Buy-Sell Agreements After Connelly and OBBBA: Who Needs to Restructure — and Who Just Needs Better Drafting

In June 2024, the Supreme Court decided Connelly v. United States and upended a funding structure thousands of closely-held businesses use: company-owned life insurance backing a redemption buy-sell. Thirteen months later, Congress passed the One Big Beautiful Bill Act and permanently raised the federal estate-tax exemption to $15 million per person — which quietly changed who actually has to care about Connelly.

Most of what you'll find online about Connelly was written in mid-2024, still cites the old exemption numbers, and stops at “consider a cross-purchase.” This piece is the 2026 version for California business owners: what the case held, who the problem still bites after OBBBA, the California corporate-law wrinkle almost nobody mentions, and the drafting fixes that matter even if the estate tax never touches you. It's orientation, not tax or estate-planning advice — the restructuring decisions here get made with your CPA and estate counsel at the table.

What Connelly held#

Two brothers, Michael and Thomas Connelly, owned Crown C Supply, a Missouri building-supply company — roughly 77% and 23%. Their buy-sell gave the survivor an option to purchase a deceased brother's shares; if he declined, the company had to redeem them. To fund that obligation, Crown held $3.5 million of life insurance on each brother.

Michael died in 2013. Thomas declined to purchase, the company's obligation kicked in, and — instead of getting the outside appraisal their own agreement called for — Thomas and Michael's son simply agreed the shares were worth $3 million. Crown paid the estate $3 million out of the insurance proceeds.

The IRS disagreed with the math. In its view, the company was worth its $3.86 million operating value plus the $3 million of insurance proceeds — about $6.86 million — making Michael's 77% stake worth roughly $5.3 million, not $3 million. The estate's answer was that the redemption obligation offset the insurance: the money came in and went right back out. A unanimous Court, in an opinion by Justice Thomas, sided with the IRS: an obligation to redeem shares at fair market value is not a liability that reduces the company's value, because a fair-value redemption leaves every shareholder's economic position intact. The insurance proceeds count; the offset doesn't. The estate owed an additional $889,914.

Two things in the opinion matter for planning. First, the Court itself pointed at the exit: a cross-purchase agreement — where the owners insure each other and buy the shares personally — would have kept the proceeds out of the company entirely, though the Court noted it carries its own burdens, including each owner paying premiums on the others and “its own tax consequences.” Second, the Court was careful about scope: it did not hold that a redemption obligation can never depress value — an obligation that forces the company to liquidate operating assets might — only that redemption obligations don't automatically offset insurance proceeds.

What OBBBA changed: who still has the problem#

When Connelly came down, the planning world braced for the federal exemption to fall by roughly half — the 2017 tax law's doubled exemption was scheduled to sunset after 2025, to somewhere around $7 million per person. That would have put an enormous number of successful business owners in Connelly's blast radius.

OBBBA, enacted July 4, 2025, eliminated the sunset. The federal estate and gift exemption is $15 million per person for 2026, permanent, and indexed for inflation from 2027 onward — effectively $30 million for a married couple that properly elects portability. (The IRS's official 2027 inflation-adjusted figure hadn't been released when this was published; it's expected in the fall adjustment cycle.) California adds nothing on top: the state has had no estate or inheritance tax for decedents dying since 2005, and nothing replacing it has been enacted since.

So does Connelly still matter? Yes — for a narrower, identifiable group, and in ways that have nothing to do with the estate tax at all:

• Owners whose estate plus the insurance approaches the line. The trap is arithmetic: corporate-owned proceeds add dollar-for-dollar to company value, and your percentage of that increase lands in your gross estate. A $13 million company with $4 million of insurance on its majority owner is, at the moment of death, closer to a $17 million company. The policy you bought to solve the succession problem is what pushes the estate over the exemption.
• Married owners relying on $30 million who never file for it. Portability isn't automatic — it requires a timely-filed estate tax return at the first spouse's death, which many estates skip precisely because “no tax is due.” An unfiled return quietly cuts the couple's effective exemption in half. (A missed election can sometimes be cured late — but that's a repair, not a plan.)
• Owners with property or domicile in estate-tax states. California has no estate tax, but Oregon's kicks in at $1 million, Massachusetts at $2 million, and Washington and New York far below $15 million. If your footprint crosses into those states, the Connelly math matters at much lower numbers.
• Everyone else — for non-tax reasons. Connelly is also a case about a buy-sell that failed at its one job. The brothers ignored their own appraisal mechanism, and the valuation fight that followed is exactly what the agreement existed to prevent. The drafting lessons below apply at any estate size.

The California wrinkle: a redemption the company can't legally pay#

Here's the point almost no national coverage touches. In California, a corporation's repurchase of its own shares is a “distribution to shareholders” under Corporations Code section 166, and section 500 permits it only if the company passes one of two tests — roughly, sufficient retained earnings, or post-payment assets at least equaling liabilities plus any preferential rights — with related solvency protections behind them.

That means a redemption buy-sell carries a structural risk no amount of insurance fully cures: at the moment the obligation matures, the company must also satisfy a statutory distributions test. A company flush with insurance proceeds will usually pass — but a company that's leveraged, has negative retained earnings, or took the valuation hit of losing its key person may find its own buy-sell obligation statutorily unperformable at exactly the wrong moment. A cross-purchase isn't subject to section 500 at all, because the buyers are the surviving owners, not the company. If your buy-sell is a redemption and your balance sheet runs lean, that's a California-specific reason to look at the structure — independent of anything the Supreme Court said.

The structures, honestly compared#

Keep the redemption — with eyes open#

If the realistic estate — company value plus death benefit plus everything else — sits comfortably under the exemption, a redemption's simplicity is still worth a lot: one policy per owner, owned and paid for by the company, no premium-sharing among owners. The Connelly inclusion is real but harmless if no estate tax results. The checklist below is still mandatory.

Convert to a cross-purchase#

Policies and proceeds sit with the surviving owners, outside the company — no Connelly inclusion, no section 500 problem, and one quietly valuable bonus: owners who purchase shares get cost basis in what they bought, where survivors in a redemption generally get no basis increase — none in a C corporation, and at best a partial one, with planning, in an S corporation. That difference compounds into real money when the company eventually sells. The cost is administrative: with more than two or three owners, the policy count multiplies fast (each owner insuring every other), premium obligations have to be policed, and unequal ages and health make premiums lopsided.

The insurance-only LLC#

The structure practitioners have coalesced around for multi-owner groups: a special-purpose LLC, taxed as a partnership, holds one policy per owner and routes proceeds to fund the purchase — solving the policy-count problem while keeping proceeds out of the operating company. It has to be built carefully: no insured member can hold the incidents of ownership over the policy on their own life (an independent manager handles that), the entity needs a genuine non-tax business purpose, and the partnership form is what keeps the transfer-for-value rules from poisoning the death benefit. One honest caveat: practitioner commentary reports the IRS has stopped issuing private rulings in this area, so the structure rests on established doctrine rather than a fresh government blessing. Done right, it stands on solid doctrinal footing; done casually, it's a pile of tax risk wearing an LLC costume.

Hybrid (“wait and see”) agreements#

A middle path worth more attention than it gets: draft the agreement so the surviving owners have the first option to purchase, with the company's redemption obligation as the backstop. The decision about which structure executes gets made at death — when the estate-tax picture, the balance sheet, and section 500 capacity are all actually known. One design note: the Connelly inclusion follows policy ownership, not the option ordering — a hybrid where the company owns the policies still puts the proceeds in company value, so the ownership question has to be answered first.

The drafting fixes everyone needs (even far below $15 million)#

Connelly's deeper lesson is that the agreement's pricing terms only protect you if they're built — and followed — correctly. Under Internal Revenue Code section 2703, a buy-sell's price controls estate-tax value only if the arrangement is a bona fide business arrangement, isn't a device to pass value to family for less than full consideration, and has terms comparable to what strangers would strike at arm's length. And then you have to actually use the mechanism: the Connellys had an appraisal process in their agreement and skipped it, which is part of why there was nothing anchoring the value when the IRS came asking.

The review-trigger checklist for a California buy-sell in 2026:

• Structure check: is it a redemption funded by company-owned policies? That's the Connelly fact pattern.
• Arithmetic check: company value plus total death benefit plus the owner's other assets — is the number within shouting distance of $15 million single / $30 million married? Remember the proceeds themselves inflate it.
• Portability check: if the plan assumes $30 million, is there an estate-counsel plan to file the return that actually elects it?
• Valuation-mechanism check: does the agreement fix price through a mechanism that satisfies section 2703 — and has every past transfer actually followed it?
• Section 500 check: if the redemption stays, can the company realistically pass California's distributions test at the moment it would have to pay?
• Community-property check: California spouses presumptively own half of what's acquired during marriage — including business interests and policies bought with community funds. Spousal consents belong in the agreement.
• Funding check: do the policy amounts still match the company's value, or was the coverage sized to a valuation from five years ago?

If the agreement predates June 2024, it was almost certainly drafted without Connelly in mind — and if it predates July 2025, its tax assumptions predate OBBBA too. That doesn't mean it's broken; it means nobody has checked. An afternoon with the document now is dramatically cheaper than a valuation fight later — that much, Connelly itself proves.