Term Sheet or Definitive Agreement: What Each Fixes

A term sheet states the shape of a transaction in a handful of printed sheets and mostly does not bind. A definitive agreement states the obligations in full and binds from signature. The term sheet exists to make disagreement cheap; the agreement exists to make obligations enforceable. On the Harivansh Packaging Limited purchase, thirteen weeks of the twenty two ran between them.

What is each of these two documents for?

Start away from transactions altogether. A family is arranging catering for a wedding. Over tea the family and the caterer settle the thing in ten minutes: four hundred guests, three counters, a fixed price per plate, service from seven in the evening. Everybody shakes hands and everybody feels the matter is closed. Then somebody sensible ruins the afternoon by asking whether all of it should be written down properly.

And the questions begin. What if four hundred and sixty people turn up? What if the venue moves the kitchen slot by two hours? Who pays if the gas cylinders do not arrive? If the food is late, does the price change, and by how much? None of those questions was answered over tea. Worse, none of them was even asked. The handshake was about the shape of the arrangement and not about the arrangement itself.

The catering handshake is the whole of the distinction in one image. The two documents are not two drafts of one document: the first settles the shape of the arrangement, and the second settles what happens when the arrangement meets a world that does not cooperate.

A term sheetA short document setting out the main commercial points of a proposed transaction, produced early so that both sides can see whether they agree before detailed work begins. is the handshake written down. On the Harivansh Packaging Limited purchase of Sundarban Polymers Private Limited, it named the price and the basis it was struck on, what was being bought, what had to happen before completion, and roughly when. The term sheet ran to a handful of sheets. Its job was to expose disagreement while disagreement was still cheap, and it did that by being short enough that both sides could read all of it in one sitting and argue about the parts they did not like.

A definitive agreementThe long, fully binding contract that the parties sign, setting out every obligation each side takes on and what happens if any of them is not met. is the built thing. The agreement is what the parties actually sign, and from the moment of signature it decides what each side must do, what each side has promised, and what either side may do if the other lets them down. The document is long, and the length is not decoration.

The most useful way to hold the pair apart is to notice that they answer different questions. The term sheet answers: is there a transaction here worth spending money to build? The agreement answers: now that it has been built, what exactly is each side on the hook for? A term sheet judged as a bad agreement has been judged against a test it was never sitting. A term sheet is supposed to be silent on most things. The silence is the point. A document that tried to settle everything would take three months to produce, and it would be produced before anybody knew whether the two sides could even agree a price.

Which document binds, and from when?

Everybody asks about binding force first, and the honest answer has two halves.

The term sheet on the Harivansh Packaging Limited transaction was bindingA term is binding when a court could be asked to enforce it. A term that merely states what the parties intend to do is not binding, whatever it says about intention. on four of its clauses and stated intention on all the rest. The four binding clauses were confidentiality, exclusivity, who bears their own costs, and which law governs the document. Which clauses bind is a drafting choice made document by document. Another term sheet could bind on more of its terms, or on fewer, and no rule fixes the number at four.

Everything else in it, and that includes the price, the structure, the conditions and the timetable, was a statement of what the two sides currently intended to do. Intending is not promising. If Harivansh Packaging had walked away the week after the term sheet was agreed, the seller would have had a grievance, a wasted month and a story to tell, but nothing on the price to enforce.

The definitive agreement is different in kind rather than in degree. All of it binds, and it binds from signature. There is no island of enforceable clauses floating in a sea of intention. Every sentence in it is an obligation, a definition that feeds an obligation, or a mechanism that measures one.

The practical reading is the one worth carrying: after a term sheet either side can walk with a bruised standing, and after signature either side walking is a breach unless a term of the agreement lets them walk. The qualification matters. Agreements do contain doors. The doors are drafted, though, with a frame and a handle and conditions attached, rather than the open field a term sheet leaves.

Signature changes the cost of changing one's mind. Before signature, a change of mind costs money already spent and goodwill already used. After signature, a change of mind costs whatever the agreement says it costs, and the agreement was written by two sets of advisers who were each imagining exactly that moment.

Why is a definitive agreement so much longer than a term sheet?

Go back to the caterer. The handshake fitted into ten minutes because it described a happy world: four hundred people arrive, the food arrives, everybody eats, everybody pays. Write the contract, though, and almost every line is about a world in which something has gone wrong. The length comes from the unhappy versions, not from the happy one.

The length of a definitive agreement is a function of unhappy futures rather than of how complicated the business is. Sundarban Polymers Private Limited makes flexible packaging films on revenue of Rs 880 crore. Flexible packaging film is not a complicated business. Sundarban Polymers buys polymer, runs extrusion and printing lines, and sells to converters and brand customers. The whole operation could be described on two sheets. The agreement that transfers it runs far longer than two sheets, and the extra length has almost nothing to do with polymer.

Every obligation in the agreement has to keep working in the case where the two sides have stopped cooperating. Surviving the breakdown is the design constraint. While everybody is friendly, a vague clause is harmless. The parties will simply sort it out over a phone call. The clause has to survive the phone call not being answered. So each obligation acquires a definition, a measurement, a deadline, a consequence and a tie-breaker, and each of those is a sentence, and the sentences add up.

The shape of where the length comes from is easy to see. Take six ordinary things that could go wrong between signing and completion on this purchase, and notice that every one of them forces the agreement to say something it would never need to say if the two sides simply stayed friendly.

What does each document actually fix?

Here is the cleanest division to carry away. The term sheet fixes the transaction, and the agreement fixes what happens when the transaction meets reality.

The term sheet on this purchase fixed four things. Price, and the basis on which it was struck: an enterprise value of Rs 1,320 crore, being ten times the Rs 132 crore of earnings before interest, tax, depreciation and amortisation (EBITDA) that Sundarban Polymers Private Limited earned. Structure: a purchase of one hundred per cent of the shares, cash-free and debt-free, with net debt deducted at completion. The conditions: what had to be true or obtained before the transaction could complete. And the timetable: roughly when each of those things should happen.

Every one of those is a statement about the deal itself. Read together, they establish what is being bought, for how much, on what basis and by roughly when. The four fix nothing about what happens if any of it turns out to be untrue.

The definitive agreement fixes that second set. Each side promises that certain things about the business are true. A promise that turns out untrue carries a consequence, and the agreement names it, including who pays and up to what limit. Each side has conduct required of it in the stretch between signing and completion. Either side may walk away, but only in the circumstances the agreement lists. And, crucially for the arithmetic, the agreement sets out how the price is finally measured. Measuring a price is a different question from stating one.

The last pair is worth sitting with. The term sheet said the price would be adjusted for working capital. The agreement defined working capital, said when it is counted and who counts it, said what happens if the two sides disagree about the count, and named the figure it is measured against. The term sheet had a policy. The agreement had a procedure.

The agreement's contents can be named without opening it. The seller's actual promises, the limits set on those promises and the moment either side could walk are covered under reading a definitive agreement. Here it is enough to know that those things exist, and that they exist only in the second document.

What happens in between, and why does it matter more than either document?

Most explanations of this pair treat the gap between them as dead time in which lawyers type. Treating the gap as dead time is exactly backwards. Agreement comes before drafting, and the drafting is shaped by findings that did not exist when the shape was agreed.

Between the term sheet and signing, confirmatory diligenceThe detailed checking a buyer does after the main commercial terms are agreed, to confirm that what it was told about the business is actually so. runs. The buyer's advisers go through the target's contracts, its receivables, its tax position, its licences, its disputes and its people. The advisers are not looking for reasons to reprice the transaction; the price was set at the term sheet. The check is whether the company described to them is the company that exists.

Whatever they find has to go somewhere, and the only place it can go is the agreement. A finding does not usually reopen the price. A finding changes what the agreement has to protect against. Suppose diligence discovers that one customer contract at Sundarban Polymers Private Limited is due for renewal shortly after completion. The renewal did not exist for the term sheet. The term sheet was written before anybody looked. The renewal exists now, and the agreement has to decide what to do about it: a promise from the seller, a condition, a specific protection, or nothing at all with the buyer taking the risk knowingly.

The ordering runs one way and cannot run the other, for two reasons. Protections cannot be drafted against facts that have not yet been found. Nor is there any sense in finding facts before it is known whether there is a transaction worth spending diligence money on. So the sequence is: agree the shape cheaply, then spend money finding out, then draft against what the spending found.

On this transaction, twenty two weeks ran from term sheet to completion, and the conditions period after signing accounted for nine of them. Subtract and thirteen weeks ran from the term sheet to signature. The thirteen weeks are where confirmatory diligence and draftingThe work of turning agreed commercial points into contract language, where each point acquires a definition, a measurement and a consequence. both happened.

The thirteen weeks do not split cleanly between diligence and drafting. The two overlapped rather than queued: drafting cannot wait for diligence to finish, and diligence keeps producing things the draft has to absorb.

Which disagreements can only drafting find?

Some disagreements no amount of talking will surface. The words both sides are using are doing the hiding. A term sheet can say that the seller gives customary termsLanguage in a short document standing in for whatever is usual, so that the parties can move on without settling it. It postpones the argument rather than resolving it., and both sides can read that sentence, agree with it, and mean opposite things by it.

Nodding at an abstraction is not agreement, and the parties find out whether they ever agreed only when the abstraction becomes a clause with a number in it.

Take the word customary. To the buyer's transaction team, customary means everything a buyer normally gets, at the limits a buyer normally gets. To the seller, customary means as little as a seller normally has to give, at the limits a seller normally sets. Both readings are sincere. Both are held by experienced people. And neither reading is visible in the term sheet. The term sheet contains one word where the agreement will contain a definition, a list, a limit and a period.

A warrantyA statement by one party that something about the business is true, given so that the other party has something to fall back on if it turns out not to be. is the classic place this happens, and it is worth being precise about why. A warranty is only interesting in combination with the limits that sit around it: how large a claim has to be before it counts, how much can be recovered in total, and for how long claims can be brought. Change any of those and the worth of the warranty changes, without a word of the warranty itself changing. The term sheet named none of them. The agreement named all of them.

The limits set on the Harivansh Packaging Limited transaction are covered under warranties and their limits. The structural point is what matters here: the limits are the disagreement, and the limits appear nowhere in the term sheet.

The household version is one most people have lived through. Two siblings agree that whoever uses the car will look after it. Everyone nods. Nobody disagrees. Six months later one of them has a view about who pays for the tyres and the other has a different view, and it turns out that looking after it was never a shared idea at all. The phrase was a word standing in for one.

What happens when the two documents disagree?

The agreement governs. The reason is already established: the agreement binds and the term sheet, on that point, did not.

The useful consequence is less obvious and catches people out. A term of the term sheet that never made it into the agreement was dropped rather than carried forward, and a party relying on it is relying on a document that was supersededReplaced. A superseded document stops having effect on the points the new one covers, so anything left out of the new one simply stops applying. on purpose.

People find this hard because it feels unfair. A term was written down, both sides read it, nobody objected to it, and then it quietly failed to appear in the long document. Surely it survives? It does not. The agreement is not a fair copy of the term sheet. The agreement is the document the parties actually chose to be bound by, and the choice not to include something is a choice.

Definitive agreements are generally drafted to say this explicitly, so the point is rarely a surprise to the advisers. The point is frequently a surprise to the commercial people, who remember the negotiation and not the paper. Somebody on the buyer's side remembers agreeing that a particular employee would stay on, or that a particular liability would be dealt with a particular way, and cannot find it anywhere in the signed document. It went. Somebody traded it away in a late round, or it fell out when a section was rewritten, or the two sides quietly discovered they had never agreed on it and moved on without saying so.

The discipline this creates is simple and worth stating plainly. If a term matters, it has to be in the agreement, in the words the agreement uses, and its presence has to be checked before signature. Remembering that everybody agreed to it is not a substitute for reading it.

What did the two documents do on this transaction?

Work the pair against the actual purchase. Harivansh Packaging Limited agreed to buy one hundred per cent of Sundarban Polymers Private Limited, and two documents did two different jobs on the way there.

The term sheet fixed the shape. Enterprise value of Rs 1,320 crore, being ten times Sundarban Polymers' EBITDA of Rs 132 crore. Cash-free and debt-free, meaning the seller keeps the cash and clears the borrowings out of the proceeds. A working capital adjustment against a normalised figure. Net debt deducted at completion. Deduct the Rs 180 crore of net debt carried by Sundarban Polymers and the shape implies an equity value of Rs 1,140 crore. The Rs 1,140 crore is what actually reaches the sellers.

Read that list again and notice what is missing. Normalised against what? Measured on which day? Counted by whom, and using which definition of working capital? The term sheet had none of those, and it did not need them. Its job was to establish that both sides accepted the idea of a working capital adjustment at all.

The definitive agreement supplied every one of them. The agreement named the normalised working capital figure of Rs 96 crore. The agreement set out the mechanism that measures actual working capital at completion against that figure. Actual working capital came in at Rs 108 crore, so the price adjusted up by Rs 12 crore. On the other side, the transaction had assumed net debt of Rs 180 crore for Sundarban Polymers and actual net debt at completion was Rs 195 crore, so the price adjusted down by Rs 15 crore.

Add the two and the equity value moves from Rs 1,140 crore to Rs 1,137 crore.

The same figure checks a second way. A figure that reconciles from two directions is a figure that can be trusted. Enterprise value of Rs 1,320 crore, less the Rs 195 crore of net debt actually there at completion, is Rs 1,125 crore. Adding the Rs 12 crore working capital adjustment lands on Rs 1,137 crore again. The two routes must agree: both are the same arithmetic entered from different ends.

Now notice precisely what moved: the shape did not change at all, and the shape acquired a measurement. The multiple is still ten times. The basis is still cash-free and debt-free. The structure is still a purchase of the whole of the equity. The agreement added the machinery that turns those words into a number somebody can pay, and that machinery moved the answer by Rs 3 crore, or 0.26 per cent of Rs 1,140 crore.

The smallness of the move is worth looking at rather than skipping past, and it is the reason the figure below is drawn twice.

How does a transaction team actually use this split?

Three sets of people read the pair differently, and each reading is worth knowing.

A buyer's transaction lead, in this case Ashwin Rege on the Harivansh Packaging Limited side, uses the term sheet as a filter and the agreement as an instrument. The filter question is: on these terms, is this worth thirteen weeks and a diligence bill? If the answer is no, the cheapest possible discovery has just been made. Once past the filter, the term sheet stops being the working document, and the lead's job changes from negotiating a shape to running two workstreams that feed each other, one finding facts and one turning facts into clauses.

A chief financial officer, in this case Devyani Kulkarni, reads the two documents for a different reason: to know what is committed and when. The term sheet tells her what the transaction would look like if it happens. The agreement tells her which obligations the company has now taken on. The two documents are very different inputs into a funding plan. Committed borrowing cannot be arranged against a document that mostly states intention, and arranging it cannot be delayed until after signature either, so the funding work has to run alongside the drafting rather than after it.

A lender reads the pair as a sequence of increasing certainty. Early on, the term sheet is all there is, and a lender will work from it while being clear that it is not a commitment on either side. As the agreement takes shape, the lender starts caring about specific things in it: the conditions that could stop completion, the walk rights that could leave the borrowing undrawn, and the mechanism that decides the final amount payable. The mechanism decides how much money actually has to be there on the day.

All three readings share one habit worth copying: they treat the term sheet as a decision document and the agreement as an operating document, and they never confuse the two. The term sheet answers whether to proceed. The agreement answers what to do, and when, and what happens if somebody does not.

For a reader who is not on a transaction desk, the transferable habit is smaller and just as useful. When something important is agreed in outline, some of the words in that outline are standing in for a decision nobody has made yet. Placeholder words are where the trouble will come from, and they are much cheaper to argue about now than later.