Force Majeure Clause Drafting in India: 2026 Guide

Last verified: 2026-06-29

On 20 April 2020, a global oilfield-services contractor walked out of the Delhi High Court holding what looked like a clean win. The court had granted an ad-interim injunction stopping the project owner from encashing eight bank guarantees the contractor had furnished on a large drilling contract. The contractor’s argument was simple and, in that first lockdown spring, very fashionable: the nationwide COVID shutdown was a force majeure event, performance had stalled through no fault of its own, and the owner should not be allowed to cash in the security while the country was frozen. For six weeks, that argument held. This is exactly the kind of moment that makes force majeure clause drafting in India look easy, and exactly the kind of moment that misleads drafters.

Then the same court looked again.

On 29 May 2020, the Delhi High Court vacated its own injunction in Halliburton Offshore Services Inc. v. Vedanta Limited, 2020 SCC OnLine Del 542. The reasoning was not that COVID could never be a force majeure event. It could. The problem was timing. The contractor had already fallen into default on its delivery obligations well before the lockdown arrived. You cannot reach back and use a later event to excuse a breach that was already complete. The pandemic did not cause the default; the default came first, and the pandemic was being used as cover. Same facts, same clause, same court. Opposite outcomes, six weeks apart.

That whiplash is the whole lesson of this guide compressed into one case. A force majeure clause is not a magic shield you wave when the world goes wrong. It is a piece of engineering. It works only if the event is genuinely caught by the wording, only if performance was actually prevented and not merely made harder, only if you were not already in breach, and only if you served the right notice in the right window. Get any of those wrong and the clause that looked bulletproof in the contract collapses in the courtroom.

Here is the part that should give every drafter pause. The contractor in that story was a sophisticated commercial party with sophisticated lawyers, and it still lost the protection because of facts that drafting and timing should have anticipated. The professionals who came out of 2020 ahead were the ones who had drafted clauses that named the right events, demanded the right notice, separated true impossibility from mere expense, and carved out exactly what force majeure could and could not touch. They did not get lucky. They drafted for the day performance fails, because that is the only day a force majeure clause is ever read.

This guide teaches you to draft that clause. It covers the doctrine that Indian courts actually apply, the eight limbs of a defensible clause, a full annotated model you can adapt, drafting variations by contract type, and a step-by-step playbook for invoking the clause when the time comes. The law sits underneath the drafting, not on top of it, so a reader who wants clause language can jump straight to it.

What should a force majeure clause include? A well-built force majeure clause should contain seven working parts: a defined list of qualifying events plus a carefully worded catch-all, a clear trigger standard (“prevent, hinder or delay” rather than only “render impossible”), notice mechanics with a firm window and prescribed mode, a consequence cascade of suspension first and then termination at a long-stop date, a duty to mitigate, the treatment of payment and cost during suspension, and the governing law, survival and severability provisions. Miss one limb and the clause leaks.

The sections below build each of those limbs from the law up. If you only want the clause language, skip ahead to the annotated model clause or the invocation playbook using the table of contents. If you want to understand why the clause is worded the way it is, start at the top with the doctrine.

What force majeure means under Indian contract law (Section 32 vs Section 56)

Most drafters reach for a force majeure clause without ever asking which part of Indian law it actually plugs into. That gap matters, because the answer decides whether the court reads your clause or reaches past it. When performance fails, the first question a judge asks is not “was this unfair?” but “which legal regime governs this excuse?” And the surprising truth is that Indian law has no codified doctrine called force majeure at all.

The phrase is borrowed from French civil law. The Indian Contract Act, 1872 never uses it. What the Act gives you instead are two doors. The first is Section 32 of the Indian Contract Act, 1872, which deals with contingent contracts: agreements to do or not do something if an uncertain future event happens or does not happen. A force majeure clause is, in structure, exactly that. It says the parties’ obligations are contingent on the non-occurrence of a defined disruption. The second door is Section 56 of the Indian Contract Act, 1872, the doctrine of frustration, which says a contract to do an act becomes void when that act becomes impossible or unlawful after the contract is made, through no fault of either party.

Here’s the thing that trips people up. These two doors are not interchangeable, and you do not get to pick. Which one applies depends on one fact: does your contract contain a force majeure clause or not?

The Supreme Court settled this in Energy Watchdog v. Central Electricity Regulatory Commission, (2017) 14 SCC 80. Where the contract contains a force majeure clause covering the event, relief is governed by that clause under Sec 32, and the court will hold the parties to the bargain they actually drafted. Where the contract has no clause, or the clause does not cover the event, the disappointed party falls back on the statutory doctrine of frustration under Sec 56. The clause is not a supplement to frustration. It is a substitute that displaces it. Draft the clause, and you have taken the question away from the court’s general doctrine and put it inside your own wording.

What does “impossible” mean once you are in Sec 56 territory? Not what most people think. In Satyabrata Ghose v. Mugneeram Bangur & Co., AIR 1954 SC 44, the Supreme Court held back in 1954 that “impossible” in Sec 56 does not mean literal, physical impossibility. It means impracticability: the underlying purpose of the contract has been so fundamentally defeated that to insist on performance would be to enforce something the parties never really bargained for. That sounds generous to the party claiming relief. In practice it is not, because Indian courts have read it narrowly ever since, and a contract that has merely become harder or costlier to perform is nowhere near frustrated.

The doctrine did not arrive fully formed. [HISTORICAL] The 1872 Act laid down Sec 32 and Sec 56 with no mention of force majeure; the 1954 Satyabrata Ghose ruling supplied the impracticability test that still governs frustration; and the 2017 Energy Watchdog decision became the modern anchor that drew the clean line between clause-governed relief and statutory frustration. Read in sequence, those three points are the entire doctrinal spine a drafter needs. Everything in a force majeure clause is built to keep your dispute on the Sec 32 side of that line, where your wording controls the outcome rather than a court’s general equity.

In practice, what experienced commercial drafters know is that the strictness of Sec 32 cuts both ways and is usually in the drafter’s favour. Because relief flows from the clause, the court will give you exactly the relief the clause provides, no more and no less. That is a feature, not a bug. It means a precisely drafted clause is more predictable than the open-textured frustration doctrine, where outcomes swing on judicial assessment of whether the contract’s purpose was “fundamentally” defeated. The clearer your wording, the less room a judge has to surprise you.

A common question practitioners raise is whether force majeure, frustration, and impossibility are simply three names for the same thing. They are not. Force majeure is your contractual allocation of risk for defined disruptions, governed by Sec 32. Frustration is the statutory backstop under Sec 56 that applies only when there is no clause or the clause does not reach the event. Impossibility, in the Indian sense, is the impracticability standard from Satyabrata Ghose that sets the bar for frustration. Keeping these three straight is the difference between drafting to your own clause and gambling on a doctrine you do not control.

The pitfall here is the most expensive mistake in this entire area: assuming Sec 56 will bail you out when a clause exists. It will not. If your contract has a force majeure clause and the event falls outside it, you do not get to argue frustration as a fallback, because Energy Watchdog says the clause occupies the field. You drafted the allocation; you live with it. Which is precisely why the wording of the clause, not the comfort of the statute, is where the real protection lives.

The Section 32 vs Section 56 decision tree: which one governs your contract

When a disruption hits, the costliest delay is spent arguing about which legal regime applies before anyone even reaches the merits. A clear decision tree removes that ambiguity. So how do you know, in a given contract, whether your clause governs or the statute does?

Walk it step by step. First, ask: does the contract contain a force majeure clause? If no, you are out of Sec 32 entirely and into Sec 56 frustration, where you must satisfy the impracticability test from Satyabrata Ghose, a genuinely high bar. If yes, move to the second question: does the clause actually list or catch the event that occurred? A pandemic clause helps you in a pandemic; it does nothing for a sanctions shock you never named and a catch-all the court declines to stretch. If the event is listed or genuinely caught, you proceed under Sec 32 and the clause’s own relief mechanics, suspension, extension, termination, govern the outcome. If the event is not caught, you are back to arguing frustration under Sec 56, usually from a weak position, because the very existence of the clause signals the parties turned their minds to disruption and chose not to cover this one.

Here is a clean comparison of the two regimes drafters should keep on the wall.

Why does this matter so much for a drafter? Because the decision tree is really a map of control. When you draft a force majeure clause, you are pulling the decision about what counts as an excusable disruption away from a judge applying a general doctrine and placing it inside language you wrote. The better approach, in our view, is to treat that as the central drafting objective: every limb you add is a way of telling the future court “here is the deal we struck on risk, apply it, do not improvise.” The clause is your instrument of control over an uncertain event.

There is one thing the clause cannot do, and drafters who do not know this get burned. You cannot contract out of the Sec 56 floor where genuine supervening illegality is concerned. If performance becomes positively unlawful, for example a fresh statute or government prohibition makes the contracted act illegal, the contract may be void under Sec 56 regardless of what your clause says, because parties cannot agree to perform an illegal act. The express clause vs implied relief question, then, has a precise answer: an express clause is almost always safer because it gives you defined, predictable relief under Sec 32, but it sits on top of a non-negotiable statutory floor for true illegality. You draft above the floor, not through it.

A common question is whether a party with no force majeure clause can still claim relief when something genuinely catastrophic happens. Yes, but only by satisfying Sec 56 frustration, and that is a far harder road than invoking a well-drafted clause. The applicant must show the supervening event struck at the root of the contract and rendered performance impracticable in the Satyabrata Ghose sense, not merely difficult or unprofitable. The practical reality is that most parties who end up arguing frustration do so because they failed to draft a clause in the first place, which is the strongest possible argument for getting the drafting right.

The pitfall to flag at this stage is over-reliance on the catch-all to do the work the decision tree should have done. A catch-all does not convert “no clause for this event” into “covered event.” Indian courts read catch-alls narrowly (more on this in the next section), so the decision tree’s second question, is the event genuinely caught, is answered by how carefully you drafted, not by how broad your residual wording sounds.

Why most force majeure claims fail in India (strict construction)

Drafters tend to assume that a force majeure clause, once in the contract, will do its job when invoked. The data from the courtroom says otherwise. Far more invocations fail than succeed, and they usually fail for reasons the drafter could have controlled. Why does a clause that looked solid on signing so often crumble at the moment of truth?

The first reason is strict construction. Indian courts read the listed events narrowly, treating the clause as an exhaustive allocation of agreed risks rather than a generous safety net. In Energy Watchdog v. Central Electricity Regulatory Commission, (2017) 14 SCC 80, the Supreme Court confirmed that a force majeure clause is construed strictly: the relief is confined to the events the parties actually specified, interpreted in their ordinary commercial sense, and the court will not stretch the language to rescue a party from a risk it accepted. If your clause lists “flood, fire, earthquake, war” and the disruption is a regulatory ban, you do not get relief by arguing the ban is “like” the listed events. It is not on the list, and strict construction keeps it off.

The second reason is the single most common drafting misconception in this field: that “more expensive” equals “impossible.” It does not. In Naihati Jute Mills Ltd. v. Khyaliram Jagannath, AIR 1968 SC 522, the Supreme Court held that a contract is not frustrated merely because performance has become more onerous, more costly, or commercially unattractive. The same logic governs force majeure relief under Sec 32. A price rise, a currency swing, a spike in raw-material cost, a more expensive shipping route, none of these is force majeure unless your clause says so in terms, and even then the court will ask whether performance was genuinely prevented rather than merely made unprofitable. Energy Watchdog drove this home when it rejected the argument that a change in Indonesian coal-export pricing frustrated power-supply contracts: the producers had quoted non-escalable tariffs and bore that risk, and a price shock did not excuse them.

So can you claim force majeure when performance has only become more expensive, not impossible? Almost never, and certainly not on a generic clause. This is where most claimants go wrong. If your concern is cost escalation rather than true impossibility, the answer is a separate price-adjustment or hardship mechanism, not a force majeure invocation that a court will reject on the Naihati Jute standard. Trying to force an economic problem through a force majeure clause is the legal equivalent of using a hammer on a screw.

The third reason claims fail is self-inflicted: a self-induced, foreseeable, or pre-existing default defeats the claim outright. This is the Halliburton lesson from the story that opened this guide. In the Halliburton dispute, the protection evaporated because the default pre-dated the pandemic; the event being invoked came after the breach, so it could not excuse it. If you were already late, already in breach, or if the difficulty was your own doing or was plainly foreseeable at signing, the clause will not save you. Courts treat force majeure as relief for genuine external supervening events, not as retrospective cover for a party that had already failed.

[SECOND-ORDER] And here is the shift most drafters have not internalised: the leading reason force majeure claims now fail is not the events list at all, it is defective or late notice. Post-COVID litigation has made notice discipline the real battleground. A party can have a perfectly valid event, genuine prevention, clean hands, and still lose because it served notice outside the contractual window, in the wrong mode, or without the required content. The non-obvious consequence is that procedural drafting, the notice limb, now matters more than the substantive events list, which is exactly the opposite of where most drafters spend their attention. We return to notice mechanics in the invocation playbook, because it is where claims are most often won or lost.

A common question practitioners raise is why so many force majeure claims fail when the clause seemed obviously applicable. The honest answer is that “obviously applicable” is a claimant’s instinct, not a court’s test. The court applies strict construction, demands genuine prevention rather than mere hardship, checks for prior default, and scrutinises the notice. Most failed claims stumble on at least one of those four gates, and many stumble on the one the drafter ignored. The pitfall, then, is drafting only the events list and treating the rest of the clause as boilerplate, when the rest of the clause is where the claim actually lives or dies.

Anatomy of a robust force majeure clause: the eight limbs

A force majeure clause is only as good as its weakest limb, and a clause missing even one of the eight working parts is a clause with a hole a counterparty’s lawyer will find. This section builds the well-built clause limb by limb, with an expert note on why each one earns its place against the strict-construction reality from the last section. What actually has to be in the clause for it to survive contact with an Indian court?

Eight limbs. Take them in order.

Limb 1, the defined events list. Name the events you want covered, expressly and specifically: epidemic and pandemic, act of God (flood, earthquake, cyclone, fire), war, armed conflict, terrorism and sanctions, government action (lockdowns, prohibitions, export or import bans, licence revocation), strikes and labour unrest, and, increasingly, cyberattack and ransomware. Because courts construe the list strictly, anything not named is presumed excluded, so the list is your primary protection, not an afterthought. Should you list specific events or rely on general wording? List them. The specificity that feels over-cautious at drafting is exactly what wins the strict-construction argument later.

Limb 2, the catch-all, and why courts read it narrowly. After the list, add a residual phrase such as “any other event beyond the reasonable control of the affected party.” But understand what it does and does not do. Under the ejusdem generis principle, Indian courts read a catch-all as limited to events of the same genus as those specifically listed, not as an open licence covering anything inconvenient. So the catch-all is a backstop for genuinely unforeseeable events of a similar character, not a substitute for naming the events you actually care about. Why include it at all, given that limitation? Because it captures the genuinely novel disruption you could not have foreseen, while the strict-construction reality means you still must list everything you can foresee. Wide catch-all or exhaustive list, which does Indian law favour? Both, in combination: a specific list does the real work, the catch-all covers the residual unknown, and the events list should be treated as illustrative-plus-specific rather than a throwaway.

Limb 3, the trigger standard. This is the limb that decides litigation outcomes, and most drafters get it wrong. There is a world of difference between a clause that grants relief only when an event “renders performance impossible” and one that operates when an event “prevents, hinders or delays” performance. “Impossible” sets the Sec 56 bar, which is brutally hard to clear. “Prevent, hinder or delay” is far more drafter-friendly because it captures genuine disruption short of total impossibility. If only part of your obligation is affected, the trigger standard is what decides whether you get partial relief or none, so draft “prevent, hinder or delay” and define what proportionate or partial relief looks like. “Impossible” vs “prevent, hinder or delay” is not a stylistic choice; it is the single most consequential word selection in the clause.

Limb 4, notice mechanics. Specify the window (for example, written notice within seven days of the affected party becoming aware of the event), the mode (email to a named address plus registered post, not “by any reasonable means”), the required content (the event, the obligations affected, the expected duration, and the mitigation proposed), and the obligation to send ongoing updates. As the last section explained, defective or late notice is now the leading cause of failed claims, so this limb is not administrative housekeeping; it is the procedural spine of the whole clause. Draft it as tightly as the events list, and arguably tighter.

Limb 5, the consequence cascade: suspension first, then termination. A well-built clause does not jump straight to termination. It suspends the affected obligations for the duration of the event, extends timelines accordingly, and then, only if the event persists beyond a defined long-stop (commonly 90 days, sometimes 60 or 180 depending on the deal), gives either party the right to terminate. Suspension vs termination is not an either-or; it is a sequence. The 90-day long-stop is what stops a temporary disruption from becoming a permanent escape hatch, and it gives both sides certainty about when the contract finally ends if the disruption does not lift.

Limb 6, the duty to mitigate. Force majeure relieves a party from performance it genuinely cannot render; it does not licence inactivity. Build in an express duty on the affected party to take reasonable steps to avoid, overcome, or minimise the effect of the event, and to resume performance as soon as practicable. This limb matters because courts will ask whether the affected party actually tried, and a documented mitigation effort is often the difference between a claim that holds and one that looks like opportunism.

Limb 7, allocation of cost and payment treatment during suspension. Spell out who bears what while obligations are suspended. Who pays standing costs, storage, demurrage, idle workforce? Are accrued payment obligations suspended or do they continue? Is each party to bear its own costs during the force majeure period? Silence here creates the surprise fight, because the parties’ instincts diverge sharply the moment money stops flowing. How payment obligations are treated during force majeure, and who bears cost during suspension, are questions the clause must answer expressly, because the default position under Indian law is unsettled enough to be litigated.

Limb 8, governing law, jurisdiction, severability and survival. State the governing law and the forum (court or arbitral seat) so a force majeure dispute does not spawn a jurisdiction fight on top of the merits. Add severability so that if one limb is struck down the rest survives, and survival so that accrued rights, the notice obligations, and the dispute-resolution mechanism outlast termination triggered under the clause. A force majeure clause does need its own governing-law and jurisdiction anchor, or it inherits ambiguity from the rest of the contract precisely when certainty matters most.

In practice, what experienced commercial drafters know is that the strict-construction reality from Energy Watchdog is the reason every one of these eight limbs exists. The list defeats the “not on the list” argument; the trigger standard defeats the “not literally impossible” argument; the notice limb defeats the “you waived it” argument; the mitigation limb defeats the “you did nothing” argument. Each limb is a pre-emptive answer to a specific way the clause could fail in court. A clause that has all eight is not over-engineered. It is correctly engineered.

A common question is whether you really need all eight limbs in every contract, or whether a short clause will do. A short clause will do until the day it is tested, and then it will not. The pitfall is the inherited two-line force majeure clause copied from an old template, the kind that lists three acts of God and says performance is “excused,” with no notice mechanics, no trigger standard, no mitigation, and no long-stop. That clause is a liability dressed as protection, and it is exactly the clause Indian courts dismantle.

Annotated model force majeure clause (with drafter’s notes)

You can read about the eight limbs all day, but a drafter learns fastest from a clause that is actually built and annotated. What follows is a model force majeure clause, India-tuned, in eight numbered sub-clauses, with a drafter’s note after each explaining why it is worded the way it is. Adapt it freely; do not delete a sub-clause without understanding what you are removing.

Force Majeure

1. Defined events. “Force Majeure Event” means any event or circumstance beyond the reasonable control of the affected Party that prevents, hinders or delays that Party’s performance, including: (a) epidemic, pandemic or public-health emergency; (b) act of God, including flood, earthquake, cyclone, fire or extreme weather; (c) war, armed conflict, terrorism, civil unrest, or the imposition of sanctions or export or import restrictions; (d) action or order of any government or regulatory authority, including lockdown, prohibition, embargo, or revocation of a required licence or permit; (e) strike, lockout or industrial action other than one confined to the affected Party’s own workforce; and (f) cyberattack, ransomware or failure of a third-party communications or cloud-services provider.

Drafter’s note: The events are named expressly because Indian courts construe the list strictly and exclude what is not named. “Prevents, hinders or delays” is chosen over “renders impossible” so the clause operates on genuine disruption short of Sec 56 impossibility. Item (e) deliberately excludes a strike confined to your own workforce, so a party cannot manufacture its own force majeure.

2. Catch-all. “Force Majeure Event” also includes any other event of a similar nature beyond the reasonable control of the affected Party, provided that the affected Party could not have prevented or overcome it by the exercise of reasonable diligence.

Drafter’s note: The catch-all is a backstop, not the main event. Courts read it as limited to events of the same genus as those listed, so it captures the genuinely novel disruption while sub-clause 1 does the real work. The “reasonable diligence” proviso forecloses the argument that an avoidable event qualifies.

3. Exclusions. A Force Majeure Event does not include: (a) any event caused by the affected Party’s own act, omission or breach; (b) any difficulty in performance arising solely from increased cost, price fluctuation, currency movement or commercial unprofitability; or (c) any default, delay or breach that arose before the Force Majeure Event.

Drafter’s note: This sub-clause is the Energy Watchdog and Halliburton armour in clause form. Item (b) shuts the “more expensive equals force majeure” door per Naihati Jute. Item (c) shuts the “use a later event to excuse an earlier breach” door per Halliburton. Without these exclusions, the counterparty will argue around the clause; with them, the argument is pre-answered.

4. Notice. The affected Party shall give written notice to the other Party within seven (7) days of becoming aware of the Force Majeure Event, by email to the address in Clause [Notices] and by registered post, specifying the event, the obligations affected, the anticipated duration, and the steps proposed to mitigate. The affected Party shall provide reasonable updates at intervals of not more than fourteen (14) days, and shall give written notice within seven (7) days of the event ceasing.

Drafter’s note: This is the limb that decides most disputes. A fixed window, a prescribed dual mode, and mandatory content close off the “defective notice” defence that sinks more claims than the events list ever does. Keep the window tight and the mode specific.

5. Consequences: suspension then termination. On valid notice, the affected obligations are suspended for the duration of the Force Majeure Event, and timelines for performance are extended by the period of suspension. If the Force Majeure Event continues for more than ninety (90) consecutive days, either Party may terminate this Agreement on written notice, without liability for the suspended obligations, save for rights and obligations accrued before the date of the Force Majeure Event.

Drafter’s note: Suspension first, termination only at the long-stop. The 90-day figure is a convention, not a rule, so set it to the deal’s tolerance. The carve-out for accrued rights ensures termination does not wipe out what was already owed before the event.

6. Mitigation. The affected Party shall use reasonable endeavours to avoid, overcome or minimise the effect of the Force Majeure Event and to resume full performance as soon as reasonably practicable, and shall keep records of the steps taken.

Drafter’s note: Mitigation plus a record-keeping obligation. Courts ask whether the affected Party actually tried; the record-keeping line turns that question into documentary evidence the Party can produce.

7. Cost and payment during suspension. Each Party shall bear its own costs arising during the period of suspension. Payment obligations that accrued before the Force Majeure Event remain due and payable. [Optional: Payment obligations falling due during the suspension are themselves suspended and become due on resumption.]

Drafter’s note: Silence here is the most common omission and the most common fight. State expressly who carries standing costs and whether payments pause or continue. The optional sentence is the pro-supplier variant; delete it for a pro-buyer position.

8. Governing law, survival and severability. This Clause is governed by [the laws of India], and disputes under it shall be resolved by [the courts at ___ / arbitration seated at ___]. If any part of this Clause is held unenforceable, the remainder continues in effect. The notice obligations, accrued rights, and the dispute-resolution mechanism survive termination under this Clause.

Drafter’s note: Anchor the clause to a governing law and forum so a force majeure dispute does not generate a jurisdiction fight. Severability protects the rest of the clause if one limb fails; survival keeps notice, accrued rights, and dispute resolution alive after termination.

Two quick variants are worth keeping in your back pocket. The pro-supplier version widens the events list, uses “prevent, hinder or delay,” sets a longer long-stop, and suspends payment during the event, so the party more likely to be disrupted gets generous relief. The pro-buyer version narrows the events list, tightens the notice window, shortens the long-stop, and keeps payment obligations alive, so the party expecting delivery retains leverage and a faster exit. Which way you draft depends on which side of the table you sit, and a senior drafter tunes the same eight limbs to the client’s risk position rather than copying a neutral template.

Does this clause cover war or sanctions? Yes, expressly, in sub-clause 1(c), because those are exactly the named events a generic clause omits and a 2026 clause should not. The drafting judgement on display here, naming the events that matter, choosing the trigger standard deliberately, and building the exclusions that pre-answer the case law, is the difference between a clause that reads well and a clause that holds.

Drafting force majeure by contract type

A force majeure clause that works for a software subscription will not work for a power-purchase agreement, and a drafter who reuses one across contract types is leaving real risk on the table. The events that matter, the notice cadence, and the consequence cascade all shift with the deal. So how should the clause change depending on what the contract is for?

The building blocks of a commercial contract stay the same, but the force majeure events you name should track the actual disruption profile of the deal. Here is how the clause should flex across the four most common contract types in Indian practice.

For supply and manufacturing contracts, the events that bite are input shortages, port closures, and supplier failure up the chain. The drafting question that trips people up is whether a sub-supplier’s failure counts as your force majeure. By default it may not, because your supplier’s problem is not an event beyond your control in the way a flood is, so if you want upstream failure covered you must name it expressly. Does force majeure cover strikes or supplier failure? Only if you draft it to, and only with the careful exclusion that a strike confined to your own workforce does not qualify, which is the kind of precise commercial drafting every boilerplate provision rewards.

For construction and EPC contracts, the dominant mechanism is extension of time rather than termination, because the parties usually want the project finished, not abandoned. The drafting subtlety is the interplay between the force majeure EOT and the price-escalation clause: force majeure should buy you time, not money, and cost relief should come through a separate escalation or change-in-law provision. Mixing the two is a classic error that leaves the contractor under-compensated or the owner over-exposed. Keep time relief and cost relief in separate limbs.

For SaaS and technology agreements, the events list looks nothing like a traditional clause, and this is where drafting is evolving fastest. [FUTURE] Early signals from current Indian commercial practice suggest that cyberattack, ransomware, and cloud-provider or data-centre outage are becoming standard named force majeure events in technology contracts, alongside a tightening separation between force majeure (true impossibility, leading to suspension or termination) and hardship clauses (economic difficulty, leading to renegotiation). Practitioners expect climate events, port closures, and sanctions or export-control measures to be named expressly more often as geopolitical and supply-chain volatility persists. The drafting takeaway: a 2026 technology clause that does not name cyber and cloud events is already behind practice, and the interplay with uptime SLAs and service credits needs its own attention so the customer is not double-compensated or left without remedy.

For lease and power-purchase agreements, the recurring issues are change-in-law and fuel-supply disruption, and the key drafting move is to separate change-in-law relief from force majeure proper. A regulatory tariff change or a new statutory levy is usually better handled through a dedicated change-in-law clause that adjusts price, because, as Energy Watchdog made clear, a price or cost shock is not force majeure. Lumping change-in-law into the force majeure clause invites the very argument the Supreme Court rejected.

This is also where the distinction between force majeure and its cousins earns its keep. Force majeure is not the same as a hardship clause: force majeure addresses events that prevent performance and leads to suspension or termination, while a hardship clause addresses events that make performance economically punishing and leads to renegotiation. And force majeure is not the same as a material adverse change (MAC) clause, which lets a party walk away from a deal (typically before closing) if something materially worsens the target’s position. Sophisticated drafting keeps these three in separate provisions, because Energy Watchdog shut the door on running economic hardship through force majeure, and conflating them produces a clause that does none of the three jobs well. These are the other boilerplate clauses worth getting right, and they work best when each does one thing cleanly.

In practice, what experienced drafters know is that the contract type tells you the events list before the client does. Read the deal, map its disruption profile, and name those events. A common question is whether one master force majeure template can serve a whole contract portfolio. It can serve as a starting skeleton, but the events list and the notice cadence must be tuned per deal, and the pitfall is the one-size template that names floods and wars for a SaaS contract while ignoring ransomware. That clause is decorative, not protective.

How to invoke force majeure: a step-by-step playbook

A perfectly drafted clause is worthless if it is invoked badly, and most invocations are. The moment performance fails, the drafting work is done and the procedure begins, and the procedure is where claims are won or lost. So how do you actually invoke a force majeure clause without handing the other side a defence?

Follow five steps, in order.

1. Check the preconditions before you do anything else. Confirm three things: that the event is actually listed or genuinely caught by your clause, that it has genuinely prevented, hindered or delayed performance (not merely made it costlier), and that you were not already in default before the event arrived. This is the Halliburton checkpoint. If you were late before the lockdown, the lockdown will not save you, so be honest about the timeline before you commit to the invocation.

2. Serve the notice in the prescribed mode and within the prescribed window. Read your own clause and do exactly what it says: if it requires email plus registered post within seven days, send both within seven days. The notice must contain what the clause demands, typically the event, the obligations affected, the expected duration, and the mitigation you propose. How do you write a force majeure notice? Precisely and early. State the event, identify the clause sub-paragraph you rely on, specify which obligations are suspended and from what date, give an honest estimate of duration, and describe your mitigation plan. A vague or late notice is the most common reason a valid event still produces a failed claim.

3. Build the evidentiary record for causation, not inconvenience. The court or arbitrator will want proof that the event actually prevented performance, not merely that the event happened. What evidence supports a force majeure claim? Government orders and notifications, correspondence with suppliers and customers, internal records showing the operational impact, photographs or logs where relevant, and a clear timeline tying the event to the specific obligation you could not perform. Causation is the link most claimants fail to evidence, because they prove the event but not its effect on this contract.

4. Mitigate, and document that you mitigated. Take reasonable steps to work around the disruption, source alternatives, reschedule, deploy a contingency, and keep contemporaneous records of every step. The duty to mitigate is not optional, and a documented mitigation effort is frequently what tips a borderline claim in your favour. A party that sat on its hands and then invoked force majeure looks like a party using the clause as an excuse, which is exactly the impression you must avoid.

5. Update, resume, or trigger termination at the long-stop. Keep sending the contractual updates, resume performance the moment the event lifts and give the prescribed cessation notice, and if the event drags past the long-stop (the 90-day mark in most clauses), decide whether to invoke termination. Letting the long-stop pass without acting leaves both parties in limbo, so diarise the date when you serve the first notice.

What defeats an invocation? The same three failures, every time: late or defective notice, no causal link between the event and the non-performance, and a pre-existing breach. The notice point is illustrated by the steel-importers dispute in Standard Retail Pvt. Ltd. v. G.S. Global Corp Pvt. Ltd., 2020 SCC OnLine Bom 704, where the Bombay High Court refused to restrain encashment of letters of credit, holding that a force majeure clause in the supplier’s contract did not assist the buyers and that distribution had not genuinely been prevented by the lockdown. The mismatch between the clause relied on and the obligation in question, plus the absence of true prevention, sank the claim. The prior-default point is the Halliburton point from the opening story, and it is the most unforgiving of the three.

Can the other side reject your force majeure notice? They can dispute it, and frequently will, by arguing the event is not covered, performance was not genuinely prevented, you were already in default, or the notice was defective. That is precisely why steps one through four matter: a clean precondition check, a compliant notice, a documented causal link, and a recorded mitigation effort leave the counterparty far less room to reject. The invocation is not a single letter; it is a process you run to make rejection as hard as possible.

Landmark force majeure cases every drafter should know

A drafter who knows the cases drafts differently from one who does not, because each landmark decision tells you exactly which argument a court will accept and which it will reject. These are the rulings that shaped how Indian courts read force majeure clauses, and each one changed something specific about how the clause should be drafted. What did the cases actually decide, and what should you do about it?

The doctrinal anchor is Energy Watchdog v. Central Electricity Regulatory Commission, (2017) 14 SCC 80. The Supreme Court held that where a contract contains a force majeure clause, relief is governed by that clause under Sec 32 and not by the general frustration doctrine; that a mere rise in price or more onerous performance is neither force majeure nor frustration; and that force majeure clauses are construed strictly. What did Energy Watchdog decide for drafters? It told you that your clause occupies the field, that economic hardship is not your route to relief, and that the court will hold you to the precise events you listed. Every drafting choice in this guide traces back to that ruling.

The frustration standard comes from Satyabrata Ghose v. Mugneeram Bangur & Co., AIR 1954 SC 44, the 1954 decision that defined “impossible” in Sec 56 as impracticability and futility of the contract’s object rather than literal physical impossibility. This is the standard that applies when there is no clause, and understanding it tells you why drafting a clause is so valuable: the Satyabrata Ghose test is narrow, hard to satisfy, and decided case by case, whereas a Sec 32 clause gives you defined relief. The case answers the perennial question of what the test for impossibility actually is, and the answer is “impracticability, read narrowly.”

COVID produced the two High Court decisions that every contemporary drafter cites. Halliburton Offshore Services Inc. v. Vedanta Limited, 2020 SCC OnLine Del 542 established that a lockdown can be a force majeure event but cannot excuse a default that pre-dated it, and it raised the force-majeure-versus-bank-guarantee question that most templates still ignore. Standard Retail Pvt. Ltd. v. G.S. Global Corp Pvt. Ltd., 2020 SCC OnLine Bom 704 established that a force majeure clause in one party’s contract does not assist a different party, and that a court will not restrain encashment of a letter of credit where performance was not genuinely prevented. Does a government lockdown automatically excuse non-performance? These two cases answer with a flat no: the lockdown is a fact, not a free pass, and its effect depends on timing, on the exact clause relied on, and on genuine prevention. Is a pandemic automatically force majeure for every contract? Again no, for the same reasons.

The “Act of God” distinction comes from Divisional Controller, KSRTC v. Mahadeva Shetty, (2003) 7 SCC 197, where the Supreme Court defined an act of God (vis major) as a natural, unforeseeable event occurring without human intervention. Is force majeure the same as Act of God? No. Act of God is a narrow common-law category of natural events, while contractual force majeure is whatever your clause defines it to be, which can be far wider, covering strikes, government action, and cyberattack that no act-of-God doctrine would reach. The drafting lesson is that you should not rely on the act-of-God concept to cover man-made disruptions; name them in the clause instead.

These cases are not clustered by accident in this guide; they are distributed across the sections where each does its work, because a drafter needs Energy Watchdog when choosing the trigger standard, Naihati Jute when excluding cost escalation, Halliburton when drafting the prior-default exclusion, and Standard Retail when thinking about notice and matched clauses. In practice, what experienced drafters know is that the case law is a drafting checklist in disguise: each ruling is a hole someone fell into, and a well-built clause is one that fills every hole the courts have already identified.

A common question is whether COVID “counts” as force majeure, full stop. The cases show the question is wrongly framed. COVID could be a force majeure event where the clause covered it, performance was genuinely prevented, and there was no prior default, and it failed exactly where one of those conditions was missing. The pitfall is treating any of these landmark rulings as a slogan (“COVID is force majeure,” “price rise is never force majeure”) rather than as a fact-specific holding that your drafting must account for.

Force majeure drafting checklist (quick reference)

This is the printable takeaway: a fast, scannable checklist you can run against any force majeure clause before you sign it. Use it as a pre-flight check on a clause you drafted or, just as usefully, on a clause the counterparty sent you.

Pre-drafting checks

• Identify the contract type and map its actual disruption profile (supply, EPC, SaaS, lease/PPA).
• Confirm whether you sit on the supplier side or the buyer side, and draft to that risk position.
• Decide what belongs in force majeure versus a separate hardship, change-in-law, or escalation clause.

Clause-build checks (the eight limbs)

• Limb 1: an express, specific events list naming every foreseeable disruption for this deal.
• Limb 2: a catch-all, understood as a backstop limited to events of the same genus.
• Limb 3: a “prevent, hinder or delay” trigger standard, not “render impossible,” with partial-relief wording.
• Limb 4: notice mechanics with a fixed window, prescribed mode, mandatory content, and update obligation.
• Limb 5: suspension first, extension of timelines, then termination at a defined long-stop (commonly 90 days).
• Limb 6: an express duty to mitigate, with a record-keeping obligation.
• Limb 7: cost allocation and payment treatment during suspension, stated expressly.
• Limb 8: governing law, jurisdiction or seat, severability, and survival.

Pre-signature review checks

• Add exclusions for self-induced events, mere cost increase, and pre-existing default (the case-law armour).
• Confirm the notice limb is at least as tight as the events list, because notice is where claims fail.
• Check that change-in-law, hardship, and MAC are handled in their own clauses, not folded into force majeure.
• Verify the clause anchors to a governing law and forum so a dispute does not spawn a jurisdiction fight.

Run that list and you have caught the failures the courts have already mapped. Skip it and you are relying on a clause that has never been tested, which is the most dangerous kind.

Frequently asked questions about force majeure clause drafting

1. What is force majeure under the Indian Contract Act, 1872? Force majeure is not a defined term in the Indian Contract Act, 1872. It is a contractual mechanism that allocates the risk of disruptive events, and it operates through Section 32 (contingent contracts) when a clause exists. Where there is no clause, the related statutory doctrine of frustration under Section 56 may apply instead.

2. Which sections of the Indian Contract Act govern force majeure? Two sections do the work. Section 32 governs a contract that contains a force majeure clause, treating performance as contingent on the defined event. Section 56 governs frustration where there is no clause or the clause does not cover the event. The Supreme Court confirmed this division in Energy Watchdog v CERC.

3. Is force majeure the same as “Act of God”? No. An Act of God (vis major) is a narrow category of natural, unforeseeable events occurring without human intervention. Contractual force majeure is whatever your clause defines, and it usually covers far more, including war, government action, strikes, and cyberattack. Act of God is a subset of what a well-drafted force majeure clause can reach.

4. What should a force majeure clause include (a quick checklist)? At minimum: a specific events list, a catch-all, a “prevent, hinder or delay” trigger, notice mechanics, a suspension-then-termination cascade, a duty to mitigate, cost and payment treatment during suspension, and governing law with survival. Add exclusions for self-induced events, mere cost increase, and pre-existing default. Each limb answers a specific way the clause could otherwise fail.

5. What is a sample force majeure clause format for an Indian contract? A defensible format runs in numbered sub-clauses: defined events, catch-all, exclusions, notice, consequences (suspension then termination), mitigation, cost and payment, and governing law. The annotated model clause earlier in this guide gives a full India-tuned version with drafter’s notes for each sub-clause. Adapt the events list and the long-stop to your specific deal.

6. Should I list specific events or use general catch-all wording? List specific events first, then add a catch-all as a backstop. Indian courts construe force majeure clauses strictly and read catch-alls narrowly under the ejusdem generis principle, so anything you do not name may be excluded. The specific list does the real work; the catch-all only covers genuinely novel events of a similar character.

7. What is the notice period for invoking force majeure in India? There is no statutory notice period. The period is whatever your clause specifies, commonly seven days from awareness of the event, with ongoing updates. Because late or defective notice is a leading reason claims fail, the contractual window and mode must be followed exactly. If the clause is silent, serve notice as promptly as possible and document it.

8. How do you invoke a force majeure clause? Check the preconditions (covered event, genuine prevention, no prior default), serve a compliant notice in the prescribed mode and window, build an evidentiary record proving causation, mitigate and document it, and then update, resume, or terminate at the long-stop. Each step closes off a defence the counterparty might otherwise raise. The notice is the most important document in the process.

9. What happens to obligations during a force majeure event, suspended or terminated? A well-drafted clause suspends the affected obligations first and extends timelines, rather than terminating immediately. Termination becomes available only if the event continues past a defined long-stop. This sequence stops a temporary disruption from becoming a permanent escape from the contract, while still giving both parties an exit if the event persists.

10. How long can a force majeure event last before termination (the 90-day rule)? There is no fixed legal rule; 90 days is a common contractual long-stop, but 60 or 180 days are also used depending on the deal. After the long-stop passes, the clause typically lets either party terminate without liability for the suspended obligations. The exact figure should track how long the parties can commercially tolerate the disruption.

11. What happens if I give the force majeure notice late? A late notice can defeat an otherwise valid claim, because Indian courts treat the notice requirement as a substantive condition, not a formality. If your clause makes timely notice a precondition to relief, missing the window may waive the claim entirely. This is one of the most common reasons force majeure invocations fail.

12. Can both parties invoke force majeure, or only one? Either party can invoke it, provided the event genuinely prevents that party’s own performance. Force majeure is invoked by the party whose obligations are disrupted, so in a contract where both sides have performance obligations, either could face a qualifying event. A party cannot, however, invoke the other side’s disruption as its own excuse.

13. Force majeure clause vs the doctrine of frustration (Sec 56), which applies? If your contract has a force majeure clause covering the event, the clause governs under Section 32 and frustration does not apply. If there is no clause, or the clause does not reach the event, you fall back on frustration under Section 56. The Supreme Court drew this line in Energy Watchdog v CERC: the clause displaces the doctrine.

14. Can I rely on force majeure if my contract has no force majeure clause? Not as such, because force majeure is a contractual creature. Without a clause, your only route is the statutory doctrine of frustration under Section 56, which requires you to show performance became impracticable in the Satyabrata Ghose sense. That is a much harder standard than invoking a well-drafted clause, which is why drafting one matters.

15. Is economic hardship or a price rise force majeure? No. A price rise, currency movement, or more onerous performance is not force majeure or frustration under Indian law. The Supreme Court held in Energy Watchdog v CERC, consistent with Naihati Jute Mills, that more expensive performance is not excused. Cost escalation should be handled through a separate price-adjustment or hardship clause, not force majeure.

16. Does COVID-19 count as force majeure in India? It depends on the clause, the timing, and genuine prevention. COVID could be a force majeure event where the clause covered pandemics, performance was actually prevented, and there was no prior default, as the COVID-era High Court decisions showed. It failed wherever the party was already in breach or performance was merely harder, not prevented.

17. Does a force majeure clause cover war or sanctions? Only if you name them. A generic clause that lists floods and fires will not cover war, armed conflict, or sanctions unless those are expressly included. Given current geopolitical and supply-chain volatility, a 2026 clause should name war, terrorism, and sanctions or export-control measures expressly, because strict construction excludes what is not listed.

18. Can force majeure be invoked against a bank guarantee or LC encashment? Generally no, and this caught many parties out during COVID. In the Halliburton and Standard Retail disputes, courts declined to restrain encashment of bank guarantees and letters of credit on force majeure grounds, treating these instruments as independent of the underlying contract. If you want force majeure to affect security encashment, you must address it expressly, and even then courts are reluctant.

References

Case Law

1. Divisional Controller, KSRTC v. Mahadeva Shetty, (2003) 7 SCC 197. Parallel citation: AIR 2003 SC 4172 (Supreme Court of India).
2. Energy Watchdog v. Central Electricity Regulatory Commission, (2017) 14 SCC 80. Parallel citation: AIR 2017 SC (Supp) 43 (Supreme Court of India).
3. Halliburton Offshore Services Inc. v. Vedanta Limited, 2020 SCC OnLine Del 542. Delhi High Court (order dated 29 May 2020).
4. Naihati Jute Mills Ltd. v. Khyaliram Jagannath, AIR 1968 SC 522. Parallel citation: 1968 SCR (1) 821 (Supreme Court of India).
5. Satyabrata Ghose v. Mugneeram Bangur & Co., AIR 1954 SC 44. Parallel citation: 1954 SCR 310 (Supreme Court of India).
6. Standard Retail Pvt. Ltd. v. G.S. Global Corp Pvt. Ltd., 2020 SCC OnLine Bom 704. Bombay High Court (order dated 8 April 2020).

Statutes

1. The Indian Contract Act, 1872. Sections cited: 32 (contingent contracts), 56 (frustration / agreement to do impossible act), 65 (restitution).

This article is for informational purposes only and does not constitute legal advice. For specific legal guidance, consult a qualified legal professional.