Federal Circuit Rejects Taxpayers' Arguments about Midco Transaction (5/17/26)

In Dillon Trust Co. LLC v. United States, ___ F.4th ___ (Fed. Cir. 2026), CAFC here and GS here [to come], the Court in 50 pages rejects Dillon Trust’s bullshit claims in a bullshit “midco” transaction tax shelter. I wrote on one aspect of the case below. Court of Federal Claims Rejects Defective § 6603 Strategy for Multiple Transferee Liabilities (11/21/22), here (discussing one aspect of the Federal Circuit opinion).

The Federal Circuit offers a reasonable description of midco transactions (Slip Op. 10 n. 5):

5 A “midco transaction” or intermediary transaction is structured to allow a seller to engage in a stock sale and a buyer to engage in an asset purchase. Shareholders sell their C corporation stock to an intermediary (or midco) at a purchase price that does not discount for the built-in gain tax liability, as a stock sale to the ultimate purchaser would. The midco then sells the assets of the C corporation to the buyer, who gets a purchase price basis in the assets. The midco’s willingness to allow both buyer and seller to avoid the tax consequences inherent in holding appreciated assets in a C corporation is based on supposed tax attributes, like losses, that allow it to absorb the built-in gain tax liability. But, if these tax attributes of the midco prove to be artificial, then the tax liability created by the built-in gains on the sold assets still needs to be paid. In many instances, the midco is a newly formed entity created for the sole purpose of facilitating such a transaction, without other income or assets. It is thus likely to be judgment-proof, and the IRS will need to seek payment from the other parties involved in the transaction to satisfy any unpaid tax liability. 

That definition is antiseptic, eliminating details of the high- or low-drama presented in the rest of the opinion where the players on the taxpayer side unsuccessfully claimed a type of willful ignorance about the transaction to justify their participation and, they hoped, avoid the transferee tax liability involved. They failed.

The Court affirmed that the involved parties had at least constructive knowledge of the bullshit gambit and therefore could be liable under New York’s Uniform Fraudulent Conveyance Act (see Slip Op. 21-26). Based on the players involved (see Slip Op. 3 nn. 1-3), some of whom testified, this was ““[a] deal done by very smart people that, absent tax considerations, would be very stupid.” Michael Graetz statement, oft quoted, see e.g., Lynnley Browning, How to Know When a Tax Deal Isn’t a Good Deal (New York Times 10/10/08).

On what these taxpayers (including the company whose stock was purchased), their transferees, their advisors, and other players in the “deal” knew or should have known, the Federal Circuit affirmed the CFC holding of constructive fraud. (In my mind, the facts affirmed by the Federal Circuit might have even permitted a reasonable inference of actual fraud in the sense that the selling shareholders in the midco transaction knew enough to also know that the taxes were going to be avoided/evaded.) Readers might want to read through the opinion to appreciate the high- or low-drama.

Key bullet points of the holding are:

• Transferee liability under § 6901 applies and invokes the NY UFCA. (Slip Op. 20-39.)
• The amount of liability included the transferor corporation’s tax, penalties, and interest. On including the penalty, the Federal Circuit noted that there “appears to be a circuit split” on including penalties. (Slip Op. 42) The Court of Appeals adopted the majority of Circuits inclusion of penalties (Slip Op. 42-44.)
• The § 6603 deposit made by the Dillon Trust could not avoid interest on other taxpayers’ liabilities. Slip Op. 44-50.)

As I view it, the parties knew or should have known that, if the IRS spotted the transaction and spent the resources to "unpeal the onion," it would likely assert transferee liability. So, it seems that they rolled the dice on the audit lottery. These taxpayers lost that roll of the dice. But there are many others who have won that gamble.