Termination Rights: When Either Side May Walk Away

A termination right is a route out of a signed transaction that the document itself created. The paper on this transaction carried four of them: a condition still unsatisfied when the stated outside date arrives, a promise in the document not kept, a defined adverse change, and agreement between both sides to stop. Every one exists only between signing and completion, and all four end on the day completion happens.

Three things settled earlier hold this up. Signing and completion are different days, separated on this transaction by a conditions period of nine weeks. A condition to completion is a gate rather than a promise, so failing one is not the same as breaking one. And a fee attached to a walk-away prices the decision rather than preventing it. Placed on a calendar, those three settle almost every question about walking away.

Where does a right to walk away actually come from?

A right to walk away comes from the document, and from nowhere else a reader can reach. The rule sounds obvious written down, and it is not how most people read a termination clause. The instinct is to treat the clause as a summary of something larger sitting behind it, a set of general permissions that the paper happens to write out. There is no such thing behind a signed transaction. A signed transaction carries no general permission to change one's mind, so everything that can be established about walking away is in the paper itself.

Think about a wedding hall booked eleven months ahead. The booking form names two situations in which the money comes back: the hall becomes unusable, or both sides agree to cancel. Nothing else. When the couple decide in month seven that they would rather have the function somewhere else, they discover that changing their mind is not one of the two. The form did not fail them and nobody misled them. The couple simply read a short list as though it were an example of a longer one.

A transaction document behaves the same way at a thousand times the size. Harivansh Packaging Limited signed a document to acquire Sundarban Polymers Private Limited, and everything either side could later do to stop that acquisition was written down in it before anybody signed. Devyani Kulkarni, the chief financial officer of Harivansh Packaging Limited, was not choosing between commercial options once the ink was dry. She was choosing among routes that had already been drafted, priced and allocated to a side.

The skill required here is therefore reading rather than law. An associate handed a document of two hundred printed sides and asked to summarise the termination position is not being asked what the law allows. The associate is being asked to find the clause, list what it created, and say who holds each item on the list. The job is finite and checkable. Whether any of those routes would survive an argument is a completely different question, and it belongs to somebody with a practising certificate looking at the actual paper.

One consequence follows immediately, and it governs every route set out below. If the clause is silent, the answer is silence. A reader who cannot find a walk-away route in the document has not read badly. The reader has found the answer, and the answer is that this document created none.

Which doors out did this document carry?

Four, and they are four doors rather than four settings of one door. The distinction matters more than it looks. A reader who blurs the four doors ends up describing a transaction as more or less terminable, and no document says that. Each route has its own trigger, its own evidence and its own holder, and one of them being unavailable says nothing about the other three.

The first route is a condition to completion still unsatisfied when the stated outside date arrives. Nothing has gone wrong, nobody has misbehaved, and the transaction has simply run out of calendar. The second is a breachDoing, or failing to do, something the signed document required. A breach is judged against the document, not against what either side hoped would happen. by the other side, needing a promise in the paper and evidence that the promise was not kept. The third is a defined adverse change, and the word defined is carrying the load: the document sets out what counts, and something that feels catastrophic but sits outside that definition does not open this door. The fourth is agreement between both sides to stop, and that route needs no trigger at all.

Because these are separate doors, a reader should be able to name which one is being used in any story about a transaction falling over, and being unable to name it usually means the story has not said. Coverage of an abandoned transaction tends to describe atmosphere: talks collapsed, the buyer got cold feet, the parties could not agree. None of that names a door. A report that a transaction ended after a regulator took longer than expected is describing door one. A report that both sides issued a short joint statement and said nothing further is almost always describing door four.

Route four deserves a second look because it is the one most readers skip. The agreement route has no trigger, no evidence and no argument attached, and it is available on any day between signing and completion for any reason at all. Two sides who no longer want to proceed have every incentive to stop cheaply rather than to construct a case under one of the other three, so agreement is also how the large majority of abandoned transactions actually end. The routes that generate the interesting arguments are the ones that get written about; the route that gets used most produces a two line statement and nothing else.

Why does every one of these routes close on the day of completion?

Because after completion there is no longer a transaction to stop. The money has moved, the shares have transferred, and the thing a termination right acts on has finished happening. The routes do not weaken, become harder to use, or start needing a better argument. All four are gone, and they go together, on one morning.

Look at the shape of the calendar rather than at the clause. Before signing, this transaction had no document, so it had nothing to terminate. In its place stood a term sheet and an exclusivityA promise that keeps one side of a negotiation talking to this counterparty and no other, for a stretch of calendar the paper names. Exclusivity restricts conversations, not the transaction itself. promise, and both restrict conversations rather than create routes out of anything. After completion, the document has done its work and the routes have expired with it. In between sits one window, and every walk-away right this document created lives inside it.

From the morning of completion the only route left is a claim under the warranties and the indemnities, filtered by the limits the document set, and the walk-away rights are simply not there any more. Those are two very different kinds of protection. Walking away means the transaction does not happen. Claiming means the transaction happened, the buyer owns the thing, and the buyer is now trying to recover money for something that turned out not to be as promised. A reader who has not registered the handover has misread where the risk sits for the entire period after completion.

What money actually moves when a transaction ends early?

Usually nothing at all, and where something does move it is because the paper created that movement in advance. Intuition says a collapsed transaction must leave a mess of money to sort out. It does not. Nothing has been paid, so nothing has to be unwound. Both sides have spent real money on advisers, on travel and on the time of their own people, and each of them keeps its own bill.

The one transfer that can happen is the one the document constructed. On this transaction that is a break feeA sum the target side agreed to pay the buyer if the transaction ends because of something on the target side. of Rs 22.74 crore payable if it is the seller side whose decision ends the transaction, and a reverse break feeThe same idea pointing the other way: a sum the buyer agreed to pay the target side if the transaction ends because of something on the buyer side. of Rs 45.48 crore payable if it is the buyer whose decision ends it. Both are constructed for this reading order against one base, the adjusted equity valueThe price the sellers ended on, once two agreed recalculations at the very end of the transaction had each moved it a little. of Rs 1,137 crore.

Every rupee figure attached to this transaction hangs from that base. The target earnings are Rs 132 crore. Ten times that is an enterprise value of Rs 1,320 crore. Taking out the target borrowings of Rs 180 crore leaves a headline of Rs 1,140 crore, the figure the sellers were first told they would receive. Two recalculations then ran at the very end: a working capital check that moved the price up by Rs 12 crore, and a borrowings check that moved it down by Rs 15 crore. Their net is minus Rs 3 crore, and Rs 1,137 crore is where the price stopped.

The four amounts worked through here are all slices of that one figure. The break fee is one fiftieth of it. The reverse fee is one twenty fifth. The cash held back in escrow after completion is one tenth, or Rs 113.70 crore. The ceiling on everything the buyer can ever recover is one fifth, or Rs 227.40 crore. Two more thresholds sit underneath: one hundredth is the basket of Rs 11.37 crore, and one thousandth is the de minimis of Rs 1.14 crore.

A terminated transaction leaves both sides holding their own costs and neither of them holding the business, and the constructed fee is the only transfer unless the document created another one. Say that out loud on a live transaction and somebody will ask whether the fee compensates the disappointed side. The fee does not compensate anybody in any complete sense. The fee prices one particular way of ending, at a number agreed in advance by people who could not know what the walk-away would eventually cost.

Is a condition failing the same thing as terminating?

No, and this is the merge readers make most often. A condition that is not satisfied is a state of the world. A termination right is a decision that somebody takes. The state and the decision usually arrive together, they get described in the same sentence, and the document frequently requires the decision even where the state is beyond argument.

Take the regulatory approval on this transaction. Suppose it simply has not arrived. Completion cannot happen on the agreed terms, and nobody chose that; it is a fact about the world, as much a fact as the weather. Nothing that follows is automatic. Somebody has to reach for the route, and the document can require them to give formal written noticeA formal message one side sends the other because the document requires it in writing, rather than a conversation between two deal teams. before anything ends. Until that happens, the transaction is still alive, still binding, and still capable of completing if the approval lands next week.

One of these two is a state of the world and the other is somebody acting on it, and the paper very often requires the second even where the first has already occurred. There is a practical reason for the design. If an unsatisfied condition ended a transaction on its own, then a delay of one afternoon would kill something both sides still want. Requiring an act leaves the choice with the people who have the commercial interest in it, and lets a transaction survive a slow week that nobody minds.

There is a second reason worth spelling out. Requiring a positive act creates a record. Once a written notice has gone out over the signature of Ashwin Rege, the transaction lead for Harivansh Packaging Limited, a dated document exists saying which route was used and by whom. A transaction that simply stopped, with each side privately believing a different thing had happened, is the worst possible input into whatever conversation comes next.

What is the outside date doing behind all of this?

The outside date is the clock. A long-stop date, to give it the name a document usually gives it, is a date after which a condition still sitting unsatisfied converts into a right to walk. Before that date, an open condition is just a wait. After it, the wait has become somebody's choice.

One dated sentence is what stops a transaction from staying open indefinitely, and that is the only reason anybody drafts an outside date. Without it a conditions period has no end. A regulator takes as long as it takes, a counterparty takes as long as it takes, and a seller who signed in March finds itself in October still bound, still unable to talk to anybody else, and still with no mechanism for saying that enough time has gone by. The outside date is how a document converts patience into a right.

Two things about the outside date are settled elsewhere. How an outside date is set, extended and argued about is covered separately. The second matters more for anybody checking these figures: this transaction record publishes a conditions period of nine weeks and does not publish an outside date at all, so no outside date is available to work with.

What does a reader check in a termination clause?

Five things, in this order, and none of them takes long once the reader knows to look. Which routes exist. Who may use each one. Whether formal notice is required before a route takes effect. The fee attached to each route, if any. And where the outside date sits against the conditions period. A date that falls before the conditions could realistically be met is a very different clause from one that falls after.

Of those five, the second one carries most of the information. A clause can list four routes and hand one side three of them, and the list on its own will never show that. Reading the routes gives the shape of the exit. Reading who holds each one gives who actually controls it.

A route available to one side only is the most informative line in a termination clause, and it is the line a reader summarising the document is most likely to leave out. Try the test on the drawing above. Both panels contain the same four routes. The left one reads as a balanced arrangement in which either side can get out on four grounds. The right one shows that one of the four sits with the buyer alone. The seller has three routes, the buyer has four, and the buyer additionally holds the route that depends on a definition the buyer negotiated.

Two smaller checks are worth the thirty seconds. First, whether a route requires notice. A route that operates without notice can end a transaction while one side is still preparing a response. Second, what happens to any fee on each route. A document can perfectly well give a side a route out and attach a fee to using it, and that leaves the side holding a right and a bill rather than a right.

Where do the routes sit on this transaction's own calendar?

Put everything on the twenty two elapsed weeks this transaction actually ran, from term sheet to completion, and the shape stops being abstract. Weeks 0 to 13 are before signing. Week 13 is signing. Weeks 13 to 22 are the conditions period. Week 22 is completion. Twenty two weeks is the whole calendar, and it belongs to this one acquisition.

Before week 13 there is no signed document. Nothing can be terminated because nothing has been created, and what stands in place of protection is a term sheet and an exclusivity promise that restricts who the seller may talk to. The stretch runs to 59.09 per cent of the whole calendar, and across all of it the word termination has nothing to attach itself to.

From week 13 to week 22 every route is live. The window is nine weeks, or 40.91 per cent of the calendar, and it is the only stretch in which any of them exists. Three conditions to completion sit inside that window. Two counterparties whose contracts change hands each have to give consentA yes from somebody who is neither buying nor selling, whose agreement the transaction needs because a contract they hold is moving to a new owner.. A regulator has to approve. And the third one is a negative: nothing meeting the document's own definition of an adverse change may have occurred. The requirements of an approval regime are settled by the authority that runs it.

From week 22 onwards every route is closed and the buyer's position rests entirely on the warranty package: a de minimis of Rs 1.14 crore, a basket of Rs 11.37 crore, a cap of Rs 227.40 crore, and Rs 113.70 crore held in escrow.

The two protections hand over rather than fade: the walk-away rights run for nine weeks and end on one morning, and the money protection that replaces them starts on that same morning and runs on a completely different clock. A reader who has not seen the handover will read a post-completion problem as though the buyer still had somewhere to go, and there is nowhere.

Nine weeks against eighteen months, and why that pair does not work?

One sentence about these two durations gets repeated more than any other. The walk-away rights live for nine weeks and the money obligations live for eighteen months. It sounds like a finding. The sentence compares two durations, puts a number on each side, and lands on a satisfying contrast. The two numbers are also sized against two different measures, and once those measures are named the sentence stops being usable.

Nine weeks is how long a right exists. Eighteen months is how long a pot of cash sits in an account. A right and a pot of cash are not the same kind of thing. The escrow is a funding arrangement, not a liability window: it says that Rs 113.70 crore is parked where the buyer can reach it for a while, and it says nothing whatever about when the obligation to answer for a broken warranty begins or ends. The escrow hold is a fact about money, the nine weeks is a fact about a right, and putting them either side of an and makes them look like two halves of one measurement when they are not.

Both halves genuinely publish one measure, elapsed weeks on this transaction's own calendar, and both can be rebuilt on that. The routes run from week 13 to week 22, so nine weeks. Eighteen months restated on the same measure is 78 weeks. The escrow release therefore falls at about week 100 counted from the term sheet. The escrow cash therefore sits 8.67 times as long as the routes ran, and 3.55 times the length of the entire transaction from first term sheet to completion.

The third bar, the dashed one, is why the rebuild was worth doing. Once the two publishable durations are on one axis, the duration a reader actually wanted is visibly missing. How long the buyer may still bring a claim against the warranties is not stated anywhere in this transaction record. In the original sentence the escrow hold was standing in for that missing duration, and it should not have been. Money coming out of an account and a right coming to an end are two separate events, and a document sets them separately.

The gap matters practically. Assume the escrow hold is the claim window and the conclusion is that everything is settled at week 100 and the buyer can close the file. If the claim window is longer, the buyer can still claim after the money has gone back, and the cap of Rs 227.40 crore is what governs, not the Rs 113.70 crore that was sitting in escrow. If it is shorter, the last stretch of the hold is protecting nothing. Which of those holds cannot be told from anything published here, and the correct thing to write in a summary is that the document does not say.

How does a lender or an analyst actually use any of this?

Three readers use a termination clause in three different ways, and none of them is drafting anything.

Start with the lender behind the buyer. Harivansh Packaging Limited put Rs 140 crore of cash from its own balance sheet into this acquisition and borrowed a further Rs 1,000 crore at a contracted rate of 9.0 per cent. The lender committed that money weeks before it was drawn, and between commitment and drawdown sits exactly the nine week window described above. The lender is not after a legal opinion. The lender wants a list: which events could send this transaction away before drawdown, who controls each of them, and whether the borrower keeps any obligation to the lender if one is used. A route the borrower controls is a very different credit question from a route a regulator controls.

Next the analyst covering the listed acquirer. Harivansh Packaging Limited trades at Rs 300/- on an illustrative basis, giving a market capitalisation of Rs 5,400 crore, and its share register is 58.0 per cent promoter and promoter group with the remaining 42.0 per cent held by everybody else. Between announcement and completion the analyst has to hold a view on whether the acquisition happens. The termination clause converts a vague question about deal risk into a countable one, and that makes it the cheapest input available: how many routes exist, how many of them require somebody else agreeing, and how many require nothing more than one side deciding.

A route that depends on a third party is a risk the transaction carries, and a route that depends on one side deciding is a risk that side controls, and mixing the two produces a completely wrong picture of how likely a transaction is to complete. An analyst who counts four routes and calls the transaction risky has not done the work. An analyst who separates the ones needing an approval from the ones needing a decision has.

Third, the same skill at a size everybody has met. An agreement to buy a flat names the situations in which the money comes back and the sale stops. The exit section repays reading before the price. If the only exits are the seller failing to produce clean title and both sides agreeing to cancel, then the buyer's loan not coming through is not an exit, and the deposit is at risk in a situation most buyers would assume was covered. Same clause, same reading order, four fewer zeros.

When the marker moves slowly through week 22 the panel does not fade. Because the whole nature of the protection changed on that one step, four dark chips go pale together and the money bar jumps from the small constructed fee to the Rs 113.70 crore escrow.