Connelly: The Life Insurance That Raised the Price It Was Buying

What happened

Two brothers signed an agreement so their company would stay in the family, with the company obliged to redeem a deceased owner's shares. When Michael died in 2013 holding 77.18%, his brother declined to buy personally, so the company's obligation was triggered, and rather than commission the appraisal the agreement called for they settled on a number. The estate reported the shares at $3 million, and on audit the estate's own accountants valued the company at $3.86 million. The IRS then counted the life insurance proceeds sitting inside the company as an asset, making it worth $6.86 million and Michael's stake $5.3 million, and both the district court and the Eighth Circuit agreed.

The case, start to finish

Two clauses, each sensible, written by different people, quietly pointing at each other.

Two brothers and a document written years early

Crown C Supply was a building supply distributor in St. Louis owned by two brothers. Michael Connelly held 77.18%, or 385.9 of 500 shares; Thomas held 22.82%. There was no market for the stock, the ordinary situation for a closely held company and the reason buy-sell agreements exist.

So they wrote one, and it did what such documents do. On the death of one brother the survivor could buy his shares; if he declined, Crown had to redeem them, at a price set by an outside appraisal of its fair market value. To fund that, Crown bought $3.5 million of life insurance on each brother and paid the premiums.

Every part of that is standard, and the reason the company owned the policies is a good one: because Crown paid, they were certain to stay in force and the money certain to be there. It was the sound operational choice, and the expensive tax choice, and nobody weighed the two together before signing.

The shortcut at the worst moment

Michael died in 2013. Thomas declined to buy the shares personally, triggering Crown's obligation to redeem. Rather than commission the appraisal the agreement called for, Thomas and Michael's son agreed, in the Supreme Court's own description, in an amicable and expeditious manner that the shares were worth $3 million. Crown paid it from the insurance proceeds.

It is easy to see why: a family has just lost someone, the two left are not adversaries, the money is sitting there, and a formal valuation feels like an expense and a delay.

The estate reported the shares at $3 million. The IRS audited. The estate's own accounting firm valued Crown at $3.86 million, deliberately excluding the $3 million of insurance proceeds on the theory that the obligation to pay them out canceled them. The IRS counted them: Crown was worth $6.86 million, so Michael's 77.18% was worth $5.3 million, not $3 million. Additional estate tax, $889,914. Section 2703 means an agreement's stated price is ordinarily not binding on the tax authority anyway, so skipping the appraisal saved a fee and settled nothing.

The loop nobody solved

On June 6, 2024 the Supreme Court affirmed, 9 to 0: a company's obligation to redeem shares at fair market value is not a liability that reduces what those shares are worth. It added the sentence the case turns on: this is simply a consequence of how the Connelly brothers chose to structure their agreement.

The fault is arithmetic. A buy-sell agreement has two halves, usually drafted by different people: a formula that sets the price, and a plan that funds it. Here the formula was fair market value by appraisal and the funding was company-owned life insurance. Each is defensible alone. Together they form a loop, because money paid into the company is part of its fair market value, so the funding raises the price it funds.

Solve it. For company-owned insurance alone to fund a redemption at fair market value, the proceeds P must satisfy P at least 0.7718 times ($3.86 million plus P). That is about $13.1 million, roughly $3.38 of cover for every $1 of the company's other value. Crown carried $3.5 million. Each extra dollar of cover bought about 23 cents of progress, because 77 cents went straight back into the number being chased, and a majority holding makes the loop far worse than a 50/50 split.

Solve the two halves together

Nothing here was a fraud, a scheme or a slip of execution. The insurance did the job it was bought for: the money was there, the family kept the company, and Thomas became sole shareholder without borrowing. It was an ordinary, competently drafted, widely used arrangement that produced a bill nobody planned for, and the damage was done at signing, by two clauses each fine on their own.

The alternative the Court itself named is a cross-purchase, where each owner holds the policy on the other and buys the shares directly. The proceeds go to the survivor and never touch the company's balance sheet, so the price of the shares is unchanged by the money that buys them. It is not free: each owner pays the premium on the other's life, awkward at 77/23 because the small owner carries the big policy; the count grows fast beyond two owners; and moving an existing policy has traps.

Weigh those with your advisers, before signing rather than at a funeral. A buy-sell agreement is a price formula and a funding plan, and they must be solved together, because the funding is an input to the price. Do the arithmetic backwards from the outcome, then follow the valuation step you wrote down.

Timeline

• Before 2013 Michael Connelly owns 77.18% of Crown C Supply, or 385.9 of 500 shares. Thomas owns the other 22.82%. They sign an agreement so the company stays in the family: when one dies, the survivor may buy his shares, and if he declines, Crown itself must redeem them. The price is to be set by an outside appraisal of Crown’s fair market value. To fund it, Crown buys $3.5 million of life insurance on each brother and pays the premiums.
• 2013 Michael dies. Thomas declines to buy the shares personally, so Crown’s obligation to redeem is triggered. Rather than commission the appraisal the agreement called for, Thomas and Michael’s son agree, in the opinion’s words, in an "amicable and expeditious manner" that the shares are worth $3 million. Crown pays that from the insurance proceeds.
• Filing and audit The estate reports Michael’s shares at $3 million. The IRS audits. The estate’s own accounting firm then values Crown at $3.86 million, a figure that deliberately excludes the $3 million of insurance proceeds, on the theory that the obligation to pay them out cancelled them.
• The assessment The IRS counts the proceeds. Crown was worth $6.86 million ($3.86M + $3M); Michael’s 77.18% was therefore worth $5.3 million, not $3 million. Additional estate tax: $889,914. The estate pays it and sues for a refund.
• 2021–2023 The district court grants summary judgment to the government. The Eighth Circuit affirms.
• June 6, 2024 The Supreme Court affirms, 9–0. A company’s obligation to redeem shares at fair market value is not a liability that reduces what those shares are worth. The Court adds the sentence the whole case turns on: this "is simply a consequence of how the Connelly brothers chose to structure their agreement."

You're in the owner's chair

You and your co-owner want the survivor to keep the company when one of you dies, and you intend to fund the buyout with life insurance. The lawyer asks a question that sounds administrative: who should own the policies? What do you say?

• Keep the company-owned policies, but write a fixed price for the shares into the agreement so the number cannot move
• Each of us personally owns a policy on the other and buys the shares directly
• The company owns both policies, pays both premiums, and redeems the deceased owner’s shares

One 77.18% stake, four different dollar figures

• What the family agreed the shares were worth: 3 $ millions
• Life insurance Crown actually carried on Michael: 3.5 $ millions
• What the IRS assessed, upheld 9–0: 5.3 $ millions
• Coverage a fair-market-value redemption really needed: 13.1 $ millions

The first three figures are stated in the Supreme Court’s opinion. The fourth is arithmetic from the Court’s own numbers, solving P ≥ 0.7718 × ($3.86M + P), and it holds because an appraisal of fair market value counts money the company holds, which is precisely what the Court decided.

Business model

A small, successful, closely held building supply distributor with two shareholders and no market for its stock. That is the setting for most buy-sell agreements: an asset the owners cannot sell to a stranger, a family that wants continuity, and a document written years before anyone needs it.

Revenue model

The opinion says almost nothing about how Crown made money, and that absence is the point. The estate tax did not attach to the trading business. It attached to a $3 million balance-sheet item that arrived on the day of death, and no amount of operating performance would have changed the answer.

Cost structure

The only cost that decided this case was the insurance premium, paid by Crown, on policies owned by Crown, naming Crown as beneficiary. The Court noted why that structure is attractive: because the company paid, the policies were certain to stay in force and the money was certain to be there. It was the sound operational choice and it was the expensive tax choice, and those two facts have to be weighed together, once, before signing.

Strategic challenge

A buy-sell agreement has two halves that are usually written by different people: a formula that sets the price, and a plan that funds it. Here the formula was "fair market value by outside appraisal" and the funding was "company-owned life insurance." Each is defensible alone. Together they form a loop, because money paid into the company is part of the company’s fair market value, so the funding raises the price it is funding.

Key decision

Two decisions, years apart, and the first one caused the second to fail. The brothers chose to have Crown own the policies rather than owning policies on each other. Then, on Michael’s death, they skipped the appraisal their own agreement required and settled on $3 million by agreement. Section 2703 of the tax code means an agreement’s price is ordinarily not binding on the IRS anyway, so the shortcut saved a valuation fee and cost $889,914.

What worked

The insurance did the job it was bought for: the money was there, the family kept the company, and Thomas became sole shareholder without borrowing. It is worth being clear that this was not a fraud, a scheme, or a mistake in execution. It was an ordinary, competently drafted, widely used arrangement, and it produced a bill the family had not planned for. There is no villain and no slip of execution here, which is exactly why it is worth studying: the damage was done at signing, by two clauses that were each fine on their own.

What failed

The arithmetic nobody ran before signing. Take the Court’s own figures: Crown’s other value was $3.86 million and Michael held 77.18%. For company-owned insurance alone to fund a redemption at true fair market value, the proceeds P must satisfy P ≥ 0.7718 × ($3.86M + P). Solve it: P is about $13.1 million, roughly $3.38 of coverage for every $1 of the company’s other value. Crown carried $3.5 million. The agreement’s own formula demanded a policy nearly four times the one that existed, and every extra dollar of cover bought only about 23 cents of progress toward the price, because 77 cents of it went straight back into the number being chased.

Risk factors

Company-owned policies inflating the very stake they fund; a majority holder, whose large percentage makes the loop far worse than a 50/50 split would; skipping a valuation step the agreement itself specifies; assuming a price written into a family agreement binds the tax authority; and a redemption structure that concentrates the whole exposure in the estate of whoever dies first, at a moment when nobody is in a mood to argue about method.

Lesson summary

A buy-sell agreement is a price formula and a funding plan, and they must be solved together, because the funding is an input to the price. Company-owned insurance is the convenient structure and the one that feeds the loop; a cross-purchase, where each owner holds the policy on the other, keeps the money off the company’s balance sheet. The Supreme Court itself named it as the alternative, and named its costs too. Whichever you pick, do the arithmetic backwards from the outcome before anyone signs, and then actually follow the valuation step you wrote down.

Key data

• 77.18% (385.9 of 500 shares) Michael’s stake in Crown C Supply
• $3.5 million Life insurance Crown carried on each brother
• $3 million Paid to redeem Michael’s shares
• $3.86 million Crown’s other value (estate’s appraiser)
• $6.86 million Crown’s value per the IRS
• $5.3 million Michael’s shares per the IRS
• $889,914 Additional estate tax
• ~$13.1 million Coverage the agreement’s own formula required
• Affirmed 9–0, June 6, 2024 Supreme Court

Sources & basis

The company here is real and named, and nothing about it was invented to make the story land. The list below is where each fact came from — public filings, court records, published reporting — so you can open a source and check it against the sentence that used it.

1. Connelly v. United States, No. 23-146, 602 U.S. ___ (2024) — slip opinion, Supreme Court of the United States (Thomas, J., for a unanimous Court), decided June 6, 2024: the 77.18%/22.82% split of 500 shares, the $3.5 million of company-owned life insurance on each brother, the $3 million redemption, the $3.86 million and $6.86 million valuations, the $5.3 million value of Michael's shares, the $889,914 deficiency, and the Court's discussion of the cross-purchase alternative View source ↗
2. Connelly v. United States — full opinion text, Legal Information Institute, Cornell Law School View source ↗
3. The ~$13.1 million coverage figure is our own arithmetic on the Court's stated numbers, not a figure from the opinion