Indemnity Caps, Baskets and Survival in M&A (India)

In an Indian share purchase or subscription agreement (SSA), the indemnity clause promises to compensate the buyer for losses caused by a breach of the seller's representations. That promise is never unlimited. Four mechanics decide what it is worth: a cap (the maximum the seller ever pays), a basket or threshold (losses must add up past a floor before a claim is allowed), a de minimis (individual claims too small to count are ignored), and a survival period (a deadline to bring a claim). Most first-time sellers and buyers negotiate the cap percentage hard and barely look at the basket type or survival period, when a "deductible" basket or a survival window that ends before problems typically surface can quietly gut a cap that looked generous on the term sheet. This guide is published by Adira, which makes contract review and CLM software, including tools deal teams use to track indemnity caps and escrow release dates, so we have a commercial stake in you understanding this well. It stands on its own regardless of what you buy afterward.

Plain meaning

Picture a ₹100 crore acquisition. The seller gives representations: the company owns its IP, its accounts are accurate, there is no undisclosed litigation, taxes are properly paid. If one turns out false and the buyer suffers a loss, the indemnity clause lets the buyer claim money back instead of being stuck with a bad deal. Four numbers decide how that claim plays out. Say the deal sets a cap of 20% of deal value (₹20 crore), a basket of 1% (₹1 crore), a de minimis of 0.1% per claim (₹10 lakh), and a survival period of 18 months for general representations. A single ₹5 lakh claim does not count at all, it is below de minimis. Claims add up until they cross ₹1 crore in aggregate before the buyer can claim anything. Whether the buyer then recovers only the amount above ₹1 crore or the full amount from the first rupee depends on whether the basket is "deductible" or "tipping", a distinction worth more in practice than most one-line basket clauses suggest. And however large the loss, the seller's liability never exceeds ₹20 crore, and no claim can be brought once 18 months from closing have passed, however genuine.

Who it protects and what triggers it

These mechanics exist because an unlimited, undated indemnity serves neither side. A buyer with no cap still has a worthless remedy if the seller cannot pay; an individual promoter rarely has ₹100 crore of headroom sitting around. A seller facing an uncapped, undated exposure can never close its books on the deal or price the risk. Together, the four mechanics turn an open-ended promise into something both sides can price, insure against, and rely on.

The trigger is a breach of a representation (or a specific indemnity for a known risk carved out during diligence, such as a pending tax dispute) that causes a quantifiable loss, discovered and notified within the survival window, and large enough to clear the de minimis and the basket. Miss any one condition and the claim, however genuine the underlying breach, goes nowhere.

What to look for

Six mechanics decide whether the indemnity architecture is fair or a trap, and most are invisible unless you read past the headline cap:

1. What the cap applies to. A single cap for everything, or a higher (or uncapped) tier for "fundamental" representations, title, authority to sign, capitalisation, and a lower one for general representations?
2. Basket type: deductible or tipping. A deductible basket means the buyer recovers only losses above the basket. A tipping basket means once aggregate losses cross it, the buyer recovers the entire amount, first rupee included. On ₹1.5 crore of proven losses against a ₹1 crore basket, a deductible basket pays ₹50 lakh; a tipping basket pays ₹1.5 crore.
3. De minimis threshold. Too low, it does nothing. Set sensibly (commonly 5-10% of the basket), it filters nuisance claims without letting a seller argue a string of ₹9 lakh losses never individually "counts."
4. Survival period, by category. General reps often survive 12-24 months; fundamental reps longer, sometimes indefinitely. Tax indemnities need to run long enough to match when a tax authority can reopen an assessment, often years longer than a general-rep window (see below).
5. Sandbagging language. Can the buyer claim for a breach it actually knew about before closing? "Pro-sandbagging" says yes; "anti-sandbagging" bars a claim on anything known. Silence invites a fight, and connects to how Indian misrepresentation law treats disclosed facts (see below).
6. Whether the cap is backed by money. A ₹20 crore cap against a seller with no escrow, holdback or insurance is a number on paper, not a recoverable amount.

The Indian position: Sections 124-125, freedom of contract, and Section 28 on survival periods

Indemnity is a defined contract type under the Indian Contract Act, 1872. Section 124 defines it as "a contract by which one party promises to save the other from loss caused to him by the conduct of the promisor himself, or by the conduct of any other person." Section 125 sets out what the indemnity-holder can recover once sued: damages, costs, and sums paid under a compromise, "acting within the scope of his authority." (See our full indemnity clause guide for how these sections work generally.) Neither mentions caps, baskets or survival periods, those are contractual architecture, which Indian courts respect under the general freedom to contract, so long as the object is lawful. There is no doctrine striking down a cap for being low, or a basket for being deductible. What you negotiate is largely what you get.

Survival periods sit on trickier ground. Section 28 of the Contract Act, as amended in 1997, voids "every agreement" that "extinguishes the rights of any party thereto... on the expiry of a specified period." Read literally, a clause that "expires" 18 months after closing, cutting off a genuine claim discovered in month 19, is exactly what Section 28 was written to catch. The Act carves out one exception, for bank and financial institution guarantees with a stated period of at least one year, but an ordinary SSA survival clause is not covered. Indian courts have not struck down commercially negotiated survival clauses wholesale, and this remains a live, somewhat unsettled area. The safer drafting choice: give fundamental and fraud-related representations a long or indefinite survival, and match tax indemnities to how long the tax authority can actually reopen the matter, rather than betting on Section 28 to rescue a late claim.

That last point has a checkable answer. Under Section 149 of the Income Tax Act, 1961 (as amended by the Finance Act, 2021), a reassessment notice can generally be issued within 3 years of the relevant assessment year, extending up to 10 years where the department has evidence that income exceeding ₹50 lakh has escaped assessment. An 18-month general survival period does nothing for a tax exposure the department can reopen in year 4 or 8, which is why tax indemnities are almost always carved out with their own, longer survival period.

Cross-border deals add one wrinkle: how much of the price can sit in escrow. Rule 9(6) of the Foreign Exchange Management (Non-Debt Instruments) Rules, 2019 caps deferred consideration, escrow, and seller-side indemnity holdbacks in a resident-to-non-resident share transfer at 25% of total consideration, for a period not exceeding 18 months from the transfer agreement. A foreign buyer wanting more, say 30% in escrow or a 24-month hold, needs specific RBI structuring or approval; it cannot simply draft around the rule.

A named Indian case: Avitel Post Studioz v HSBC PI Holdings

Avitel Post Studioz Ltd & Ors v HSBC PI Holdings (Mauritius) Ltd, (2021) 4 SCC 713, arose from a share subscription agreement where HSBC invested in Avitel based on representations about a genuine business contract that, the tribunal found, Avitel's promoters had fabricated. An SIAC tribunal awarded HSBC USD 60 million in damages for fraudulent misrepresentation, calculated by reference to the loss actually caused, not the original investment size. Avitel resisted enforcement in India, including on the ground that it offended Indian public policy. The Supreme Court disagreed and allowed enforcement. See the judgment on Indian Kanoon.

Why this matters for drafting: Avitel is the clearest signal that a genuinely fraudulent misrepresentation is not something a seller can expect a contractual cap to shrink down to size. Market-standard Indian SSAs carve fraud and wilful concealment out of the general cap entirely, and "our cap covers everything, including fraud" should read as a red flag, not a reassurance.

Red flags

Normal	Red flag	Why it matters
Cap for general reps around 10-25% of deal value; fundamental reps higher or uncapped	Cap so low (under 5%) it barely covers one serious breach	Representations become decorative; the seller has priced in getting away with a breach
Basket clearly labelled "deductible" or "tipping"	Basket type left unstated or ambiguous	A deductible basket can pay half of what a tipping basket pays on identical losses
De minimis around 5-10% of the basket amount	De minimis missing, or set so low it excludes almost nothing	The buyer loses the protection the basket was meant to give
Fundamental reps and fraud carved out of the general cap, or given a much higher one	Fraud and title claims subject to the same low general cap	Directly the Avitel problem: it discounts a fraudulent seller's own dishonesty
Survival matched to risk: 12-24 months general, longer for fundamental, tied to Section 149 for tax	Survival shorter than when the problem typically surfaces, especially tax	A valid tax claim surfacing in year 4 is worthless against an 18-month clock
Escrow sized close to the cap, released only after survival ends	Escrow released early, or sized far below the cap	A cap with no money behind it means chasing a seller who has spent the proceeds
Sandbagging position stated explicitly	Clause silent on sandbagging	Silence turns a disclosed issue into a fight over what "known" meant

Bad clause → better clause

Bad: "Seller shall indemnify Buyer for all Losses arising from any breach of the representations and warranties in this Agreement, subject to a cap of ₹20,00,00,000."

What is wrong: one cap for every representation, including fundamental and fraud claims; no basket or de minimis, so any claim of any size can be brought; no stated survival period; and no mechanism for actually collecting the ₹20 crore from the seller.

Better: "Seller's aggregate liability for breach of the General Representations (Schedule 3) shall not exceed 20% of the Purchase Price. Seller's aggregate liability for breach of the Fundamental Representations (Schedule 2) and for Losses arising from fraud or wilful concealment shall not exceed 100% of the Purchase Price. No claim may be brought for an individual Loss below ₹10,00,000 (De Minimis), and such Losses shall not count toward the Basket. Buyer may only claim once aggregate Losses exceeding the De Minimis cross ₹1,00,00,000 (Basket), following which Buyer may recover such Losses from the first Rupee (Tipping Basket). Claims for General Representations must be notified within 18 months of Closing; Fundamental Representations within 3 years; Tax Representations within 30 days of expiry of the applicable limitation period under Section 149 of the Income Tax Act, 1961. Buyer's knowledge of a matter fairly disclosed in the Disclosure Schedule shall not bar a claim, except as expressly agreed for specific disclosed items. ₹10,00,00,000 of the Purchase Price shall be held in escrow for 18 months from Closing to satisfy claims under this Clause."

What changed: the cap is split by category so fraud and title are not quietly capped at the general-rep level; a de minimis and a labelled tipping basket replace an all-or-nothing threshold; survival periods match the actual risk, including a tax link; sandbagging is addressed head-on; and escrow backs the cap with real, reachable money.

How it interacts with related clauses

• Indemnity. Cap, basket and survival sit on top of the underlying promise. See the full indemnity clause guide for how Sections 124-125 shape what these mechanics limit.
• Representations and warranties. The cap, basket and survival period only matter because a representation was breached. See Representations vs Warranties in Indian Contracts for how sandbagging connects to Indian misrepresentation law.
• Limitation of liability. In a services contract, indemnity typically sits inside a general liability cap; in an SSA, the cap-basket-survival package usually is the limitation of liability mechanism. See our limitation of liability guide.
• Escrow. A cap is only as real as the money backing it. Size and duration the escrow to cover claims likely near the end of the survival period, not just the headline cap.

Map how your cap, basket and survival clauses line up against each other in a document, for free, using Weave, before you send a term sheet or draft SSA back for negotiation.

US and global contrast

US private M&A runs on the same four mechanics, cap, basket, de minimis, survival, but with a more standardised market range: general-rep caps commonly sit around 10-15% of deal value, and representation and warranty insurance (RWI) is common enough on mid-market US deals that it sometimes replaces a meaningful seller indemnity altogether. RWI is still rare in Indian private M&A, which puts more weight on the seller's own balance sheet, and on escrow, to make an Indian buyer's cap collectible. Sandbagging defaults also differ by US state, Delaware leans pro-sandbagging by default, while Indian contracts have no default and must state a position expressly. India's Section 28 uncertainty around survival periods, and the FEMA 25%/18-month rule on cross-border escrow, have no real US equivalent; a US buyer moving into an Indian deal is often surprised that a cap and escrow structure it considers standard needs RBI-compliant restructuring to work at all.

FAQ

What is the difference between a deductible basket and a tipping basket? A deductible basket means the buyer recovers only the amount of losses above the basket. A tipping basket means once aggregate losses cross the basket, the buyer recovers the entire amount from the first rupee. The same basket number produces very different recoveries depending on which type it is.

Why do fundamental representations get a higher cap or no cap at all? Fundamental representations, ownership, authority to sell, capitalisation, go to whether the buyer actually gets a valid deal at all. A breach here is not a normal business risk, which is why these are typically carved out of the general cap and often capped at or near the full purchase price, sometimes uncapped for fraud.

Can a survival period in an Indian SSA be challenged as void? Potentially, under Section 28 of the Indian Contract Act, 1872, which voids clauses that extinguish a party's rights on expiry of a specified period, subject to a narrow exception for bank and financial institution guarantees. This is not a settled bar on all commercial survival clauses, but it is a real, checkable risk for very short periods.

What does "sandbagging" mean and does Indian law have a default rule on it? Sandbagging is bringing an indemnity claim for a breach the buyer already knew about before closing. Indian law has no default rule the way some US states do; it interacts instead with Section 19 of the Contract Act, where a fact within the buyer's means of discovering the truth can affect a misrepresentation claim. An Indian SSA should state its sandbagging position explicitly.

Does the FEMA 25%/18-month rule apply to all Indian M&A, or only cross-border deals? Only to a transfer of equity instruments between a person resident in India and a person resident outside India, under Rule 9(6) of the FEMA (Non-Debt Instruments) Rules, 2019. Purely domestic Indian share transfers are not bound by it, which is one reason domestic escrow structures can be sized and timed more freely.

This guide gets you to understanding what a cap, basket, de minimis and survival period do in an Indian SSA, and the statutory limits, Section 28, Section 149, and the FEMA 25%/18-month rule, that shape how they should be drafted. It does not tell you whether the numbers in your deal are fair, market-standard, or enforceable on your facts, that depends on the deal, the disclosure schedule, and drafting history, and is not legal advice. Talk to an M&A lawyer before you agree to, or walk away from, a cap and basket structure in a live negotiation.