Indemnification, Escrows and R&W Insurance in Private Deals

The risk-allocation stack in an SPA

A purchase agreement for a private company answers one question many ways: if something the seller said turns out to be untrue, who pays, how much, and for how long? The answer is a stack of interacting terms. Read them together, because a generous cap means little if the survival period is short, and a large escrow means little if the basket swallows most claims.

Two related mechanisms sit beside the stack and are often confused with it. A purchase price adjustment trues up working capital, cash and debt at closing; it compensates for a number being different, not for a representation being false. An earn-out defers part of the price against future performance. Both can be netted against indemnity claims if the agreement says so.

This article describes contract terms, published deal data and Delaware case law. It is not legal advice; the drafting of any purchase agreement belongs with counsel.

Baskets, escrows and survival: what recent data show

The largest public window into private-deal terms is SRS Acquiom's annual deal terms study. The 2026 edition "analyzes 2,300+ private-target acquisitions, valued at $569 billion" that closed between 2020 and 2025 for which SRS Acquiom provided services ^[1]. The full study is gated; the figures below come from Fasken's published summary of it ^[2].

Indemnification escrows as a percentage of deal value (SRS Acquiom data via Fasken)

The gap between the "No RWI" and "RWI" rows is the clearest number in private M&A: when the buyer has insurance, the seller's indemnification escrow shrinks from about a tenth of the price to a fraction of a percent at the median ^[2].

Other 2025 data points from the same summary:

• RWI incidence. RWI was identified on approximately 46% of deals. Because buyers do not always disclose buy-side RWI, the "no RWI" sets "likely include some deals where RWI was used."
• Baskets. Deductible baskets fell from 39% of deals in 2024 to 32% in 2025; first-dollar baskets rose to 40%, and no-basket structures rose to 28%.
• Walk-away deals. "No survival" constructs in non-RWI deals, meaning the seller has no post-closing liability for general representations, dropped from 18% in 2024 to 11% in 2025.
• Survival in RWI deals. 43% of RWI deals included survival of the seller's representations and warranties, the lowest share in four years.
• Earnouts. Earnouts appeared in 24% of deals in 2025, up from 22% in 2024.

All of these are from Fasken's summary.

What the dataset is not. It covers deals where SRS Acquiom provided services, and Fasken notes that 9% of 2025 deals exceeded US$750 million in value. It is not a sample of the lower middle market, and nothing in the published summary reports these terms by deal size for small transactions. Treat it as a description of the institutional private-target market, not as a benchmark for a $5 million acquisition.

How long a claim can live: survival and 10 Del. C. § 8106(c)

Survival clauses are the contractual layer. Statutes of limitations are the legal floor or ceiling beneath them. In Delaware, the default limitations period for contract actions is three years from accrual under 10 Del. C. § 8106(a). Subsection (c) lets sophisticated parties go longer: an action based on a written contract "involving at least $100,000" may be brought "within a period specified in such written contract," provided it is brought before 20 years from accrual ^[3].

In practice, purchase agreements use survival to do three things:

1. Shorten the window for general representations, so the seller's exposure ends on a known date and the escrow can be released.
2. Lengthen the window for fundamental representations (title to equity, authority, capitalization) and often tax, sometimes to the statutory maximum the governing law allows.
3. Define what "survive" means: whether a claim must be noticed before expiry or filed in court before expiry. That difference has decided cases; it should be expressed in plain words, not left to an inference.

A buyer should map the survival schedule against when problems tend to surface. Aon's 2026 claims study found that approximately 51% of North American R&W claims were notified more than 12 months after closing, and 24% more than 24 months after closing ^[4]. A 12-month general survival period would have closed the door on roughly half of those notices had they been seller indemnity claims.

Fundamental reps and fraud carve-outs

Every indemnity regime has a hole cut in it for fraud. The question is how big.

Delaware's answer comes from ABRY Partners V, L.P. v. F&W Acquisition LLC, 891 A.2d 1032 (Del. Ch. 2006), applied by the Delaware Supreme Court in Express Scripts, Inc. v. Bracket Holdings Corp., No. 62, 2020 (Del. Feb. 23, 2021). In the Supreme Court's summary, ABRY concluded that a contracting party cannot, as a matter of public policy, "limit . . . exposure for its own conscious participation in the communication of lies," but a buyer that "knowingly accepted the risk that the [counterparty] would act with inadequate deliberation" may not escape contractual limits "by attempting to show that the [counterparty] acted in a reckless, grossly negligent, or negligent manner" ^[5][6].

Express Scripts is also a case about RWI. In a $187 million securities purchase agreement, the buyer agreed that, "absent deliberate fraud, its sole and exclusive remedy for breach of the representations and warranties was the R&W Policy," apart from certain fundamental representations. The buyer recovered $13 million under the policy in arbitration, then won a jury verdict of over $82 million for fraud in the Superior Court. The Supreme Court held that instructing the jury it could find liability for recklessness was not harmless error, because "[a] deliberate state of mind is a different kettle of fish than a reckless one," and reversed and remanded ^[5].

Practical consequences for the stack:

• Define fraud. If the agreement says "Fraud means actual and intentional fraud with respect to the making of the representations," recklessness claims are likely out. If it says nothing, expect litigation over what "fraud" means.
• Separate fundamental reps. Title, authority and capitalization usually sit outside the basket and the general cap, with longer survival.
• Know who carries the fraud carve-out. In an RWI deal with a sole-recourse clause, the carve-out for intentional fraud may be the buyer's only path back to the seller. Check whether the policy preserves the insurer's subrogation rights against the seller for fraud; read the policy.

What R&W insurance does and where it stops

RWI, as discussed here, is a buyer-side policy that pays covered losses from breaches of the seller's representations, above a retention and up to a limit. It replaces most of the seller's indemnity, which is why escrows shrink.

Aon's 2026 Transaction Solutions Global Claims Study gives the best public view of how North American policies respond:

Three caveats keep these numbers honest. First, they come from Aon's own "internal, aggregated and anonymized claims databases", so they describe Aon-placed policies. Second, the payment figures rest on roughly 40 payments in one year, a small base for a median. Third, a policy that pays $8 million median claims is written on deals much larger than most search-fund or independent-sponsor acquisitions ^[4].

Where RWI stops is policy-specific, and no generic list substitutes for the binder. Ask how the policy treats matters the buyer knew about before closing, diligence findings, forward-looking statements, purchase price adjustments and any deal-specific exclusions the underwriter adds, and ask the broker which quality of earnings findings the underwriter will treat as known.

Small deals: when RWI economics and escrows diverge

The public datasets above skew toward larger transactions, and none of the sources verified for this article publishes RWI premium, retention or minimum-deal-size figures for the lower middle market. Anyone quoting a precise "RWI is uneconomic below $X million" threshold should be asked for the source.

What can be said without invented numbers:

• Fixed costs bite harder on small deals. Underwriting requires diligence reports the insurer can rely on. A buyer who would not otherwise commission a full quality-of-earnings report, legal diligence and tax review will add those costs to the premium.
• Retentions are sized to the deal. The retention is the buyer's first-loss layer; on a small deal the absolute amount may be modest, but so is the buyer's equity cushion.
• Seller financing changes the calculus. Where a seller note or earn-out is part of the price, the buyer already holds money it may be able to set off against claims, if the agreement grants a set-off right. That can do some of the work an escrow does.
• Lenders have views. Senior lenders and SBA lenders may set their own requirements on escrows and seller notes; check before the LOI fixes the structure.

The decision is a quote-driven comparison: escrow size and release timing versus premium, retention and exclusions. Get real quotes from a broker and counsel's read of the exclusions.

Checklist: settle these before the LOI fixes the structure

None of these answers is standard. Each is priced, explicitly or implicitly, against the others.

Worked example: one breach, two structures (fictional)

Illustrative example. The company, numbers and policy terms are invented to show mechanics; they are not market data.

Harrow Valve Company, Inc. is sold for $20,000,000. Eighteen months after closing, the buyer discovers that a customer rebate program was not accrued, so the audited financial statements overstated income. Agreed loss: $1,500,000.

Structure A. By month 18 half the escrow ($1,000,000) has already been released to the seller at month 12. The buyer notices the claim just inside the survival period. Recovery is $1,500,000 minus the $150,000 deductible = $1,350,000. The remaining escrow covers $1,000,000; the buyer must pursue the seller directly for $350,000, possibly by setting off against a seller note if the agreement allows.

Structure B. The seller's 12-month survival has expired and its escrow has been released, so the seller pays nothing. The buyer absorbs the uncovered part of the retention, here $200,000 (because the seller's $100,000 escrow layer is gone). The insurer pays $1,300,000, subject to the policy's terms and to the breach not being excluded or known before closing.

The comparison shows the trade: Structure A keeps the seller on the hook but leaves the buyer collecting from a person it may still be working with; Structure B gives the seller a clean exit and moves the risk to an insurer and to the buyer's retention. Timing decides a lot of it. The same breach noticed at month 20 would get nothing in Structure A and the same insurer payment in Structure B.

Diligence findings that feed the indemnity negotiation

Specific indemnities, carve-outs from the policy and the size of the escrow are all negotiated around what diligence finds. Some of that evidence is available before a letter of intent, from public records:

Findings like these move items from "general representation" to "specific indemnity" or "known matter", which is exactly where an RWI policy will exclude them. Recording each finding with its source and retrieval date, as in a research handoff evidence packet, lets counsel decide early which risks need their own line in the agreement.

Docket's agents compile that kind of quoted, source-linked evidence on targets before outreach; they do not replace legal diligence.