The Definitive Agreement: Share Purchase, Business Transfer and the Clauses That Matter

A definitive agreement is the signed document that settles what is being sold, at what price, on what promises and on what conditions. Harivansh Packaging Limited signed one to buy Sundarban Polymers Private Limited. Buying shares moves the company whole, with everything already inside it. Buying a business moves only the assets and liabilities the document lists. The choice of structure decides what the buyer inherits.

Underneath that answer sits a habit worth building early. Every figure anyone ever quotes about a transaction was written into a clause by somebody, and the clause is the only place the figure can be checked. A press release can say Rs 1,320 crore. A conversation in the office can say Rs 1,320 crore. The document is where it becomes clear that Rs 1,320 crore is the enterprise value and that the sellers of Sundarban Polymers Private Limited were paid Rs 1,140 crore. Rs 180 crore of net debtBorrowings minus cash. It is the amount of borrowed money the business would still be carrying if it used every rupee of its cash to pay lenders down. stayed exactly where it was, inside the company. Finding the clause is the whole skill.

What does a definitive agreement do that a term sheet does not?

The term sheet, covered separately, is the short document that records the shape of a transaction before anyone commits. A term sheet says the two sides intend to buy and sell at a price built on a stated multiple, subject to work still to be done. Almost none of it binds. The whole document is a statement of intention written while both sides are still hopeful.

The definitive agreement is where intention becomes consequence, and that single change of purpose is why one document runs to a few printed sides and the other to a few hundred. A term sheet can say the price is Rs 1,320 crore of enterprise value. A definitive agreement has to say what happens if the working capital on the day of completion is not what everyone assumed, what happens if a warranty turns out to be untrue nine months later, what happens if a required approval never arrives, and what happens if one side simply stops answering. Each of those is a branch, and every branch needs its own words.

The difference between telling a landlord that the flat will be taken and signing the rental agreement makes the point. The first conversation covers the rent and the move-in date, genuinely most of what matters to a tenant. The signed paper covers the rent, the move-in date, and then several printed sides about what happens when the geyser stops working, when the tenant wants to leave early, when the landlord wants to sell, and when the water bill turns out to be higher than either of them expected. Nobody enjoys the second document. The rental agreement exists because life supplies the branches whether or not anyone wrote them down.

So a useful way to hold the two apart is this. A term sheet answers the question what is being done. A definitive agreement answers the question what happens if that is wrong. Reading the second one well means reading for the branches, not for the summary. The summary is already in the term sheet, and two hundred printed sides were never needed to get it.

What is a Share Purchase Agreement, and what actually moves?

In a share purchase, what changes hands is ownership of the company, and nothing inside the company is touched at all. The same customer contracts sit in the same drawers. The same employees arrive on Monday. The same lenders are owed the same money on the same terms. The only change is who holds the shares.

Because nothing inside the company moves, the buyer inherits everything inside the company, and the borrowings are the part that shows up in the price first. Sundarban Polymers Private Limited carries Rs 180 crore of net debt. Harivansh Packaging Limited did not pay that off and the sellers did not clear it. The Rs 180 crore simply came along, and that is precisely why the Rs 1,320 crore of enterprise value became Rs 1,140 crore of equity value paid to the sellers: the buyer takes on Rs 180 crore of obligations along with the shares, and that Rs 180 crore is 13.64 per cent of the enterprise value being talked about.

The buyer also inherits the rest of the ladder without any of it being written down as a transferred item. Sundarban Polymers earns Rs 880 crore of revenue, earnings before interest, tax, depreciation and amortisation (EBITDAEarnings before interest, tax, depreciation and amortisation. A rough measure of what the operations throw off before the effects of how the business is financed and how its assets are written down.) of Rs 132 crore, and after Rs 34 crore of depreciation and amortisation earnings before interest and tax (EBIT) of Rs 98 crore. In this structure there is no list of transferred assets, so none of those figures appears in one. The figures come with the company because they are the company.

Here is the reading discipline that follows, and it is short enough to remember. In a share purchase, the question to ask about anything is whether it is inside the company. If it is inside, it moved. A lawsuit filed against the company two years ago is inside. A tax assessment nobody has resolved is inside. A supplier who is owed money is inside. The absence of a list is exactly why the promises in the document matter so much in this structure: the buyer has no list to check, so the seller instead promises that certain things are true about the whole of what is inside.

What is a Business Transfer Agreement, and what stays behind?

A business transfer works the other way round. The company itself does not change hands at all. Instead the document lists assets and liabilities, and those listed items move from the seller to the buyer. The plant. The machinery. The inventory. Two named customer contracts. Whatever else the parties wrote down.

The list is the transaction, and anything the list leaves out stays behind with the seller. The list is the whole reading discipline for this structure, and it inverts the one set out above. In a share purchase the question is what is inside the company. In a business transfer the question is what is on the list. A supply arrangement that everyone assumed was part of the business, but which nobody wrote into the scheduleA list attached at the back of an agreement, referred to by the clauses in the main body. Keeping long lists out of the clauses is a drafting habit, not a sign the list matters less. of transferred assets, does not move. The buyer does not have it. The seller still does.

Picture a running tea stall outside a college gate. One way to buy it is to buy the small company that runs the stall, in which case the buyer gets the stall, the lease, the regular customers, the standing it has in the neighbourhood, and also the two months of unpaid electricity that nobody mentioned. Another way is to buy a list: one kettle, one counter, one stove, the stock of tea leaves. The second way leaves the unpaid electricity with the seller, and it also leaves the lease with the seller unless somebody remembered to write the lease onto the list. Buyers who take the second route and forget the lease discover, on the first morning, that they have bought a kettle.

Which of the two structures suits any particular transaction is a legal and a tax question. The company law side of a share transfer and of a transfer of a business sits with the Ministry of Corporate Affairs at mca.gov.in. The tax cost of either structure to the seller is settled by the Central Board of Direct Taxes at incometaxindia.gov.in.

Which clauses in a transaction document actually move money?

A document of two hundred printed sides is not two hundred sides of equally important material. Most of it is machinery: how notices are given, what language governs, what happens if one clause is unenforceable, how many copies were signed. The machinery matters when something goes wrong, and almost never before.

Four clause groups carry all the money in any transaction document, and everything else in the document exists to support them. Held as a group, the four can be found in a document never seen before, in any order, under any heading anyone chose to give them.

The first is what is being sold, and to whom. In a share purchase this is often one sentence: the sellers sell and the buyer buys the shares. In a business transfer it is a sentence plus the schedule that does the real work. Either way this clause fixes the structure, and the structure fixes what the buyer inherits.

The second is the price and the mechanic that finalises it. Almost nobody pays a fixed number. There is a headline figure, and then a mechanic that adjusts it once the actual position on the day of completion is known. On this transaction the headline was Rs 1,140 crore of equity value and the mechanic produced Rs 1,137 crore. Both figures are real. The two figures answer different questions.

The third is the promises, meaning the warranties the seller gives about the business and the indemnityA promise to make good a specific loss in money if a named thing turns out badly, rather than a general promise that something is true. arrangements sitting behind them. The promises are where a buyer converts uncertainty into a claim it might one day make.

The fourth is the limits on those promises. A promise with no ceiling is a different commercial arrangement from a promise capped at Rs 227.40 crore, or 20.0 per cent of the Rs 1,137 crore actually paid. The seller who agreed to that number was agreeing to a very specific worst case.

The four groups do something useful for anyone handed a document and an afternoon: what is important has already been decided. The reader takes those four groups and the definitions that feed them, and leaves the machinery for the day something breaks.

Where does each figure in this transaction actually live?

Now take the transaction apart and put each figure back in the clause that defines it. Placing every figure back in its clause turns a reader into someone who can be trusted with a document, and it takes about twenty minutes on a real one.

Every figure anyone quotes about a transaction has exactly one clause behind it, and the clause is where the figure is either confirmed or contradicted. Eight figures ran through this purchase. Eight clauses defined them. A reader who cannot point at the clause behind a figure does not yet know what the figure means, however confidently the figure is being repeated.

The ordering is structural rather than numerical: which clause carries which figure, and in what order they are opened. None of that changes from one transaction to the next, so a single fixed map does the work, and it can be taken in on one pass.

Read the map below slowly. Two things should stand out. The first is that considerationThe formal word for what one side gives the other under an agreement. In a purchase it is normally the money, and the clause carrying it is where the price lives. appears twice, once for the headline enterprise value and once for what the sellers actually received, and those are not the same clause doing the same job. The second is that three separate figures come out of the completion accounts machinery alone. A summary that mentions the price clause and stops has skipped two thirds of what decided the money.

The bridge from the first of those figures to the second is the single most common trap when talking about a transaction, so it deserves its own picture. Look at the three bars below and read them top to bottom.

Why does the definitions section decide what half the other clauses mean?

Here is the part almost nobody reads and everybody should. Near the front of a transaction document sits a list of words in bold with meanings attached: working capital, net debt, cash, EBITDA, loss, business day. The list of defined words is not a glossary for the reader's convenience. The list is operative text.

A defined term means what the document says it means, and nothing else, however strongly the word means something else everywhere in the world. If the document defines working capital as inventory plus receivables less payables, then working capital in that document is inventory plus receivables less payables, even where an accounting standard, a textbook and every colleague in the room would include something more. The clause that adjusts the price does not run on the ordinary meaning. The clause runs on the definition.

The definition is why the completion mechanic on this transaction worked at all. The agreement set normalised working capital at Rs 96 crore. Actual working capital at completion was Rs 108 crore. The price moved up by Rs 12 crore. The comparison of Rs 96 crore with Rs 108 crore is only honest because both numbers were computed the same way, on the definition the document gives. Computing the actual figure some other way and arriving at Rs 118 crore instead does not find a bigger adjustment. The other method compares two different quantities and produces a number that means nothing.

Think of two cousins agreeing to split the wedding costs equally. Splitting equally sounds complete. The agreement is not complete until somebody writes down whether the jewellery counts, whether the flights of the guests coming from abroad count, and whether the venue advance already paid counts. Every argument that follows is a definitions argument wearing the costume of a money argument. The document that writes those three lines down before anyone spends anything has done the same job the definitions section does.

What does the document say about the transaction not happening?

Roughly half a transaction document describes a world in which the purchase completes and everyone behaves. The other half describes the world in which it does not. The second half is where the conditions sit, along with the routes by which either side can stop.

On this purchase there were three conditions to completion: a regulatory approval, the absence of a material adverse change, and consents from two counterparties whose contracts change hands. A condition precedentSomething that has to happen before an obligation becomes live. Until every one is met or waived, the parties are bound to the document but not yet to the exchange it describes. is not a promise that something is true. A condition precedent is a gate: until it opens, nobody has to complete.

A document carries not only what happens if the transaction proceeds but what happens if it does not, and the second half is where each side's real worries become visible. For someone reading a real document for the first time, the second half is the more revealing one. Nobody negotiates protection against a risk they are not thinking about. So the conditions, and the routes out, are a map of the fears in the room. A long list of consents says somebody was worried the contracts would not travel. A carefully drawn material adverse change gate says somebody was worried about the nine weeks between signing and completion.

The detail of each route, and what the two sides can do to each other on the way out, is a separate subject and is covered separately. The point is that the clauses exist, that they sit in the document alongside the four money clause groups, and that a reader who stops before reaching them has read the optimistic half of a document written for the pessimistic half.

How are two hundred printed sides read in one afternoon?

Five things get read, in an order the document does not use. The five are the whole method, and they work because the four money clause groups plus the definitions feeding them are perhaps thirty printed sides of the two hundred.

The definitions come first, precisely because they are the least interesting reading in the document. Every clause read afterwards is built out of them, so twenty minutes there prevents a misreading of everything downstream. Then the price clause and its mechanic together, never separately. The headline and the mechanic answer different questions, and only the pair shows what was paid. Then the limits on liability, turning the promises from open-ended into a specific worst case. Then the conditions, showing what could still stop the transaction. Then the warranty schedule read against the disclosure schedule, the only way to find out which promises were quietly hollowed out before signing.

The reading order is not the order the document is printed in, and reading front to back is exactly how an afternoon disappears with nothing to show for it. A document opens with a cover sheet, the parties, and the recitalsThe short paragraphs near the front that set out the background: who the parties are and what they have agreed in outline. They explain the setting rather than create obligations.. The recitals explain the background and change nothing about the money. Then it works outward in the order somebody chose to type it. Read that way, four hours reaches the mechanics of signing and no further.

What does the whole transaction look like when it is mapped onto the document?

All of it comes together on one transaction. Every figure below comes from the same invented purchase, and each one is placed against the clause group that defines it. Work down the table and check the arithmetic line by line. Naming the clause behind each figure is the habit worth building.

Eight figures, eight clauses, and a reader who cannot point at the clause behind a figure does not yet know what the figure means. Two rows are worth pausing on, and both are places where careful people go wrong.

The first is the most the sellers can ever receive. Add the Rs 60 crore earn-outA part of the price that is paid later and only if the business hits a stated result. It is a way of settling a disagreement about the future by waiting to see who was right. to what was actually paid at completion and the answer is Rs 1,197 crore. Strike the same Rs 60 crore on the Rs 1,140 crore headline and the answer becomes Rs 1,200 crore, overstating things by Rs 3 crore. The headline was never what the sellers were paid. The base a figure is struck on is part of the figure. Naming the base every time removes the mistake.

The second is the amount paid above the target's net assets. Sundarban Polymers Private Limited has net worth of Rs 320 crore. Against the Rs 1,140 crore headline that leaves Rs 820 crore above net assets, or 71.93 per cent of that base. Against the Rs 1,137 crore actually paid it leaves Rs 817 crore, or 71.86 per cent. The two percentages differ by 0.07 points and both round to 71.9 per cent, so at one decimal the base discipline looks free. The discipline is not free, and the rupee route says so plainly: Rs 820 crore against Rs 817 crore is a difference of Rs 3 crore, exactly the net adjustment, with no rounding anywhere in the working.

The treatment of that amount above net assets on the buyer's books, and how much of it is allocated to identified intangibles before anything is called goodwillThe part of a purchase price that is not matched by identifiable assets and liabilities. It sits on the buyer's books as a single figure until somebody tests whether it is still worth carrying., is an accounting exercise. The Institute of Chartered Accountants of India, at icai.org, is where that treatment is set out.