Early IRA liquidation vs ACA subsidy

[footenote]

I've read all the early (post-59.5, pre-RMD) IRA liquidation / conversion threads I can find on this site. (Shout-out to sengsational for many good references and comments.)

I then:
- read Kitces and Welch (i-orp) on RMD minimization
- used extended i-orp to get specific IRA liquidation recommendations
- run regular, spousal and non-spouse bennie* RMD calculators on current IRA amounts

Yikes! RMDs, particularly bad for DH if I check out first, running as high as $80k / year.

i-orp recommends I liquidate $90k the first year - another "yikes!" tax-wise, plus goodbye ACA subsidy.

Also considered the Kitces "bucket" route to liquidate only up to the edge of [whatever] bracket. This lessens the taxes, but even these lower liquidations eliminate the ACA subsidy.

So if I'm going to lose $6k in subsidies either way, maybe I should bite the tax bullet now to get to the "zero IRA funds by 70" Promised Land?

Have any of you taken the full i-orp-recommended liquidation route in your fifties and/or sixties?

*There is a chance (which I do not count on when running calculators) my step-mother may leave me a six-figure IRA inheritance. So I could have non-spouse beneficiary RMDs as well. Not complaining, but when I consider all possible RMD scenarios, it does appear wise to take at least some tax hit now to start minimizing them.

 

[Sunset]

There is nothing wrong with a $80k / year. I would love to have 5 or 6 of those.

I believe it is a mistake to actually get to "zero IRA funds by 70" , just look at what your RMD would be like if you got to $200,000 IRA , it would be ~ 7K for the first year, a pretty non-taxable amount.
Yes it is dependent upon your other income, so the tax paid will vary, but it will be in the lower brackets for OP.

My final reason for not wanting to go to zero IRA is that large medical and nursing home expenses, etc are still deductible. So when I go into a nursing home and am paying $70K per year for it, it will be nice tax-wise to pull that $70K from the IRA and deduct it all.

 

[zinger1457]

Don't see why you wouldn't just take IRA distributions from (59.5-65) to get you to the max ACA MAGI and still get a subsidy. Is emptying out your IRA during those years worth the $6K/year (~$36K) hit?

 

[GTFan]

QCDs are the answer to this problem.

 

[Col. Klink]

+1 to both Sunset and Zinger1457's comments. I am also this year while still at age 58 going to do a Roth conversion. The amount of this will be determine by my tax accountant, but the goal will be to move some money but still ensure we get a tax benefit on our 2019 ACA spending. I don't take the subsidies.

 

[RunningBum]

QCDs are the answer to this problem.

Not even close. First, the OP is talking about ACA subsidies, so they must not be 70.5 yet.

Second, giving away all of your retirement money is not really a sound strategy for managing taxes. It's certainly not an answer unless they planned to give away that much.

If one is still itemizing deductions, giving away appreciated assets is usually better than QCDs. It not itemizing deductions, doing QCDs is the best way to handle the charitable contributions they plan to make.

 

[RunningBum]

I don't use i-orp. I just did a quick run and it doesn't even consider the ACA subsidy. And it seems awfully aggressive to me on doing Roth conversions. I like using different tools but I always make sure I understand how they work, whether they consider all factors, and whether the output actually makes sense to me.

I don't have that much in my tIRA, so I think I can get pretty close by converting up to the subsidy cliff until 65, then doing the rest before taking SS, probably at 70. If I had a lot more, I'd consider giving up the subsidy I saw the probably tax savings at a lower rate now exceeded the subsidy. It's all a matter of using the best method to preserve the most of my money, I have no qualms about giving up the subsidy if it makes the most sense.

 

[walkinwood]

I've tried to model this scenario - roth conversions to the 12% bracket v/s aca subsidies and ACA subsidies always win even if I go right up to the 400% of poverty level. We are both in our mid-late 50s.



Our traditional IRA accounts are all invested in intermediate bond funds, so that may be why I get those results.



I once asked Kitces this question on his blog and he suggested that ACA subsidies were probably the way to go if you have that option.

 

[Dtail]

I don't use i-orp. I just did a quick run and it doesn't even consider the ACA subsidy. And it seems awfully aggressive to me on doing Roth conversions. I like using different tools but I always make sure I understand how they work, whether they consider all factors, and whether the output actually makes sense to me.

I don't have that much in my tIRA, so I think I can get pretty close by converting up to the subsidy cliff until 65, then doing the rest before taking SS, probably at 70. If I had a lot more, I'd consider giving up the subsidy I saw the probably tax savings at a lower rate now exceeded the subsidy. It's all a matter of using the best method to preserve the most of my money, I have no qualms about giving up the subsidy if it makes the most sense.

I thought that I-ORP does consider the ACA subsidy calculation in its Extended version.

 

[footenote]

Don't see why you wouldn't just take IRA distributions from (59.5-65) to get you to the max ACA MAGI and still get a subsidy. Is emptying out your IRA during those years worth the $6K/year (~$36K) hit?

In my case it would only be three years worth of ACA subsidy ($18k-ish). (I've already set the income / subsidy levels for 2019.)

I will play with the Kitces model, but it doesn't take much income to push me out of subsidies. That doesn't mean I couldn't moderate the withdrawal to get some subsidy, however....

 

[footenote]

There is nothing wrong with a $80k / year. I would love to have 5 or 6 of those.

I believe it is a mistake to actually get to "zero IRA funds by 70" , just look at what your RMD would be like if you got to $200,000 IRA , it would be ~ 7K for the first year, a pretty non-taxable amount.
Yes it is dependent upon your other income, so the tax paid will vary, but it will be in the lower brackets for OP.

My final reason for not wanting to go to zero IRA is that large medical and nursing home expenses, etc are still deductible. So when I go into a nursing home and am paying $70K per year for it, it will be nice tax-wise to pull that $70K from the IRA and deduct it all.

I did not consider this! (See, I knew you guys would broaden my perspectives on this.) I've always planned on "leaving" about $200k for LTC expenses toward the end, so that's a really good catch. Thank you!

 

[footenote]

I've tried to model this scenario - roth conversions to the 12% bracket v/s aca subsidies and ACA subsidies always win even if I go right up to the 400% of poverty level. We are both in our mid-late 50s.

Our traditional IRA accounts are all invested in intermediate bond funds, so that may be why I get those results.

I once asked Kitces this question on his blog and he suggested that ACA subsidies were probably the way to go if you have that option.

Thank you - I will model that and experiment with the i-orp ACA option in the extended version (which I saw but haven't fully explored).

I was taking too much of an "either / or" approach when this might be optimal:

1) Up-to-12% bracket conversions + as much ACA subsidy as possible from 62-65

2) More aggressive conversions 66-69 (once on Medicare)

 

[RunningBum]

I thought that I-ORP does consider the ACA subsidy calculation in its Extended version.

You're right, it does.

 

[GTFan]

Not even close. First, the OP is talking about ACA subsidies, so they must not be 70.5 yet.

Second, giving away all of your retirement money is not really a sound strategy for managing taxes. It's certainly not an answer unless they planned to give away that much.

If one is still itemizing deductions, giving away appreciated assets is usually better than QCDs. It not itemizing deductions, doing QCDs is the best way to handle the charitable contributions they plan to make.

The argument here is to ignore IRA WDs/conversions and then do QCDs at RMD time if it's a tax problem, therefore your ACA subsidy is not affected until Medicare kicks in. That's the approach we're taking when we hit 59 in a few years (admittedly we have taxable assets we can draw from so it's not an issue for us).

Or just wait until 65 and then start doing WDs or conversions, then QCD the excess you don't want at 70+. Don't get why folks are worried about taking excess RMDs, QCDs are the answer to that problem and are better for society as a whole.

 

[footenote]

Thanks to everyone who educated me.

Liquidating about 30k annually until I'm Medicare age and then maxing out annual liquidations up to [whatever tax bracket] until I reach 70 appears to be optimal.

This approach will
a) forfeit some but not all ACA subsidy
b) leave an appropriate amount of qualified funds for late-in-life LTC costs if needed
c) minimize RMDs in general and reduce possible double-tax torpedo if I should receive a beneficiary RMD

Thanks again

 

[Texas Proud]

I think others have hinted at it, but I will be specific...


Do IRA conversions at a lower rate than you suggested so you still can get an ACA credit... might not be as much as you had liked, but still probably substantial...


Then live with the RMDs....


BTW, I plan to do a lot of conversion between 65 and 70 and start SS when 70....

 

[Montecfo]

Thanks to everyone who educated me.

Liquidating about 30k annually until I'm Medicare age and then maxing out annual liquidations up to [whatever tax bracket] until I reach 70 appears to be optimal.

This approach will
a) forfeit some but not all ACA subsidy
b) leave an appropriate amount of qualified funds for late-in-life LTC costs if needed
c) minimize RMDs in general and reduce possible double-tax torpedo if I should receive a beneficiary RMD

Thanks again

I also consider that to be right strategy. Good luck!

 

[sengsational]

I'm not getting the huge recommendation for conversion in my case out of i-orp. Just my scenario, I suppose. But there are two things to make it give a more tame result. First, remove cash you have set aside for spending in the current year; you don't want i-orp to plan something for that stash if you're going to spend it. Second, make sure all tax buckets have the same percent equities. James confirmed that since the return on bonds is lower, optimization will drain that first. But most of us have an asset allocation that goes across tax categories. So setting equity percent the same across all categories will make taxes the lever, not the rate of return. This was some schooling that I got from someone on this board, and I was surprised at the magnitude of the change in i-orp recommendation. I lobbied for a warning in the help and a default of making equity percent equal, but didn't press the issue.