Written by the house from every document he has on this theme. It carries no figures of its own by design: the sources carry the numbers and the live registers carry the current ones.
The marital deduction for property in trust is a bargain with deferred payment, and every term matters. When a QTIP election is made under Sec. 2056(b)(7), the estate defers tax at the first death in exchange for including the property in the surviving spouse’s estate at the second death. That deferral comes with mechanical requirements — an affirmative election, a qualifying income interest, inclusion later — and the tax code enforces every one of them. Missing the election collapses the entire structure, as The Marital Deduction a Widow Almost Lost to a Checkbox shows. Making the election but terminating the trust early triggers gift tax in unexpected places, as The Agreement That Cost Two Children [figure withheld: see the cited source and the live register]Million Each demonstrates. The bargain holds, but only if you honor its exact terms.
A QTIP termination is a taxable event even when everyone agrees. Sec. 2519 treats the surviving spouse’s disposition of a qualifying income interest as a deemed transfer of the remainder, and Sec. 2207A creates a reimbursement right for any resulting gift tax. When the McDougall family terminated Clotilde’s QTIP trust and gave all [figure withheld: see the cited source and the live register] to Bruce outright, the children were the donors — they surrendered remainder interests worth [figure withheld: see the cited source and the live register] each, not Bruce, who was already treated as owning the property under the QTIP rules. The Agreement That Cost Two Children [figure withheld: see the cited source and the live register]Million Each makes plain that unanimous family consent does not eliminate the tax consequences; it just determines who bears them.
Valuation of trust interests turns on state law and the governing instrument, not federal tables. The IRS argued in McDougall that Sec. 7520 actuarial tables must control the value of the children’s remainder interests. Judge Halpern held otherwise: Sec. 7520 says values “shall be determined” under the tables “for purposes of this title,” and a trustee dividing assets on termination under state law is not making a determination for purposes of the Internal Revenue Code. The court anchored this in Morgan v. Commissioner: state law creates the interests, federal law decides which ones get taxed. A trustee can consult the tables but is not bound by them. The Agreement That Cost Two Children [figure withheld: see the cited source and the live register]Million Each establishes that Washington state law and Clotilde’s will controlled what her children were entitled to, and those sources — not a federal actuarial shortcut — drove the [figure withheld: see the cited source and the live register] valuation.
A missed QTIP election can be fixed, but the fix is discretionary relief, not a right. The Marital Deduction a Widow Almost Lost to a Checkbox reports that the IRS granted 9100 relief to an executor who failed to check the QTIP election box, allowing a late election because the mistake was made in good faith reliance on advisors and did not prejudice the government. But 9100 relief costs the price of a private letter ruling, takes months, and depends on facts that stay clean. It is a rescue, not a plan.
The protective election is the fiduciary discipline. Both sources converge on this: when there is doubt about whether a trust needs a QTIP election, make it. The Marital Deduction a Widow Almost Lost to a Checkbox says the cost of an unnecessary election is close to nothing and the cost of a missing one is the whole deduction. The Agreement That Cost Two Children [figure withheld: see the cited source and the live register]Million Each implies the same principle for trust modifications: the time to price a remainder interest and map the tax consequences is before the family signs, not after. A fiduciary’s job is to close the gap before it opens.
He starts with the statute’s exact language and traces what it obligates. The marital deduction sections are not analyzed as policy or history but as a set of mechanical trades: defer now, include later, elect affirmatively, satisfy each condition. When the IRS or a taxpayer tries to simplify — tables must govern, everyone agreed so nothing transferred, the election was implied — he goes back to what the Code actually requires and what it makes conditional. He treats statutory language as a set of if-then gates, and he assumes each one enforces.
He reads cases for their legal holdings, not their equities. McDougall is not presented as a family that made a mistake; it is presented as a ruling on what Sec. 2519 treats as a transfer, who Sec. 2207A makes liable, and whether Sec. 7520 binds a trustee acting under state law. The sympathetic facts — everyone consented, no one evaded — get acknowledged as a counterargument, but they do not soften the analysis. The statute did what it said it would do.
He values procedure as substance. The checkbox on a Form 706 is not treated as a technicality that a reasonable IRS should waive; it is treated as the condition on which a multi-million-dollar deduction depends. The protective election is not overcautious paperwork; it is the move that eliminates the risk that a discretionary rescue will be necessary. He consistently elevates the boring mechanical step — filing timely, checking the box, getting the valuation before signing — to the level of fiduciary duty, because that is the step that prevents the seven-figure problem.
He refuses to let agreement substitute for arithmetic. The McDougall family unanimously consented to terminate the trust, and he acknowledges that families have legitimate reasons to do so. But unanimous consent does not make the remainder interests worthless, does not eliminate the deemed transfer under Sec. 2519, and does not waive the children’s exposure. He separates what people want from what the tax code does to them when they get it.
He treats state law as the input and federal tax law as the function that acts on it. Morgan v. Commissioner is quoted for exactly this principle: state law creates the interests, the Internal Revenue Code taxes them. A Washington trust termination is governed by Washington law and the terms of Clotilde’s will; the Sec. 7520 tables are federal and advisory. A trustee’s duty is defined by state fiduciary law; the tax consequences follow. He does not merge the two systems; he keeps them stacked and checks each in order.
The gap between what families intend and what the tax code does to them. In The Marital Deduction a Widow Almost Lost to a Checkbox, a widow hired professionals and thought she was protected; a missed checkbox nearly cost her the marital deduction. In The Agreement That Cost Two Children [figure withheld: see the cited source and the live register]Million Each, a family agreed to simplify and thought unanimous consent meant no consequences; the children each made [figure withheld: see the cited source and the live register] gifts. Both pieces circle the same idea: the estate and gift tax system does not care what you meant or that everyone agreed. It cares what you filed, what you elected, and what transfers the statute deems to have occurred. The unwritten piece is the one that walks through a typical family estate plan — QTIP trust, bypass trust, powers of appointment — and shows, step by step, every place where good intentions produce taxable events the family never predicted.
The protective move that costs almost nothing and forecloses the expensive problem. The Marital Deduction a Widow Almost Lost to a Checkbox says make the protective QTIP election when there is any doubt. The Agreement That Cost Two Children [figure withheld: see the cited source and the live register]Million Each says price the remainder interest and map the tax consequences before the family signs the termination agreement. Both are the same advice applied to different instruments: do the defensive thing early. The unwritten piece is a checklist or decision tree for executors and trustees — not general principles, but the actual if-then: if the trust could be a QTIP, elect; if the termination could trigger Sec. 2519, value the interests first; if the power of appointment is limited, check whether it reduces fair market value under state law. It would be boring and it would save families.
The distinction between discretionary relief and reliable planning. He has now written twice about IRS relief that worked — the 9100 ruling in The Marital Deduction a Widow Almost Lost to a Checkbox — and tax consequences that could not be undone in The Agreement That Cost Two Children [figure withheld: see the cited source and the live register]Million Each. The tension is that one mistake was fixable and one was not, and the difference was not the size of the error but the nature of the election. A QTIP election is something the Code explicitly allows to be made late under the right circumstances. A Sec. 2519 transfer is a deemed gift the moment it occurs. The unwritten piece is the taxonomy: which estate and gift tax mistakes are eligible for 9100 relief, which ones can be cured by amended return, and which ones are final the day they happen.
None found.
Whether the McDougall reasoning on Sec. 7520 will hold in other circuits or survive IRS challenge in different contexts. Judge Halpern held that the actuarial tables do not bind a trustee valuing interests for distribution under state law, because the trustee is not making a determination “for purposes of this title.” That is a clean statutory reading, but it is also a memorandum opinion from the Tax Court, and the IRS had argued the opposite. The question is whether this holding will be followed when the tables would produce a dramatically different result, or when the taxpayer is using the state-law valuation to reduce a federal tax.
What the children in McDougall actually understood when they signed. The piece says they spent four years litigating whether their gift was worth [figure withheld: see the cited source and the live register]or [figure withheld: see the cited source and the live register], which means whatever they understood in 2016, they did not understand that. But there is no detail on what they were told before signing, whether they received independent tax advice, or whether the termination agreement itself disclosed the potential gift tax exposure. That gap matters for the fiduciary question: was this a failure to explain or a failure to hear? The record is not in the piece.
How far the Sec. 2207A reimbursement offset extends. The Tax Court in McDougall held that what the children gave up was their distribution net of the avoided Sec. 2207A(b) liability — cutting roughly [figure withheld: see the cited source and the live register] off each gift. That is a straightforward application of the statute, but it assumes the gift tax would have been paid and the reimbursement claim would have been enforceable. If Bruce had insufficient liquidity, or if the children had structured the transfer to avoid triggering Sec. 2207A, would the offset still apply? The piece does not say, and the opinion apparently did not reach it.
Whether protective QTIP elections carry any downside he has not yet written. The Marital Deduction a Widow Almost Lost to a Checkbox says the cost of an unnecessary election is close to nothing. What it does not say is whether that is true in every case — whether there are fact patterns where electing QTIP treatment when it was not needed creates a problem later, at the surviving spouse’s death or in trust administration. If the cost is truly zero, the advice is simple. If there are edge cases where it is not, they belong in the analysis.
How the nonjudicial settlement agreement statutes differ across states in ways that change the tax outcome. The Agreement That Cost Two Children [figure withheld: see the cited source and the live register]Million Each says Washington state law and the Uniform Trust Code drive what a beneficiary is entitled to on early termination, and that North Carolina, Delaware, South Dakota and Tennessee each enact it differently. But the piece does not map those differences or say which states would have produced a different result in McDougall. That is the practical question for a trustee deciding whether to allow a termination: does the state I am in treat this remainder as something or as nothing?
This is reasoning about the mechanical operation of the marital deduction and the tax consequences of trust modifications. It is not advice on how to structure a marital trust in the first place, when to use a QTIP versus an outright bequest, or how to draft powers of appointment to preserve flexibility without creating Sec. 2519 exposure. It is not guidance on how to calculate the Sec. 7520 tables when they do apply, or how to prepare a Form 709 when a deemed gift has occurred. It is not analysis of portability, bypass trusts, or the interaction between the marital deduction and the estate tax exemption.
A reader would still need counsel to draft the trust instrument, to determine whether a particular trust qualifies for QTIP treatment, to decide whether a nonjudicial settlement agreement is advisable in a specific family situation, and to prepare the private letter ruling request if 9100 relief becomes necessary. The writing explains what the statutes do and how courts have applied them. It does not replace the lawyer who applies them to a client’s facts.