This post uses hypothetical scenarios for illustrative purposes only. It does not describe any actual client, transaction, or representation, and is not legal advice.
Here is how this usually shows up in 2026 diligence. The quality-of-earnings report on a target — say a staffing company or a restaurant group — flags a line in other income from 2023: a payroll tax refund in the high six figures. The seller’s CFO explains it cheerfully enough. A consultant called during the pandemic recovery, said the company qualified for the Employee Retention Credit, took a percentage of the recovery as its fee, and filed amended payroll returns. The money arrived. Everyone moved on. Nobody in the building can locate the eligibility analysis, and the consultant’s firm no longer answers its phone.
Three years ago, a buyer might have shrugged at that story. In 2026, it should reprice the deal or restructure the indemnity — because Congress rewrote the enforcement rules, and the audit tail on that refund now runs years past the survival period in a standard purchase agreement.
OBBBA turned the ERC into a six-year liability
The One Big Beautiful Bill Act, enacted in July 2025, contains a section — § 70605 — devoted entirely to ERC enforcement. Three of its moves matter to deal lawyers.
First, the assessment window stretched. The Act amends section 3134(l) of the Code to give the IRS six years to assess ERC-related liabilities, double the standard three-year employment tax period. The six years run from the latest of the original return’s filing date, the deemed filing date, or — critically — the date the refund claim itself was submitted. A target that filed its amended 941s in early 2024 can be audited into 2030 or beyond. The wage-deduction side is matched: if the IRS disallows the credit, it gets the same extended window to restore the corresponding deduction adjustments, so the exposure compounds across payroll tax and income tax.
Second, the claim window closed permanently. No refund is allowed for any claim filed after January 31, 2024. That means the pipeline is frozen — what the target claimed is what it has, and any “pending” claim filed after the cutoff is now a dead asset that diligence should value at zero.
Third, the promoters who manufactured the aggressive claims are now themselves enforcement targets, facing six-figure penalties and listed-transaction reporting obligations that reach back to 2020. The IRS’s ERC program page reflects the posture: as of mid-2026 the Service was still working through roughly 20,000 claims in various stages of review, audit, disallowance, and appeal. Promoter reporting gives the IRS a roadmap of exactly which employers used which advisors. If the target’s claim came through a contingency-fee mill, the odds that the claim file surfaces in an IRS database are meaningfully higher than they were two years ago.
Why the standard purchase agreement misses this
The typical private-deal architecture handles taxes with a pre-closing tax indemnity, a survival period of twelve to twenty-four months for general reps, and a longer tail — often the statute of limitations plus sixty days — for the tax reps. That template fails on the ERC in a quiet way: many agreements define the tax-rep tail by reference to “the applicable statute of limitations” without anyone checking what that statute actually is. Where the credit is in play, the applicable period is now six years from a date that may be well after the return year. An escrow that releases at eighteen months, a tax indemnity that assumes a three-year tail, and a seller who plans to dissolve or distribute proceeds — put those together and the buyer holds a liability with no one left to collect from.
In a stock deal the analysis is blunt: the employer entity is the taxpayer, the entity is what the buyer bought, and the repayment obligation — credit, interest, and a potential 20 percent accuracy-related penalty — lands on the buyer’s new subsidiary. In an asset deal the payroll tax liability generally stays behind with the seller entity, but buyers should not treat structure as a complete answer. The seller entity that keeps the liability is usually also the entity that liquidates after closing, which is precisely why the diligence and the indemnity architecture matter more than the deal form.
What the diligence request should actually ask for
A generic “all tax returns and correspondence” request will surface the amended 941s and little else. The useful requests are narrower, and they belong in the first diligence list rather than a later supplement. Ask for the eligibility memo — the contemporaneous analysis of which quarters qualified and why, whether under the gross-receipts decline test or the full-or-partial-suspension test, with the specific government orders identified. Ask for the promoter’s engagement letter, because a contingency fee calculated as a percentage of the credit is the single loudest audit-risk signal the file can contain. Ask for the computation of qualified wages, the PPP interaction analysis, and any IRS correspondence — disallowance letters, Letter 6612 audit notices, or voluntary-disclosure filings. And ask the question nobody likes to answer: was the deduction for wages reduced by the credit amount on the income tax side, as required? If it wasn’t, there are two problems in the file, not one.
If the answers come back thin — no memo, a vanished promoter, a fee agreement tied to the refund size — the buyer does not need to conclude the claim was wrong. It needs to conclude the claim is undefendable, which for pricing purposes is nearly the same thing.
Structuring around a claim you cannot verify
Where the exposure is real, the tools are familiar; the calibration is what changes. First, a standalone ERC rep — not folded into the general tax rep — covering eligibility, documentation, promoter involvement, and the wage-deduction adjustment. Second, a special indemnity for ERC recapture that sits outside the basket and cap architecture entirely, because this is exactly the kind of known, quantifiable, seller-era liability that baskets and deductibles were never meant to dilute. Size it at the full credit plus interest plus the accuracy-related penalty, and run its survival to the end of the six-year assessment window — not to the generic survival date.
Third, secure it. A special indemnity from a seller who dissolves is a press release, not a remedy. That means a dedicated escrow or holdback with a release schedule tied to the assessment window or to an IRS no-change closing, whichever comes first. Sellers will resist a six-year escrow, and reasonably so; the usual landing zone is a stepped release with a long-stop tied to audit commencement. Do not assume representation and warranty insurance fills the gap — by 2026, ERC exposure appears on most underwriters’ standard exclusion lists, and where it doesn’t, the underwriter will ask for the same eligibility file the seller cannot produce.
Fourth, deal with any refund still in the pipeline. If the target filed before the January 2024 cutoff and the money hasn’t arrived, the agreement should say who owns the refund if it pays post-closing, who controls the audit or appeal if it doesn’t, and whether the purchase price flexes either way. A pending claim is simultaneously an asset the seller wants credit for and a liability the buyer may inherit; leaving it to the boilerplate wires-crossed provisions is how post-closing disputes get born.
The founder’s side of the table
None of this is only buyer’s medicine. A founder who claimed the credit on solid ground — a genuine suspension order, a documented receipts decline, a file that holds up — should assemble that file before going to market, for the same reason sellers run sell-side quality of earnings: the absence of documentation gets priced as if the claim were bad. The likely outcome for a well-papered claim is no escrow at all, or a modest one. The likely outcome for an undocumented claim is a dollar-for-dollar holdback the founder funds personally. Courts and the IRS will ultimately sort out which pandemic-era claims were legitimate; the M&A market sorts sellers much faster, and it does it at the closing table.
If you are buying or selling a company that claimed the Employee Retention Credit, feel free to reach out to my firm manager, Magda, at Magda@montague.law, or fill out our contact form. Mention you read this post.
