AMT and State Tax Refund Inclusion

[JohnH]

I'm questioning my way of determining whether any of the state tax refund is taxable when a client was deep into AMT in the prior year. What I'm doing is to re-run the prior year return, reducing the Line 5 amount on schedule A by the amount of that year's state tax refund with a negative entry, and verifying that the new (lower) Line 5 figure transfers to the 6251. I then compare the net refund or tax due on this revised return to the same figures on the original return. If there is no change, then I know there was no tax benefit gained by the Line 5 entry in the prior year, and I can thus exclude the entire state tax refund form income in the current year.

I was reading a discussion of this matter yesterday, and the writer suggested the correct way to do it is to use Worksheet 2 of Pub 525. I don't think that worksheet can handle AMT, and I also think the Pub 525 says not to use it for AMT calculations. Can anyone enlighten me on how you do this calculation? Am I doing it incorrectly?

[Lion]

My software has a worksheet. It does the math. I review.

[Nashville]

Don't Think It will Fly

John, I'm not aware of any allowance to reduce the taxability of a state refund if the deducted taxes from previous year created dollars for the AMT. Lion from CT has posted that her software allows for it, but I haven't seen a cite or anything in instructions. Doesn't seem fair at all, but understand fairness is oblivious to IRS thinking.

A couple of related notes with respect to state tax refunds.

1) Many of my high-income clients will never receive a state refund, but will always insist that it be paid in for estimated state taxes. Until some time ago, that meant that IRS could not tax a state refund. However, in absolute antithesis to the concept of "cash basis taxpayer", they changed the definition of "refund" to include offsets, entitlements, blah blah etc. so they could collect taxes on an overpayment that was never received. In recognition of this, however, they did allow a like-dollar adjustment upward to the deduction of estimated state taxes in the following year.
2) Another unfair treatment is for full taxation of a state refund for a year when the deduction for state taxes has been eroded away for phase-out. In other words, full taxation of the refund when only a portion of the offsetting deduction was allowed. This one even went to court and the court decided it was the intent of Congress that "this is the price for having high incomes." On this issue, however, the IRS relented and now you can factor down the taxability of the refund in the same proportion that the deduction was limited. My guess is that the states applied pressure to the IRS. Governments and large institutions can influence them, while pleas from ordinary folks are ignored.
3) There has been discussion on this forum that in situations where the AMT robs taxpayers of the tax deduction, that taxpayers simply not deduct taxes on Schedule A to begin with. Whereas this may appeal to people who think they are "outsmarting" the system, I don't see where the math can lower the net tax liability at all by doing this. The very best they can do is break-even, if I follow the math.

I tuned in to an interesting comment from one of my instructors in a seminar last December about the future of the AMT. Whereas tax preparers and analysts have been clamoring for the eradication of the AMT for years, instead the people who are advising Congress are solving the problem in a different manner. This "manner" involves elimination of the AMT adjustments over a period of time where the AMT vanishes because there are no adjustments remaining. The "catch" however, is that instead of lowering the AMT income to match the "regular" income, the "regular" income can be raised to match the alternative income. As you might imagine, congressmen are so happy with this, they are squirming around in the floor in total delirium.

[JohnH]

Ron:
I have been following instructions on the middle of Page 25 of Pub 525, which adjoins the Worksheet 2. Those instructions provide a way to calculate if there was any tax benefit derived from the State Tax Deduction. It can happen when someone is just barely into AMT, but generally doesn't matter if they are deep into AMT.

As for the other issue (non-AMT cleints), I've always reported the state tax refund as income on the 1040 and reported the applied overpayment as state taxes paid on line 5 of Schedule A. This keeps the computers happy when the 1099-G matching commences. And it's representative of what really happened - constructive receipt and all that stuff. And then, any time the client is in a marginal range, you need to calculate the true tax benefit of the refund by bumping it against the sales tax deduction allowable.

[dtlee]

John,

You are correct on the tax benefit rule and I tend to agree with you on the issue of AMT. I am not sure if there is a better way to do this, but here is what I do (I am using an actual return for this, so the numbers are not nice round numbers):
Suppose there was an $8,037 State Income Tax Deduction in the prior year, and the State Tax Refund from that year is $1,011 and the amount of AMT on the return is $1,139. Based on the client's tax bracket, I do a quick computation of how much of a deduction was needed to eliminate that AMT Tax. For example, if the AMT marginal tax rate is 26%, I make a rough estimate that the amount of state income tax that does not produce a benefit would be $4,381 by dividing the AMT Tax by the AMT marginal tax rate. Thus, a refund of over $4,300 would have been needed to offset the AMT Tax. If this were closer, I would do as you describe and go back into the return to determine the exact impact. As I indicated, if I know that the refund is around $1,000 and the tax free amount is over $4,000 there is really no need for the exacting detail.

I do have some concern about just taking the entire amount of the refund as supplied by the state, though, since nearly half of my clients have only a partial or no tax benefit from the state income tax deduction. This is at least partially because the state sales tax here (Long Island, NY) is so high. Another reason is that many of my clients pay estimated taxes to the state. Even though I insist that they pay the state estimates by December 31, most of them do not. When they have a refund and they paid a Q4 estimate to the state in January, the computation of the taxable refund becomes more complicated (worse if multiple states are involved). It actually not only affects how much needs to be reported but how much of that January payment can be deducted. If you are using a product that correctly handles reducing the deduction, you are basically including that adjustment amount twice.

I am obviously in the minority when it comes to caring about this topic. I did a few posts several years ago asking how others handled certain non-AMT state tax refund situations. I ran the same scenario through four different packages and did not get the same result on any two of them. Only one of them seemed to handle the proper computation without any intervention. I did not get anyone volunteering to run the scenario through other packages to see how they did.

[JohnH]

Doug: Thanks for the confirmation and for the shortcut. I knew there probably had to be one to eliminate the obvious situations where the STR is all excludable. When I get a chance I'm going to work on a spreadsheet to do this and then do some back-testing. That would be much easier than simply re-running the entire return, comparing 5 or 6 entries for the proper adjustment, and then looking at the tax effect. It's never going to be easy, but it would be nice to have a shortcut to weed out the obvious ones and then only re-run those which MIGHT need some limited adjustment. I very seldom find an AMT situation in which any of the STR is taxable, but I still dutifully do the math "just in case".

One the other issue of taxability of the STR in general, I agree that it seems to be all over the place. The sales tax deduction has muddied this water a bit, and it's really difficult to glance at it and know for sure. I'll bet there are many returns filed showing the entire amount taxable when in fact at least a few hundred dollars could be lopped off. I use Drake, and I THINK it handles this calculation correctly. Their worksheet does take into account the Sales Tax Deduction in determining the proper amount to be taxed. I was pleased with how it handled them the year I switched to Drake. But this conversation causes me to want to run a few of my own cross-checks just to be sure.

[dtlee]

I am not one to bash Drake (I've been a user for over a decade). They have some very good computations with regard to the taxable refund of state taxes and some real shortcomings. In addition, they have a rather radical (some might call it foolhardy) treatment of state refundable tax credits. If you have some time, you might join the Drake forum here:

http://forums.drakesoftware.com/forums/index.php

I had quite a few detailed posts there a few years ago about some scenarios which were not handled well (i.e., properly) which, though rare, users should be aware of. In fact, my concerns about their poor handling of New York refunds had one user there referring to me as a "baby" for wanting the software to work properly. She and another user both wanted me banned from their forum for the blasphemy I was spouting about this perfect product.

If you do sign up there or are already a user there, find a post about state tax refund and reply to it asking me to point you to the posts about more complicated scenarios I was mentioning here and I will be glad to post the links there. Be aware, my blasphemous posts were hardly exciting. Drake has no problem with valid, open discussions about their product and they have never asked me to change or delete a post nor to leave. However, some of their users have been known to bristle at the thought of anything negative being posted about their beloved product. Any vendor would be envious of the devotion in some areas of their user base.

[Roland Slugg]

A real easy way to see if there was a tax benefit is to just go back to the prior year's return and reduce the state tax deduction by the amount of the refund. If the prior year's tax, including the AMT, doesn't increase, then there was no tax benefit, and none of the refund is taxable.

[dtlee]

A real easy way to see if there was a tax benefit is to just go back to the prior year's return and reduce the state tax deduction by the amount of the refund. If the prior year's tax, including the AMT, doesn't increase, then there was no tax benefit, and none of the refund is taxable.

Hence this thread goes full circle.

[JohnH]

Doug: I like the route it has taken, and at least I feel better about the method I've been using. But I do intend to play around with your shortcut. I think developing a spreadsheet to eliminate the slam dunks, and just running it on the second monitor while the return is printing might be a real time saver.

I think I've seen some of your posts over on the Drake site regarding this issue. I've been a Drake user for 3 tax filing seasons now, but since my client base is primarily NC & SC, I haven't identified any real problems. I'd propbably take more interest in the issue if I had clients in other states. And i know about the true believers - every vendor has them. I still participate in the Unofficial ATX Community for reasons other than the software, and there are a few true believers over there as well. Say something negative about ATX, and they jump right on it with a vengeance. When they get emotional, it usually makes me wonder if they are trying to convince me or themselves.