to some place like Minnesota or North Dakota). There are also rather hefty property taxes here that somewhat counter the lack of income tax.

 

[Cat-tirement]

As I was doing my own analysis, I realized that New York state excludes the first $20k of retirement income (including IRA withdrawals).

Surely that can't last. Back when I lived in NY state for a few years, it seemed like they kept finding ways to tax almost anything.

 

[Likes_to_Lurk]

Surely that can't last. Back when I lived in NY state for a few years, it seemed like they kept finding ways to tax almost anything.

Cat,
I hope this will stand for awhile. There are sooo many "wrinkles" in the NY state tax code, maybe it will take a few years to find this one!