DENNIS: The first question that we're going to deal with is parallel GRATs. The “play” with a parallel GRAT deals with a revocable spousal interest and a revocable spousal interest was not mentioned in Sec. 2702. It also wasn't mentioned in the proposed regulations and made its first appearance in 1992 in the final regulations.
As you recall, we've had two cases dealing with the validity of the valuation of revocable spousal interests. The Schott[????? case held that a revocable spousal interest had value, but the Cooke??? case held that it didn't have value for purposes of determining the valuation of the remainder interest in a GRAT. The 2005 final regulations under Sec. 2702 that addressed the Walton case also addressed revocable a spousal interest; the Walton case adopted the Cooke approach.
As you look at the validity of the parallel GRAT technique, pay particular attention to example 8 in the final regulations. Now, let me describe for you the technique referred to as a “parallel GRAT”:
My wife is Vicky. Vicky and I have an estate that is in the $5 million range. We're concerned that we won't have enough assets to live on, but we're also concerned that our children will have to pay estate taxes. So what Vicky and I will do is take $2 million each. I'll take my $2 million and put it in a GRAT and reserve a two-year term that will pay me 6%.
The first term will be pay me at a GRAT amount — an annuity amount — equal to 6%. That two-year term will terminate either at the end of two years, or upon my death. Following the two-year term, there will be a term for Vicky that will last 25 years or the shorter of her life. So the second interest is Vicky's 25% — a 25-year interest for a 6% amount that will terminate either after 25 years or on Vicky's death — whichever occurs first.
At the same time, Vicky will take her $2 million and make a parallel GRAT that will have a two-year term for Vicky or her death — whichever occurs first, followed by a 25-year term for me, or the shorter of my death. In other words, if either one of us dies, one GRAT terminates.
In both of those spousal interests — the GRAT that I create for a 25-year term for Vicky — I will reserve the right to revoke Vicky's interest by will at any time during the entire term of the GRAT. In Vicky's GRAT, she will retain the same right to revoke my interest at will at any time during the GRAT.
We put our $2 million in with a 6% payout — with two 60-year-old individuals ????????? But neither Vicky nor I are 60 years old yet. So there will be a taxable gift of 37.47%. In other words, the remainder interest will be 37.47%.
Now, assuming that that GRAT runs its full term and has a yield of income and appreciation of 6%. At the end of the term there will be approximately $1 million in it. In other words, the payout will equal that, so at the end of the term, the children will get $1 million from each GRAT — or $2 million total. The value of the GRAT will hold its value.
What we've done in this instance is transferred $2 million, with a gift tax cost of 37%. Vicky and I will get that payment as it goes out. If I die early, the children will win because the GRAT that Vicky created for me will terminate early and the children will inherit the funds before the 25-year term expires.
It is important that I exercise my power to revoke Vicky's interest in my will if she dies in that 25-year term, because that 25-year term to Vicky will not qualify for the gift tax marital deduction. And so, if I have not exercised that power to revoke or if I have not retained that power to revoke, then there's a taxable gift equal to that 25-year term to Vicky that will not qualify for the federal gift tax marital deduction.
One thing that we always do when we're looking at techniques like this is to assume just a flat-out gift. So in other words, rather than give the children $2 million in a GRAT that doesn't take effect until 27 years for a taxable gift of about $700,000 bucks, let's do this: Let's make a taxable gift of $700,000 and let's assume that there's a 6% return on that $700,000. What is it worth at the end of the 27-year term?
That $700,000 at the end of a 27-year term with a 6% tax-free yield would end up being in the $3 or $4 million range, so it will significantly outperform. If you assume that you want to put it on a tax-free basis and it's going to be a 4% return, make a gift of $700,000 to the children. The children invest it at 4% and what is it worth at the end of 27 years? About $2 million. Now, the “play” is if I die during the term. So, what you want to do with these techniques is to make sure you have a sufficiently long term that the grantor is expected to live, but hope that one of the grantors dies during the middle of the term. Then, the children are the winners.
There is a lot of discussion on this in the recent development material. There are several problems that are mentioned. Remember how I suggested that you look at Example 8 in the regs particularly carefully, because what happens in Example 8 is that it discusses reserving a right to revoke a spousal interest and approves it. But that revocation is reserved during the existence of the term — in other words, during my two-year term. The reservation of a right to revoke doesn't last during that entire 25-year term for Vicky.
So what happens is that if we followed Example 8 and you reserve the right to revoke Vicky's term, which would only expire at the end of my term, then we would be making a taxable gift to Vicky when I revoke that right. It will not qualify for the gift tax marital deduction because it's a terminable interest and I would have done what none of us want to do — -made a taxable gift to my spouse that will not qualify for the marital deduction.
So, the first issue you have is, is the Service correct that the duration of my right to revoke Vicky's interest can only last while I have an interest in the GRAT, or can it last during the entire term of the GRAT?
What the material sets forth is that there's no support for the Service taking that position. We need to remember that Dick was right on the Walton GRAT and if you use this technique, let's hope that Dick's right on this one, too.
The second issue you have is the reciprocal trust doctrine. The material also discusses whether the reciprocal trust doctrine would apply to this technique and the material concludes that there's a good case that it does. We believe the answer is “no”. And so you need to do your own research. And you need to be satisfied with that. A parallel GRAT could work where you have individuals who need funds and do not want to make an outright gift. You've got someone that's mortality disadvantaged and you want to take a shot at it, but keep in mind that you need to do your own research and make sure that you feel comfortable with that.
JEFF: First I want to address the question asked on Monday about discounting in a property-settlement context for built-in income tax liability. Mark Chorney????? pointed out to me that in Chapter 14, p. 65, f 283, he actually answers that question and says that most trial courts will not consider the built-in income tax liability because the taxable event lies so far into the future, that the discounting would be mere speculation. Both Carlin McCaffrey??? and Max Gutierres??? confirmed this answer based on their practice in this arena.
DENNIS: Carol's going to talk about the grantor trust rules, now.
CAROL: We have a question here about Sec. 675, sub 4 — the power to reacquire property held by a donor or another. The question is, “rumor has it that the IRS recently has refused to rule that the swap power held by a donor in a non-fiduciary capacity will not trigger a 2036 inclusion. Any comments?”
I have heard the same thing about the IRS not ruling on this issue. It is true that the Jordall????? power was held in a fiduciary capacity and one of the main concerns there, of course, was also Sec. 2042. The statute talks about it being held in a non-fiduciary capacity.
If that's a factual question, I think any drafter who's got half a brain can solve that problem. You just say that this is held in a non-fiduciary capacity. It seems to me that that would kind of resolve the question.
You never know quite how to take a refusal to rule, however. You know, the Service can rule positively; they can rule negatively. You rarely see negative rulings because once they tell you they're going to rule negatively, you say “okay, that's all right, I just won't have you rule.”
So their refusal to rule is often something else. Perhaps they may have something under study. They may not be willing to go on record one way or the other. It can mean they just don't want to aid and comfort the enemy. I've talked to people on the Service — some of whom of course are no longer there — and they feel that rulings are within their discretion. So I don't know how excited I get about their refusal to rule, other than knowing that I'm not going to be able to get whatever guarantee private letter rulings give me.
By the way, you should all know that private letter rulings can be revoked and they can be revoked retroactively to the extent that the taxpayer has not relied upon them. So they're not the guarantees that we sometimes think they are. They're not often revoked, but just a little trivia there.
The power to reacquire has been raised as to whether this is a Sec. 2036 issue, and there have also been some rumors bandied about that the Service may be considering the position that the power to reacquire property is a Sec. 2036 power. Well, I thought I knew what Sec. 2036 meant. Clearly, I don't anymore, but I do think that it is as a legal matter, not possible for that power to be a Sec. 2036 power. There is no economic benefit by being able to substitute property of equivalent value in my view. But I'm not the IRS nor the courts, so I can't definitively tell you that that's not going to happen someday. Certainly the law in the Sec. 2036 area is evolving and we're not quite sure where it's going to go.
It has been suggested that the power should be a Sec. 2036 power — to avoid any risk that the power could be held by another — because Sec. 675 says that a power held by any person to do certain things causes a trust to be a grantor trust. Sec. 675 gives a list of things, and one of the things is the power to reacquire. So it's been suggested that another person can hold a power to reacquire.
I have always been a little concerned about that. I realize that the “any person” language that precedes this would seem to indicate that I could ask any person on the street to be able to reacquire the property, and then, that should cause it to be a grantor trust. But somehow, that just doesn't seem logical and I'm not convinced that the “any person” used in that context could “reacquire.” “Reacquire” indicates to me that you had to have had it in the first place.
So, I've always been a little reluctant to rely on the “any person” language, despite the fact that there are at least two private letter rulings that say that it's okay. Remember, private letter rulings are issued only to the person who asked for them-you can't rely on them. And if the IRS decides to take a contrary position with you, you're out of luck.
Another power that people use is the power to borrow without adequate security or interest. I think that no one would reserve a power to borrow without interest. But you can certainly reserve a power to borrow without adequate security.
JEFF: I'm not the IRS or the government either but it seems to me that Carol's right about the Sec. 675 power to buy assets. When you talk about reacquiring for property of the same value, you're just talking about the ability to buy assets for their fair market value; I can't imagine that would successfully trigger Sec. 2036.
The next question asks whether you should include more than one option or can you get by with just one? This is an aspect of the grantor trust rules that hardly ever gets discussed.
For example, Prof. Donaldson on Tuesday didn't have time to dig into one additional element I want to mention — the “portion rules.” These are found in Sec. 671 of the regulations. They make you aware that different grantor trust provisions cause the grantor to be deemed to be the owner of different portions of the trust income. It's defined using the DNI carryout rules.
And so you have, for example, income that is allocable to corpus that is different from income that is available for immediate distribution. You may also have the ordinary income portion triggered, and it may be carried as it were to the grantor by one defect but not by another. And so it's important to think about what it is you're trying to accomplish when you're drafting a defective grantor trust.
By way of example, if you wanted capital gain and loss to be taxed out to the grantor, but you really didn't want the ordinary income on a cash flow basis to be taxed out, you would be looking for a defect that triggered the portion rule that applies to the income allocable to corpus.
By way of example — and I'm not sure this is the defect you would want because it would almost certainly cause estate tax inclusion — but 677A1 is the grantor trust rule that says you will be treated as the grantor if you've retained the right to receive the income, but the portion of the income you are taxed on is only the ordinary income that would get allocated to the income account for fiduciary accounting purposes and that would not get the gain out to the grantor. For that purpose, you'd need to trigger 677A2, which is the right to receive income accumulated for future distribution. That's really the income allocable to corpus.
The other aspect worth mentioning is that under the regulations, we use a DNI model and under that, I believe it is the case but there is scant ( WORD MISSING? ) on the notion that you can't ever get losses in excess of gains carried out.
That's part and parcel of the rule that would say in normal fiduciary income tax administration, losses get captured in the entity until 642H applies in the year of final termination. And so you need to be very careful to read the regulations here. Not only do they have these quirky little rules about what portion is triggered, but also they purport to apply a DNI model that I'm guessing many people are not paying attention to.
CAROL: You think we're not reading the regulations?
JEFF: I think some people are not reading the regulations, Carol.
CAROL: I don't think the government's read them since about '69.
JEFF: That's another aspect. The other fascinating thing here is that you've got to remember the grantor trust rules were crafted at a time when they punished the taxpayer and so they are extremely broad in their drafting. The fact that you could borrow from a trust 6753 and trigger the grantor trust rule is actually a great thing today when you want defective grantor trust treatment, but back then it was meant as a punishment.
Oftentimes, when you say, “gee, this language just makes no sense,” you've got to remember it was drafted in 1948 when the Service was trying to hurt taxpayers by codifying what would then be the Clifford regulations.
DENNIS: Jeff reminds us that there are good regulations and there are bad regulations. Let's now go to another GRAT question that asks, “What special provisions would you include in a grantor trust if the grantor is a corporate insider?”
This deals with an area that we need to be aware of, and we need to consult experts when we're drafting a GRAT or a charitable remainder trust using corporate securities and dealing with an insider. Section 16B of the Securities Exchange Act of 1934 provides for recovery of insider trading profits. What that means is that if I'm an insider with inside information, I can't buy stock today, sell it next week and make a profit. Similarly, if I'm an insider with inside information, I can't sell stock today and buy it next week, even though the profits would go to the corporation. When litigation is commenced over insider trading, plaintiffs' lawyers typically look at the corporate filings made with the SEC — under the last few years, plaintiffs' attorneys have picked up on some of the filings that dealt with GRATs.
And if you think, well how could that ever occur? Well, is the GRAT payment or is the GRAT substitution that Carol was talking about under Section 6754, a sale or a purchase of the stock? In other words, let's say we have a GRAT and I'm a corporate insider. I'm the trustee and the grantor trust provision in the GRAT gives me the power to reacquire in a non-fiduciary capacity. If what I do is put the stock in what it's selling for-say, 2 bucks a share — and there's going to be an IPO in the next six months and the stock skyrockets, then after it skyrockets what I would do is pull the stock out, freeze the value, and put cash in. And because it's a grantor trust, it will not be a taxable event and I'll lock in the sales.
Well, in 1997 the SEC issued a no-action letter called Peter J. Knight SEC No Action Letter, Federal Securities Law Reporter CCH para. 77,403. And the SEC stated that it would not take any action with respect to the transactions involving a GRAT because it's not really a sale or exchange.
Well, in 2003 there was a district court case in the state of Washington called Dreiling v. Jain, in which the court imposed a $247 million award as a result of many transactions involving a corporate insider taxpayer — one of which was a GRAT. Unfortunately, the case was settled. So what's left is a district court case that says there's an issue with respect to having the power to acquire and exercising that power with respect to insider trading. There's also another case like this that was settled — Morales v. Quintiles Transnational, 25 F. Supp. 2d 369.
I believe that a power to reacquire is problematic. In other words, the problem that the power that Carol was discussing earlier, 6754, is problematic when you're dealing with a corporate insider. What I would recommend is to use some other device or technique or power to make it a grantor trust.
One possibility is to provide that the annuity amount payable to the grantor would be payable from the income and to the extent income's not sufficient, from principal. There are two private letter rulings that would indicate that that would make that a grantor trust with respect to income and principal. The payment of the annuity amount to the grantor coming from income and principal.????????? Those private letter rulings are 94, 48, 018, and 94, 49, 013.
So if you're dealing with a corporate insider, watch the power that you use to make the GRAT a grantor trust. Watch who you're using as a trustee. I would recommend not using the corporate insider as the trustee during the GRAT term or while the GRAT is holding insider stock. You could run into the same problem with a charitable remainder trust. So keep that in mind.
JEFF: Right. We have several questions dealing with the application of the reciprocal trust doctrine. The one that seems to be of most concern these days is whether the reciprocal trust doctrine could apply outside the context of Sec. 2036 — it was historically created in the Grace case — but could it apply to a reciprocal irrevocable life insurance trust?
Let's imagine that a husband creates an islet??? with wife as trustee and the wife and two children as beneficiaries. At the same time, imagine that they didn't divorce the timing of the creation. That's one of the ways you could easily dodge the reciprocal trust doctrine. Let's imagine at the same time, the wife created an islet for the benefit of her husband and the two children, with her husband as the trustee. The question is whether the reciprocal trust doctrine would apply in that circumstance.
The question presented to us is that the husband and the wife as beneficiaries of these two trusts will have slightly different enjoyments, because each of them will receive distributions limited by an ascertainable standard. But in the trust the wife created for husband, there is a monetary cap on the amount he will receive.
For me, I don't think I could reach the conclusion that the monetary cap differential is sufficient to trigger the rule that says that if the trusts are not substantially identical, the reciprocal trust doctrine won't apply. I'm not sanguine whether that would be enough.
We saw a private letter ruling last year in which the government said that if the trust in one case contained a non-general power that the other trust lacked, that would be enough of a difference. That basically confirmed the Levy case from several years ago. I just don't know how much difference is enough but I think that if it were me, I'd want more of a disparity than just this particular cap on the amount that could be distributed.
Part of the question presented is whether a reciprocity to the reciprocal trust doctrine can apply in more than just Sec. 2036. I think we agree on the panel that the answer is probably “yes” but there's very little authority on that question as well. If this is a “done deal” and you've already got these facts and circumstances, one thing you could consider is having either or maybe both spouses resign as trustee.
That might still leave you with some uneasiness about a Sec. 2036 application because each spouse is also named as a beneficiary. Either or both also might consider disclaiming their interest as a beneficiary. That would likely have more tax liability. At least you'd need to value their right to receive distributions limited, by the ascertainable standard and in the one case, by the monetary cap.
I don't know whether I would go that extra mile if all you're worried about is Sec. 2042 — incidents of ownership in a fiduciary capacity. If these trusts were uncrossed, it would be treated as if the husband created the trust — he's the trustee who owns the policy on his life. There, you would worry about Sec. 2042 inclusion, so just getting out as trustee would be adequate in that circumstance.
DENNIS: It would be a question whether you would disclose that on a gift tax return to start the statute right. We've had that discussion before. Some do, some don't. It looks kind of funny when you do disclose it, so most generally don't, but you do have to worry about a statute issue as you go forward.
CAROL: I'm going to take a group of questions about family limited partnerships now.
On the issue of family limited partnerships, one question asked whether we would use them for future planning, and what should we do now with those existing partnerships.
On the issue of what do you do now, obviously you need to respect the partnership you've created. It's very important that the rules that you have set up are followed.
So if there are restrictions on liquidation and getting back the partnership assets, it looks really bad every time somebody wants something they get one out of the partnership,???????? a la Strangi. So please stop doing that.
On the issue of liquidation, obviously you need to have restrictions on liquidation to prevent the person who is dying from unwinding the whole partnership.
On the rest of this, I think the analysis is perhaps less legal and more strategic and the analogy is to the old joke about outrunning the bear: You don't have to be faster than the bear. You just have to be faster than the other guy who's running with you.
We're seeing the IRS litigate a lot of bad cases. Obviously the ideal situation is never to be audited because it's a lot less expensive for our clients not to have to argue, but if you are audited, you'd like the Service to see you as not being a prime prospect for being eaten by the bear.
So, we are counseling our clients today regardless of what we think the law is, that it would be in their best interests to give up holding any general partner units or being the managing member of an LLC. We've had several questions about whether they have to give them all up. Do they want a minority interest?
Frankly I don't see the point of having a minority general partnership interest. What our clients typically want is control and if they're not going to keep control, then there's no point in holding any of the GP units in a minority position typically. And, you do have a rule in Sec. 2036 that says if you can exercise these powers jointly with others, you're still in. So ideally you would counsel your clients to give up the GP units altogether and get out of the position of controlling distributions in particular.
We've been asked whether you can retain the ability to control investments; I think you should be able to retain those rights. But remember, if you can get your client to give up all control, it's better. I think the case law supports strongly the notion that controlling investments should not be a Sec. 2036 power.
Jeff has some thoughts about the lack of rigorous legal analysis that's going on. It's more of a “smell test” and when the Service starts looking at partnerships that are not bona fide in their view, they seem to lower the bar regarding “control”. So you've got to make some judgment calls with our clients and warn them of the risks.
I think that regarding limited partnership units, the issue is Sec. 2036A2, the ability to liquidate the partnership as a limited partner by voting with all of the other limited partners to close down the partnership.
I think that a lot of us feel that it's a stretch that the tax court at this point is not willing to go, other than in Strangi, which I described earlier as “gratuitous violence.” It doesn't mean though, that it's not a risk. We think as a practical matter it will mean that you could never give away anything and retain any corporate stock or anything else, without having an issue, because you could liquidate it. You'd never get a discount on anything. But you don't really want to have that philosophical debate I with an IRS agent when there's a great deal of money on the table.
So one thing you can do is get rid of all the units. Sell them. Give them away. If you need the economic benefit, sell them for a note to a grantor trust and keep that income stream of payments. Or, put it into a trust that is irrevocable but an incomplete gift by retaining a testamentary limited power of appointment; then you can keep the economic string of payments and with an independent trustee of that trust, you would not be voting those limited units.
I think that is a more complex analysis than most clients need, but again, if a great deal of money is on the table, I think that does eliminate that Sec. 2036A2 risk.
Dennis, you were going to tell us what you're doing in your practice for new clients who have not created these partnerships already. They're coming in to see you for estate planning advice. What are you telling them?
DENNIS: First, I agree with what you stated about control. But let's take the circumstance in which I have a client that is going to fund a limited partnership and is going to create a corporate general partner and will own 100% of the stock in the corporation. The client then decides to make gifts of limited partnership interest.
One question is whether a gifted partnership interest can be pulled back into the decedent's estate under Sec. 2036. My client has 100% control over the general, corporate general. He has a 99% limited partnership interest and then gifts interests in the limited partnership. What if he dies three years, four years, five years later? Can those gifts come back in? The answer is “yes”: Those gifts can come back into his estate under Section 2036 and under an implied agreement.
You have Cinda[???????], you have Abraham and you have a whole series of cases that state that if there's an implied agreement at the time it's created that the donee is going to give the money back, there's inclusion under Sec. 2026. Well what gives rise to that implied agreement? Well, if I put all my assets in there and I need the distributions from that entity and I'm making gifts away, then you have to think there may be an implied agreement. If what I do is put all those assets in there and the entity makes distributions to me for my needs and/or makes distributions to me to make gifts, I've got a problem.
Well let's assume that in this instance that what my client does is only put in 20% of his assets in the limited partnership. He still controls the general partner, makes gifts of the limited partnership interest and at death still holds 100% interest in that general partner. Are those gifts includable in the estate?
They shouldn't be, but I worry significantly about them. The reason why they shouldn't be included should be the same as if I create a corporation and give away shares of stock, and still hold control over the corporation. They should not be includable. But I would not take that risk, particularly when I'm dealing with financial assets in the entity, because of the ”smell test”. I worry that Sec. 2036 will run right into that and the courts will construe it as a transfer with a retained life estate.
So when I'm dealing with clients that have financial assets in a family limited partnership, they should never have control over the general partner at the date of death. Going back to my example, I've got a client that has a corporate general, 100% stock. He makes gifts. We're still going to get the discounts for gift tax purposes if it's structured properly. After a period of time, it would be appropriate for the client to get rid of control over the corporate general.
One way of doing that is to gift 50% of the stock to the surviving spouse in the corporate general. Is that sufficient? I would much rather the client have no interest in the corporate general. Otherwise, you're potentially running into Sec. 2036A1. But do I have to get rid of all my limited partnership interest to avoid Sec. 2036?
I think you may have to avoid Sec. 2036A2 and if Sec. 2036A2 is a significant risk, you may want to use a technique that Carol described on Monday in which I take my limited partnership interest and convey it to an irrevocable trust that is an incomplete gift, so that at my death, the trust assets are includable in my ?????? — not the limited partnership interest.
Now when my clients have businesses like active real estate in the limited partnership, I am less worried. You may ask whether it matters if I'm dealing with passive investments or active businesses, but I think that after reading the cases, it does matter. So when I'm dealing with clients that have business activities in there, I point out the risks but they're less inclined to get rid of everything. Part of it is that with passive investments, clients don't mind giving up control as much, compared with active businesses.
But Carol's point is well taken that this is an area that is constantly changing. Stay aware of what's going on in the tax court. So advise your clients about techniques and strategies to minimize risk, but be aware that there is still risk involved.
And Jeff, can we tell these people that there's really no opportunity cost. In other words, can I tell my client, to go ahead and do it and if he loses he's right back to where he started?
JEFF: The issue that Dennis is forecasting is Sec. 2043 and it's application or potential application here.
Most of us never studied Sec. 2043 in law or accounting school. When we teach the wealth transfer tax provisions and we get into the Strangi rules, transfer with retained enjoyment or retained power, Sec. 2036 is the relevant provision here. Historically, we've never encountered a situation in the FLP arena in which there's a transfer and retained enjoyment or control. That, at least, is the argument. But the transfer was made in exchange for consideration.
Here, we've transferred underlying assets in exchange for interests in the entity, and when the IRS says that the transfer fails the “adequate and full consideration” exception to Sec. 2036, we are left in the following situation. But The IRS has not yet been raising this issue.
In fact, last September, in Atlanta, during the IRS's annual CPE, I asked the government why it hasn't been pursuing the Sec. 2043 aspect of these cases. The response I got from the Service was basically that they never studied Sec. 2043 either.
The ugly reality is they're coming alive on this issue. Indeed, I had a call from one of their individuals who's charged with responsibility for prosecuting these cases who said he thought Sec. 2043 had a different application than it does. They were confusing the so-called “purge and credit” rule that is found in Sec. 2001B with the “consideration offset” rule in Sec. 2043 and had applied the purge and credit rule in a prior case.
Let me remind you of that rule. It would be applicable in the Abraham case: I've got an interest in a partnership that I transfer to my child and then it's brought back into my estate. The purge and credit rule says that the gift of the interest is eliminated from my tax base if the interest itself is brought back.
That is not our situation. In the situation in which I created — the FLP or LLC — I've transferred my good marketable assets and got back an interest in the entity.
Where Sec. 2036 applies in this circumstance is the following: First, at the date of my death, I own those partnership or LLC interests. They are an asset like my pocket watch or my automobile that is going to be includable probably under Sec. 2033. I own it and it's includable.
In addition, Sec. 2036 says that the transfer of the underlying assets, the marketable assets, for example, my dwelling, comes back into my estate because of my retained control or enjoyment of that property.
Now that represents double inclusion because they're treating me on one hand, as if I did the deal, and on the other hand, as if I didn't do the deal. They're bringing all the underlying assets back into my estate. This is not the purge and credit rule that would be triggered if my creation of the entity had been a gift.
The Sec. 2001B rule would say we eliminate the gift on creation — but we're not dealing with gift on creation. We're dealing with true double inclusion as if I did, and then didn't, do the deal. Congress understood the impropriety of that double inclusion through the Sec. 2043 consideration offset rule.
But what I'd like to do is give you a hypothetical that illustrates why this rule doesn't work right. It's this failure to work properly that's the substantial downside.
In fact, the impropriety of this rule is so tremendous that the government has said to me the result can be so ugly to the taxpayer that it worries about raising it, because when a court sees how bad it is, it might reject the Sec. 2036 argument at the front end.
And so I don't know whether the IRS is likely to raise this in the future. But here's the hypothetical. I start with 100 X of good marketable assets that I transfer into an entity, and I receive back interest or units that are worth 99 X. That is to say, I missed “full and adequate” consideration by just 1 percent.
Now, the full and adequate consideration exception is an all-or-nothing concept, so I've failed the exception that has been the root of so many of these cases we've been describing. At death, what's going to be includable in my estate is the value of the assets that I transferred and the value of the entity interest that I received back.
Now, let's imagine that nothing changes between date of creation and date of death. What's includable? The 100 of assets I transferred and the 99 of interest I received. I'll get a consideration offset for the 99 of assets that I received back, the interest and the entity. That puts me back in the position I would have been in had we done nothing at all.
And so in a close-to-death FLP situation, the Service really has no incentive to raise this issue. But let's change the facts and assume that I transfer assets worth 100 X and they double in value before I die so that in the entity, my assets are worth 200 X.
Imagine that if the inside assets double, my interest in the entity also doubles, so that at the date of my death, they are includable at 198 X. I've got 200 X includable because of Sec. 2036 and I have 198 X includable because I own the assets when I die.
The question is, how does the consideration offset rule operate? The answer is that the offset amount under Sec. 2043 is the value of the interest I got back, valued on the day we did the deal. That's 99 X. Which means that all the appreciation in the 99 X of interest that I received back is includable-that's not the same as if I had never done the deal.
Why does the rule operate this way? Sec. 2043 has been flawed this way from its inception, and the government has been made aware of the impropriety we've just described, but it refuses to change the rule. Why? It's not just because it's a “gotcha“ provision. It's because 2043 is an anti-tracing rule. Here's what I mean:
Let's say I create the entity. I take back the interest and then sometime before I die, I give it away. I sell it. Maybe I sell it and take the proceeds and invest it in something else, and that something else is includable in my estate. We don't want to have to have to figure out what it is that I own at death that represents these interests that I got back when I did the deal at the beginning.
Sec. 2033 and the rest of the inclusion rules apply to whatever I own at death, and then we make right the double inclusion with the consideration offset, and to make that rule work, we apply it using the value we know which is the 99 X at the time of creation of the entity.
And when you understand this result, it's the old FLPs that raise the greater risk than the new close-to-death ones, and the question is, will the government ever come alive on the Sec. 2043 application and stick a taxpayer with the way it's meant to be applied and produce a very substantial, in some cases, negative result to have gotten into the deal in the first place? The way to avoid it, of course, would be get rid of the FLP interest so that at death, you no longer have the Sec. 2033 inclusion that produces the inappropriate double inclusion.
JEFF: This is really ugly.
DENNIS: Continuing on with an FLP question. The state owns an FLP limited partnership interest. The FLP owns $1 million in municipal bonds with a pre def?????? basis of $1 million, 30% valuation discount accepted by the IRS on audit. Does Sec. 743/754 require a step-down in the inside basis? The answer is “no”, but let's remind us why we need a Sec. 754/743 election.
Let's assume that my brother and I create a partnership. We each put in $50,000, so the partnership has $100,000. The partnership buys a $100,000 piece of real estate. Ten years go by and the real estate's now worth $1 million. I die. My 50% interest has a net asset GAAP value of $500,000. Let's assume no discounts. The partnership's basis in that real estate is the $100,000 acquisition cost, less depreciation.
So what we have is the inside basis in the partnership that is significantly less than my basis, my estate's basis outside. So if the partnership made the sale, then we could potentially have gain flowing out, even though I've got a higher basis outside.
The way you get around that is to have in place a Sec. 754 election, which is an election whenever there's a transfer of an interest, sale or death. Not gift, but sale or death, that the partnership's basis is stepped up, the inside basis is stepped up with respect to the transferring partner.
So in the real estate example, if there's a 754 election in place, my estate's basis would be stepped up to my outside basis of $500,000. My brother's basis would remain the same and for depreciation purposes going forward and for sale purposes going forward, we'd have two different basis. Now that can work against you when you have basis going in and then you have a significant reduction in value, such as the example posed by this question.
Where you have a $1 million municipal bond with a $1 million basis and then the estate's basis becomes $700,000, if there's a Sec. 754 election in place, the partnership's inside basis would be reduced to $700,000. It would be $300,000 of income recognized at some point in the future. What's the answer? Not to have a Sec. 754 election in place. You could revoke a Sec. 754 election but the point to remember is that we've always made sure that there would be Sec. 754 elections in place — they need to be in place in the tax year during which the transfer occurs so you could make them in the tax return later. But you want to make sure whether you want to make a Sec. 754 election, so be careful and look at the basis before you see whether you want it in place or not.
CAROL: Well I have a series of GST questions.
I was asked about the Robertson CLAT that has no charity named as an annuity beneficiary. The trustee had the discretion to select charities every year and at the end of the time period, assume the CLAT passes to grandchildren. The question posed asks, since there isn't a charity that has an interest and no taxable termination at the end, why would there not be a direct skip going into the trust? The reason is that the trust has nobody with an interest. There's a special classification rule for that in Sec. 2612:
In this example, because there were beneficiaries who had a possibility of getting money — such as the charities — and charities in general are non-skipped persons even though they technically didn't have an interest, that would prevent a trust where no one had an interest, from being a skipped person. So that's why these was no direct skip going in.
There is also a question about adopting a non-relative. Generally, descendants of a parent of a transferor or the transferor's spouse or former spouse are assigned under the family rules There is also a part of the rule that applies to anyone married at any time to a descendent of a parent — a grandparent of the transferor or transferor spouse or former spouse — who are also classified in the same generation as their husband or wife.
The third part of that rule is that legal adoption is treated the same as blood.
You then have the assignment of people who are not assigned under those rules — we call them “non-family” rules and those are based on age. Age is irrelevant if you are assigned under the family rules age.
Let's say I adopt someone who is 40 years younger than I am. That person is unrelated to me in any way and would have been a skipped person. But if I adopt her, then because of the rule that says she is treated the same as a blood relative and becomes my child, she is now assigned under the family rules.
There isn't a dual classification because the family rules trump the non-family rules. There is a rule about dual assignments but it doesn't apply here because you can't have a family and non-family dual assignment. So as long as I legally adopt someone, that person becomes my child for this purpose. But you have to be careful if you're adopting an adult: In some states, the adoption may not be recognized.
In fact, there's an Illinois case in which a person adopted someone with whom he lived and had a personal relationship with. The adoption was not recognized and it's unclear to me whether that type of an adoption would be recognized for GST purposes
But in the less controversial situation, I think if you adopt a minor whom you're taking care of, or if you're adopting a non-related person (even an adult) with whom you have a family relationship with, that non-related person would be treated as your child.
There is also a glitch in the new rule about adopting family members that applies to descendents of a parent, not a grandparent. So there are still some family members that you could adopt, like a grandchild of your cousin, who wouldn't come under the family rule. They would get a dual assignment and you would not treat them as a child.
There was also a question about GST tax versus estate tax rates. Let's just run through it quickly. Rates are flat for federal purposes and with the enactment of state estate taxes or inheritance taxes, the GST tax may be the winner here, although there are some states like Illinois that impose a state GST tax.
There were many states that enacted the credit, which was a pick-up type credit for GST under Sec. 2604. That credit has been repealed like the state death tax credit has been repealed, but some states, like Illinois, have decoupled from that system.
So if you have a taxable termination or taxable distribution in Illinois that occurs at the same time, and as a result of an individual's death you pay not only a GST tax if you don't have an inclusion ratio of zero, but also you pay a 5% state generation skipping tax. So on one hand, that's probably going to be less than most state death taxes so I think GST is the winner on this call.
GST also has lateral transfers or upward transfers, so I can transfer from sister to brother to sister to mother and never have any tax. But if I subject something to estate tax, every time I do that, if everybody's got a general power, they're going to pay a federal estate tax. So GST has always been the winner as far as lateral or upward transfers.
With regard to creditors — if I give someone a general power of appointment versus a limited testamentary power of appointment, and he actually wants to exercise that power, GST here is the winner because I can use a special power and creditors cannot get at it. But in most states, in Illinois it's certainly true, if you have a general power of appointment and it's actually exercised, that property is subject to your creditors' claims. It's considered part of your estate. So you should be a little careful about that if you have clients with, shall we say, deadbeat children.
On the federal estate tax side of the equation, you still have a very generous and getting-better federal estate tax exemption. If somebody has a smaller estate which would otherwise be subject to generations that have been taxed, and he'll also get a chance to allocate that person's GST exemption.???????
You have to make some kind of a judgment call and that's why I think people will still continue to use the ability to give an independent trustee the power to give somewhat general power so you can delay making that decision. Remember as a drafter what I said on Monday: When I don't know what to do, I try to put this decision off on somebody else's shoulders, so I don't have to make it. And the theory is, of course, that he'll have more information at that time and be able to make a better call.
DENNIS: Well, but the key is flexibility.
CAROL: It really is. I think you want to put these decisions off because we've seen graphically what a fluid situation we're living in today in a way that we never dreamed of 20 or 30 years ago. We thought this system was stable. We were wrong. Things are changing faster than we can imagine and we have no idea what's going to be the system in a couple of years, let alone in five or 10 years.
When you think about the scale of these things, giving people general powers of appointment because some child who's 20 years old is going to be subject to a generation skipping tax when he dies at age 90, do you think this system is going to be here in 70 years? I don't think so.
JEFF: But on the other hand you may want to give the child that for control purposes.
CAROL: But of course we never did before. We always gave him at the most, broad limited powers of appointment. We didn't give him general powers. So I mean the scale of this is really quite amusing when you think about it.
I think the answer is, in your planning, when you're talking about things that are going to potentially happen a long time from now, if nothing else, realize that the system is probably going to be different and we can't predict what it is and so flexibility really is the key.
JEFF: And I think flexibility in this sense means somebody must have a power of appointment or a power to alter the terms of the trust.
Here's one question about decoupling. This posits that you're in a decoupled estate tax jurisdiction, such as New York. Do you see any problem giving an independent trustee the power post-mortem to increase the amount passing to the credit shelter trust? The notion would be that you could increase $1 million to $2 million based on whether you want to or do not want to pay state death tax at that time.
You may remember the government challenged a similar kind of authority that we came to know as an equalizer marital. It said that post-mortem, you would make some adjustments based on how you valued assets and determined the value of the estate of the surviving spouse by virtue of those post-mortem activities you didn't really know at the death of the decedent what it was the surviving spouse would receive.????? And the government said that meant that the interest of the surviving spouse was a non-deductible terminable interest.
The government ultimately lost that argument in the Smith case, the Lorin case, and the Meeskie case, but here, my guess would be the situation is different. It's different because the equalizer was based on a formula that was established and certain at the date of death.
Here, the amount passing to the spouse or to the credit shelter trust would be in the relatively unfettered discretion of the fiduciary, the executor, the administrator of the trustee, however it was being established. So my greatest concern here would be that if you did this, you wouldn't qualify for the marital deduction in the first place.
Now the way to get around this instead is through partial QTIP elections — whether it's a Clayton QTIP in which the non-elected property will flip out into the non-marital trust, or maybe you have partial elections with division of the QTIP-able trust and identical provisions going forward. I think we agree that it is the QTIP election that probably poses the better opportunity than discretion in the personal representative of the straight-out variety that's being described in this question.
CAROL: And in that vein, we had a question about the Clayton flip saying that it sometimes held that a special executor, an independent executor, must make that election when the flip is going to occur and as a practical matter how is that affected.
One way to get around it is to direct in the will that the executor must follow the direction of the trustees of the revocable trust in making any QTIP elections, and then in the revocable trust provide that the trustees can — if there isn't an independent trustee that they can appoint one for the purposes of making any — exercise any discretions applicable to an independent trustee which would include the QTIP election. Then, if an independent trustee is acting and a QTIP election is not made, then it would flip and we would cover both the QTIP election not being made for federal or state purposes.
We also have a question about the issue of the surviving spouse having the power, as opposed to an independent trustee and whether that's really necessary.
I don't believe that tax elections should ever result in some kind of a taxable gift or a Sec. 2036 power or anything else. I think if Congress wanted people to be able to exercise tax elections and have a tax result come of it, it should be able to say that in the statute. And in my view, property rights should never be determined for purposes of gift or estate tax until the elections have been made. Up until then it's not clear what those rights are.
Nonetheless, because a lot of dollars could be on the table and because we don't like to get into these arguments with people, at the moment that's the safer thing to do. I think the nice thing would be if the government would issue some kind of a revenue ruling that clears this up. The government has never raised this issue, ever. So I have to believe that the Service doesn't think it's much of a winner either.
DENNIS: A follow-up question to answer relates to the cross border practice of law with estate planners, particularly when discussing planning for clients who live in a decoupled state in which the draft person/lawyer is not registered.
WHOIS SPEAKING HERE¿??????????? Is Dennis saying that one should not draft documents for clients who reside in another state? And the second part is, what is the practice at his firm????????????
DENNIS: First, the law has not shaken itself out in the decoupled states as to how the applicable exclusion amount would get allocated and how discretionary funding will be handled. The concern that I have, therefore, is that I will be drafting a will for an individual in a decoupled state and although I may not get hit with a charge of the unauthorized practice of law, I may be creating a disastrous tax standpoint for my client because of the laws in the decoupled states.
I'll share with you two examples of where that came up in my practice. Recently I represented a large estate. It had three beneficiaries. One of the beneficiaries did not have an estate plan and asked me to do one for her. I prepared standard documents for her. Then her sister, who lived in another state, asked me to prepare the same documents for her.
I advised her to hire her own lawyer, and told her I would obtain her sister's permission to send my prepared documents in PDF file, so that her own lawyer wouldn't be starting from scratch. I sent the documents up to her own lawyer and when I spoke with him, he asked me whether I was aware that the state the sister lived in didn't have an applicable exclusion amount that was equivalent to the federal. I said “no.” So, what he was creating were three trust documents. If I were to just duplicate the documents for my client's sister, I could have been creating a state estate tax problem for her. So my concern is the malpractice risk of being unfamiliar with the laws of other states, not the unauthorized practice of law.
My second point relates to when I was trial counsel for a lawyer who had been sued for malpractice. The lawyer went from Virginia to West Virginia, met with a client, and created an estate plan that consisted of a family limited partnership and a charitable remainder trust, funded with marketable securities.
The client died. The family liked the family limited partnership and all the discounts they received, but didn't like the charitable remainder trust, even though both were created on the same day by the same lawyer.
The family relied on many things to challenge the charitable remainder trust, one of which was the unauthorized practice of law.
All the experts reached different conclusions based on the same set of facts. And I remember the judge stating in his ruling that there was no unauthorized practice of law because it's federal tax law.
So it's really not the unauthorized practice of law to be concerned with — it's the malpractice exposure that exists because of not being familiar with the rules in the decoupled state.
CAROL: We have a question about the GST and estate tax exemptions being the same: should you still draft for three separate trusts or do you assume that the bypass trust will be the same as the GST trust, and rely on severance language and if circumstances change.
I think you can do that, and certainly in smaller situations it's a perfectly workable way. But I don't do it that way. There are a number of reasons why I like to spell out the split, one of which has to do with a nuance in the regulations under Sec. 2654 that says that if you don't have direction in the document to sever, that the terms of the severed trusts in the aggregate have to be the same. I've always been a little unsure of exactly what that means and so I'm just very conservative on this point.
As far as the mismatch, let's say I have a $5 million estate (disregard the estate death taxes for a second.) If I have a $2 million exemption, my family trust would be $2 million; my marital trust would be $3 million.
Now what's my GST answer? Well, if I made no taxable gifts and allocated no GST exemption, my family trust is fully exempt. My marital trust is not.
But that doesn't happen in my world very often, so let's assume that the person has used some gift tax exemption of $500,000 and he hasn't used any GST exemption. Well, that is a familiar pattern. We'd have a family trust then of $1.5 million, a marital trust of $3.5 million, and we would allocate just the exemption of the family trust. We'd have a half a million left over, so we'd do our traditional reverse QTIP election of half a million dollars. It would be in a separate reverse QTIP trust and again you can get there by a severance or you can get there by a direction to split it.
But what about the last situation which we frankly ignored for many years, which is that you can use GST exemption without using any gift tax annual exclusion. You could be allocating to a CRUMMY withdrawal trust. Let's assume here that somebody's been making annual exclusion gifts to a GST dynasty trust and he's used a half a million dollars because he has several children and grandchildren and he's been doing this for a long time, but never used any gift tax exclusion.
Now, the family trust is $2 million but I don't have enough GST exemption to cover it. I only have $1.5 million. So you really have to — and we always should have probably — allowed for that possibility that it might be the family trust that's getting split up, not the marital trust. Of course you don't have to make any reverse elections there. It works just fine. And there are other ways, of course, today that we use a GST exemption that we might not have done before with the downstream split where we could allocate to the trust following a GRAT, for example, and we might use it up that way.
So I think in our drafting now, we are allowing for the possibility that we don't know which of those trusts is going to be split up. Also, because of the state death tax credit, we have an interim contingent marital trust, estate marital trust — whatever you want to call it. We really have a possibility of four trusts because it's just more complicated that way.
Another question we received involves a couple ???? estate. Assume that the will of the first spouse to die provides the old-fashioned more traditional formula, which says the marital share is the smallest amount necessary to reduce the federal estate tax to zero, taking into account the state death tax credit, but only to the extent the state death tax is not thereby increased.
Let's assume that the first spouse dies in 2006. The question is, should the direction not to increase the state death tax credit be read instead to apply to the state tax deduction? And that raises some interesting issues not only in terms of how you apply this but what you want to do in your drafting.
If you have a taxable estate in which your state exemption is the same as your federal and no one is holding property in other states, then there's no problem. But in the more complex situation where you may have a mismatch, you have an issue here.
One issue is, do you just ignore that language? Probably, yes, because the state death tax credit is gone. You can't take into account a credit that doesn't exist anymore and you would have a $2 million family trust. That might not be what you want to do, however, you could have a few different situations.
For example, assume that your state death tax is not based on the federal taxable estate net of the deduction. Some states are. Some states are going to let you deduct their own taxes in determining the taxable estate which is an inter-related computation, in which case I think you'd get back to the $2 million answer again. But assume that when the state imposes a tax, it adds on the taxable estate plus whatever deduction that it would be for their own taxes.
So, assume that at a $2 million level, you would pay $99,000 in state death taxes, but if you bump it up to $2,099,000, you pay another $10,000 in taxes. And so the question is, what do you do? Would you bump it up to $2.1 million because you're going to be subtracting off those state taxes and get down for federal purposes back to your $2 million? But that will actually increase state death taxes by some amount.
And the answer is, change your formulas. Create an interim trust and push this decision off down the road when more facts are known and people can decide about the relative tax rates. Things are changing. Even with the prospect of repeal, in essence gone, if rates drop to 15%, it's still not clear. And because exemptions are also increasing, you have to consider that even if a person died right now, he'd pay a tax. In two or three years when the surviving spouse dies, the exemption could be huge. So paying a tax now is a much tougher call even if it's relatively low rates.
DENNIS: The general rule is always defer, defer, defer.
And the estate may eliminate it. The individual may die in a state that doesn't….
CAROL: They might use it all up. I mean if I had $100 million estate, I'm not going to assume that the surviving spouse is going to get $100 million exemption. I don't think the repeal is going to occur, at least in the near term, and that's a little easier decision because now you're just deciding do you want to do, which rates do you want to play at. But I think in the more typical situation with estates up to $10 million, it's much harder to decide So wait to get more information and push that decision off.
DENNIS: One thing we're seeing is with clients with $50 million and up, they're giving strong consideration to moving to a state that doesn't have state estate taxes.
CAROL: Well, that's correct, and Florida is becoming a very popular residence.
JEFF: Let's answer some questions relating to fiduciary selection. First, in a Q pert???????????, could an adult child be a co-trustee? I think it's harmless and indeed, he could be sole trustee. Could the spouse be trustee? I think that answer is also “yes.”
A related question concerns a parent creating a trust for the benefit of descendants. The trust will not own insurance, so there's no issue about incidence of ownership in a fiduciary capacity. Can the settlor, the parent in this case, be trustee? Sure, if you're careful in your drafting.
The authority for what constitutes careful drafting is at least half a century old. In Sec. 204l, naming a beneficiary as trustee causes you to use an ascertainable standard and we're pretty familiar with the health education maintenance support standard for ascertainable Sec. 2041 standard purposes. The case law under Sec. 2036 is even more generous. That's surprising because it ought to be easier to trigger Sec. 2036 than Sec. 2041 but it's the opposite.
My problem is that case law is so old that I'm not sure that if it were me, I would retain the powers trustee even collared???????? with the standard of a trust that I expected to be excluded from my estate at death.
What about naming a beneficiary as trustee? There, you certainly want to use the ascertainable standard the way you would if the settlor was the trustee. In addition, in both of these circumstances, you probably want to preclude the government from raising what I believe is a wrong argument, but is sprinkled all throughout the regulations known as the “discharge of obligation of support” theory.
The standards that we're talking about typically protects distributions to myself, but the government says that if I have the power to make distributions to my dependent child, that is an indirect benefit to me and the ascertainable standard exception does not protect me there. It only protects distributions to me.
So, in addition, you want to add a prohibition that says you cannot make any distribution that would satisfy your legal obligation of support. Finally, I think that you want to have a recognition that if I'm trustee of a trust that pays me income for life and I make a distribution to a third party, the government's position is that I'm making a gift of my income interest in the corpus I distribute. So the way to avoid that is with the ascertainable standard.
DENNIS: The last question. I have a Florida resident that has property in New York in a Q-pert.????? If the settlor dies before the term ends, the New York property reverts to the settlor's estate. That would cause a New York estate tax. Can a Q-pert????? form a single member, limited liability company to earn the real estate? That way, if the settlor dies, an LLC interest will revert to the settlor, avoiding the New York tax. Call a New York lawyer and ask. I expect that the answer is “maybe,” but I do recommend that you call a New York lawyer.
Carol, Jeff and I thank you for paying attention. We thank you for your questions and I thank Carol and Jeff.
CAROL: Thank you.
HOWARD: Okay, we are now going to take the lunch break. Please remember to take all your personal effects. The advisory board has gleaning rights and will keep anything they find that they want. Also you may now turn on your PDAs and cell phones.
[END OF QUESTION & ANSWER SESSION]