Roth Basis and First Time Homebuyer
Decided to start a new thread to continue this discussion.
Alan,
The reason you must enter Roth contribution basis on line 22 of Form 8606 if the distribution is partly NQ is so the tax on that portion can be calculated. The ordering rules apply to NQ distributions. I don’t understand how you derived no tax on earnings with an account closed. You must enter total distributions on line 19 including the Q first time homeowner distribution. The Q distribution is then backed out and line 22 basis is subtracted from that amount leaving the NQ taxable portion of the distribution. The table on pg. 8 of the Form 8606 instructions is for tracking basis. I rarely need to refer to this because my software does an excellent job of tracking.
With a previous year Q first time homeowner distribution and nothing taxable, line 22 of that 8606 would be blank. It is blank because the ordering rules do not apply to a Q distribution for the purpose of taxation. The Roth basis is still subtracted from the Q distribution and will be accounted for in later distributions, even though there is no entry on line 22. As you point out this shows up as an apparent recapture, although it was always there. Form 8606 is confirming that ordering rules still apply to Roth basis for a Q distribution. It also crossed my mind regarding a recovery from a total disability. I agree that this would be handled in the same manner. I think the only reason the first time homeowner distribution is singled out on Form 8606 is because of the 10,000 limit.
Ed C.
Permalink Submitted by Alan Spross on Wed, 2009-07-22 22:58
The table on p 8 is critical to determining the amount of tax basis to enter on line 22, and therefore it determines any Roth taxation for the entire year. Let’s break this down into 3 consecutive steps:
Assume a Roth balance of 30,000 composed of 20,000 of basis and 10,000 of earnings. Taxpayer drains the Roth before year end but a 10,000 qualified first home distribution is part of it. Assume this was in 2007 for p 8 chart purposes.
Ex #1) Form 8606 completed as follows: Line 19: 30,000; Line 20 10,000; Line 21 20,000. For line 22 the Inst say to enter -0- plus all your regular contributions over the years. These contributions were 20,000 as indicated above, so 20,000 goes on line 22. Line 23 is -0- (20,000 less 20,000 = 0). Since this is -0-, the 8606 is complete and there is no tax whatsoever because the earnings were effectively addressed on lines 19 and 20. The qualified portion of the year’s distributions absorbed them.
But what happens in the more typical case where the taxpayer ONLY takes out the 10,000 qualified part.
Ex #2) Form 8606 works out as follows: LIne 19 10,000, Line 20 10,000. Line 21 -0-. Line 22 would have been 20,000. Again, no tax, but look what happens when taxpayer moves to some following year, 2008.
Ex #3) Ex #2 8606 for 2007 is last 8606 filed. In 2008 later taxpayer takes out NQ 20,000, which is his basis, but this is where the recapture comes in. This 8606 would show 20,000 on line 19, 0 on 20, and 20,000 on 21. But now the Inst for line 22 goes to the chart and it says to enter on line 22 the figure from the PRIOR 8606 excess of line 22 over 19. This excess would be 10,000. You add -0- to this from the 3rd column of the chart since you made no new regular contribution. Line 23 then equals 10,000 (20,000 less 10,000). The chart has therefore reduced the basis by 10,000, due to the 2007 qualified distribution. This is where you were initially correct in that the ordering rules were essentially restored, even though they did not apply to 2007. The 20,000 distribution would now have an amount of 10,000 that was taxable, essentially erasing the benefit of the prior year.
But this did not happen when the entire Roth was drained in 2007 in Example 1. In that example, taxpayer drained the Roth with no taxes.
I think these examples show how difficult to explain this is, and why Pub 590 makes absolutely no attempt to summarize this under their definition of the ordering rules. This retroactive reduction of basis seems to agree with Q&A 4 of the IRS Reg you cited. But that Reg does not address what happens when the Roth is drained in the year of the qualified distribution, which is entirely different. Therefore, my original statement of the qualified distribution coming out of earnings only describes the result if the Roth is drained in the same year, not the more typical case where it continues.
Permalink Submitted by Edward Czapor on Thu, 2009-07-23 02:22
Alan,
I don’t think we have any disagreement here, only in how the end results are explained. I now see what you mean by earnings coming out tax free in example 1. When I first tried this I did not have the right scenario. I would explain these results by applying the ordering rules, as my contention is that the ordering rules always apply. For example 1, the ordering rules require regular contributions out first (20,000). This leaves 10,000 earnings as the Q distribution resulting in no tax.
In example 2 for 2008, the year after the 10,000 Q distribution, the figure from the table is actually the remaining basis after 10,000 of basis ( contributions out first) was taken up by the 2007 Q distribution. This is tracked by my software, and thats why I don’t have to refer to the chart. I did not mean to infer that this figure was not critical. The two very different outcomes are also due to the ordering rules. In example 2, the 2007 distribution came from basis leaving 10,000 of remaining basis, and finally the 10,000 earnings. This results in the earnings being taxed in 2008.
Thanks,
Ed c.
Permalink Submitted by Alan Spross on Thu, 2009-07-23 03:31
Ed,
Glad you questioned my original over simplification, because it produced a better understanding of what happens in different scenarios. Initially, it never occurred to me that the timing of the distributions could significantly change the outcome with no assumed investment results, and I never really paid that much attention to use of the p 8 chart.
Alan
Permalink Submitted by Edward Czapor on Sat, 2009-07-25 00:20
Alan,
From my interpretation of the ordering rules I did not think earnings out first was possible and therefore forced me to further invesigation. You enlightened me and probably many others to the fact that the first time homeowners exception could be worthless in some situations. This is probably the likely reason you deduced that earnings must come out first. This is logical and common sense. This is the way it should be, but we see it is not. A friend of mine who deals with environmental issues told me that trying to apply common sense to those laws and regulations often leads to trouble. Perhaps the same can be said of the tax code and regulations.
I’m sure you are aware, but I want to point out that in your example 1, the account would not have to be drained if the earnings were greater than the 10,000 absorbed by the Q distribution. You would have a taxable amount if more earnings were withdrawn, or if not, the additional earnings would remain in the account. This is what initially confused me about draining the account to remove earnings tax free, which did apply to your example.
Since the ordering rules appear to take preference with any distribution, it would make sense (could be trouble) to opt out of the homebuyer exception if this is permitted when there is no benefit, such as in your example 2.
If I had any say, the following instructions would be added to the instructions for line 20 of Form 8606:
In some situations first time homeowner expenses may not benefit you. Do not enter your qualified first time homeowner expenses on line 20 if after including those expenses on line 19, line 23 or line 25a are zero or less, and any converted amounts distributed have been held for five years or longer.
This may not be the best way to phrase it, but would effectively opt out of the exemption, and preserve this option should the taxpayer become requalified in the future.
Ed C.
Permalink Submitted by Alan Spross on Sat, 2009-07-25 01:37
Good point. No point wasting any part of the lifetime limit if there is no benefit, either from use of earnings tax free or waiver of the penalty. I think use of the exception is optional if it is beneficial to save it.
I think we have contrasted the differing affects of the exception with the two examples, and those two examples were selected to be simple ones. In most cases, if the taxpayer is getting any benefit at all, such as waiver of early withdrawal on recent conversions, they should bank it. Yet another question is if the ordering rules take the taxpayer into a recent conversion by 1,000 with a 10,000 distribution, can taxpayer simply opt to put 1,000 on line 20 when 10,000 was the qualified first home amount taken out? Obviously, the IRS will track the line 20 amount to determine when the lifetime limit has been exhausted.
Suffice it to say that this particular exception would make a good case for the tax simplification project if it ever happens.
Permalink Submitted by Edward Czapor on Sat, 2009-07-25 02:24
If you opt out in that situation you would have no entry on line 20. The total distribution would be entered as if there was no homebuyer exemption. You would pay no tax, but the 10% penalty would still apply to the $1,000. Or in other words you would have to pay $100 to preserve your homebuyer exemption for possible future requalification. I don’t think you could take a partial qualification on just the $1,000, because the ordering rules require contributions first.
Permalink Submitted by Edward Czapor on Sat, 2009-07-25 02:49
I will take that back. It looks like it is possible if the distributions are on separate 1099R’s or the total is on a single 1099R. My tax program assumes the $1,000 Q homebuyer comes from the conversion eliminating the penalty. So you should be able to preserve the remaining $9,000.
Permalink Submitted by Alan Spross on Sat, 2009-07-25 03:50
To get more than one 1099R, there would have to be more than one IRA account. If an amount is entered on line 20, there should be no need to add a 5329 to claim the penalty exception, as that would be redundant. I am not aware of any guidance on the subject, but it seems like the first item below should not be a problem, but either of the other items are pushing the envelope:
1) Opt not to claim ANY of a qualified FH distribution to save it for later years
2) Opt to claim PART of the qualified distributed amount on line 20 based on only one 1099R
3) Opt to claim PART of the qualified distributed amount on line 20 with more than one 1099R when all Roths are considered as one for tax purposes.
If both spouses had Roths, selection of which Roth to use or how to split the needed amount over the two provides even more planning options, and might provide flexibility with respect to the 3 items above.
Can you imagine the mess next year when someone does a conversion taxable in 2011 and 2012, and then takes a first home distribution which will accelerate the taxes on the conversion into the year of the FH distribution. You would then have the accelerated conversion taxes on the re designed 8606 as well as the first home distribution reported on the distribution section. Even if the FH distribution is tax free, it still accelerates conversion income into that year.
Permalink Submitted by Edward Czapor on Sat, 2009-07-25 16:15
If you can opt out there should be no problem with a partial Q distribution or the number of 1099R’s. I doubt the IRS would question such a transaction, since they would be relatively rare, and there is little revenue to be gained. Even If opt out was denied on any portion producing no benefit, no additional tax or penalty would be due in that year. Only lost revenue would occur on the chance that taxpayer requalified in a future year and would benefit from the first time homebuyer exception.The 2010 conversion options certainly ratchet up complexity and make planning mind bogling.
My statement of penalty elimination on the $1,000 converted amount was in comparison to not using the partial Q distribution on line 20. Simplification is usually accompanied by the inevitable grandfatherings, phase-ins, exceptions, and a creation of a new set of unforseen complexities. We can only wait to see to what degree true simplification may be advanced.