Demerger vs Spin-Off: Who Ends Up Holding What, and Why

A demerger and a spin-off both hand the separated business to the people who already hold the company, so neither one brings a rupee in from outside. The route taken differs, and so does what those holders finish with. Neither route touches the Rs 900 crore sitting on the books of Meghdoot Coated Products Limited, an invented manufacturer, and settling which word means what before either one gets used saves a painful argument later.

The background needed here is short. A struggling borrower cares about two numbers: what the business earns and what it owes. A separation moves neither of them. All three routes are set out together under separation structures, along with the debt arithmetic in full, so the comparison here narrows to two of the three and reuses that arithmetic rather than rebuilding it. Meghdoot Coated Products Limited has a creditor list running to Rs 900 crore, of which Rs 620 crore is secured and Rs 280 crore is not; a year brings in Rs 60 crore; so its leverage multipleBorrowings measured against a single year of earnings, so a bigger number means more years of earnings would be needed to clear what is owed. stands at 15.0 times.

What do these two routes actually have in common?

The larger part of the answer is the part people skip, so start there. Both routes end with the separated business sitting in the hands of the people who already held the original company. Nobody outside that list of holders is asked to write a cheque. There is no buyer at the other end of a negotiation, no price to argue about, no completion payment, no consideration of any kind moving inward.

Because no outside party pays anything, neither route produces a single rupee, and so neither of them is an answer to a debt question. The absence of any outside payer does the heavy lifting in this comparison, and it is worth being blunt about why. A business at 15.0 times has a gap between what it owes and what it can service. Closing that gap needs one of three things: more earnings, less debt, or money arriving from somewhere. A separation of either kind delivers none of the three on its own.

A household version of the same thing makes the point. Suppose a household runs a grocery counter and a small tailoring counter out of one rented shop, on one loan taken years ago. One evening the two people running it decide to keep separate books, separate cash boxes, separate stock. By the next morning they have two businesses where there was one. Has the loan changed? Not by a rupee. Has anybody handed them money for making the change? No. Separate books are useful for clarity, for accountability, perhaps for eventually letting one counter borrow on its own strength. Separate books are not a repayment.

The confusion arises because a separation feels like a transaction. There are documents. There are advisers. There is an approval process. Something visibly happens. But look at where the money moves and nothing does. The same people hold the same underlying business before and after; only the packaging around it changed. The one separation structure that does bring cash in is the one where a slice is sold to somebody outside, and that route is set out under separation structures.

What does a demerger do to the company itself?

In a demerger, the company is the thing being cut. One legal entity becomes two, and the holders who had shares in the original finish holding shares in both resulting entitiesEach of the two businesses that is left in existence once a division has taken effect.. Nobody is bought out and nobody is diluted; the same register of holders now appears twice.

Because it is the company being divided rather than a shareholding being moved, every asset, every contract and every borrowing has to end up on one side of the new boundary, item by item, and somebody has to decide which. Sorting every item onto one side or the other is the real work of a demerger, and it is why the route is slow. A machine sits somewhere. A lease is in one name. A supply contract has a counterparty who may or may not consent to being assigned. A borrowing was taken by one entity against security that may sit across what will become the boundary. None of that sorts itself out.

The exercise has a name, allocationThe exercise of deciding which side of a division each asset, contract and borrowing lands on, taken item by item rather than in aggregate., and it is the same discipline as drawing a perimeterThe line drawn around what is inside a transaction and what is deliberately left outside it. around a transaction. Drawing that perimeter is set out under the separation process. The difference is that a perimeter in a sale has a buyer on the far side pushing back on every line. In a demerger both sides end up with the same holders, so there is nobody on the far side. Having nobody on the far side sounds easier, and in one way it is. In another way it is harder. When nobody is arguing, a loose allocation can sail through unchallenged and surface later as a dispute between two sets of lenders who did not think they were on opposite sides of anything.

Return to the grocery and tailoring counters for a moment. A demerger is what happens when the two counters are genuinely pulled apart: two rent agreements where there was one, two electricity connections, two sets of accounts, a decision about which of them keeps the delivery scooter. Somebody has to sit down and go through the list. Nothing about that list is glamorous and all of it is the actual job.

What does a spin-off do instead?

A spin-off starts from a different place. Here the business being separated is already a distinct entity, held by the parentThe entity that sits above another and holds its shares, which is often where a group's borrowings already are. as a shareholding. Nothing needs to be cut apart. The cut was made at some earlier point, when the subsidiary was set up or acquired. In a spin-off the parent hands that shareholding out to its own holders. After the distributionHanding shares out to the people who already hold the company rather than selling them to anybody., the holders sit with two shareholdings side by side where they previously had one, and the parent no longer sits above the other entity.

The separation already existed as a matter of structure, so nothing is divided in a spin-off and the only thing that moved was who holds the shares. The absence of anything to divide is the whole difference in mechanism, and almost everything else that differs between the two routes follows from it.

The godown makes this concrete. Suppose the same two people running the shop have always had a separate lease on a small godown around the corner, taken in its own name, kept in its own books, with its own tiny set of dealings. Handing one partner the keys and the lease to the godown is not the same operation as splitting the shop counter in two. There is no list to work through. The godown was never mixed in with anything. Only the holder of the godown changes.

Having nothing to allocate is why a spin-off tends to be quicker to describe and quicker to execute, and why people reach for it when a group already has a tidy structure. The same fact is why a spin-off can be unavailable for reasons that have nothing to do with what anybody wants. If the business to be separated is not already a separate entity, there is nothing to distribute, and the route simply is not open. The separate entity would have to be created first, and creating it puts the group back into the allocation work the spin-off was meant to avoid.

Why do the two words keep drifting into each other?

Now the awkward part. The two words above have been used here with two specific meanings, and those meanings are not universal. Usage moves between markets and it moves between advisers inside the same market. In some places a demerger is the umbrella word and a spin-off is one variety of it. In others the two are treated as synonyms and the structure has to be read out of the rest of the document. In still others the word carries a meaning tied to a particular statutory route. A meaning like that is fine locally and travels badly.

Two parties who have agreed on a word have not yet agreed on anything, so a written treatment, a term sheet or a board paper states which meaning it is using at first mention. Two meanings have been stated above and used consistently: a demerger divides the company itself, a spin-off distributes a shareholding in something already separate. Neither meaning is a definition that anybody else is bound by.

So what does the ambiguity cost? Not much at the stage where everyone is talking. The cost lands at the stage where a document has to be drafted against what was agreed, and there it is heavy. A term sheetThe short document that records what the parties have agreed in outline before anybody drafts the long form. that says the parties will proceed by way of a demerger has recorded a word. If one side pictured the company being cut in two and the other pictured a subsidiary being handed out, the term sheet recorded a disagreement in a form that looks like agreement. Nobody raised a query because the word matched.

There is a habit worth building here, and it costs one line. Wherever either word is written, its meaning should follow immediately after, in plain words, naming the company being divided or the shareholding being handed out. The reader on the other side then either agrees or objects, and either way the disagreement surfaces early. Reading works the same way in reverse: when either word turns up in somebody else's document, the safe course is to assume nothing and find the sentence that says what is being divided. If there is no such sentence, that absence is the finding.

Where does the debt come to rest, and does the total move?

Here is where the arithmetic starts, and here is where the two routes visibly part company. Against a claim list of Rs 900 crore, Meghdoot Coated Products Limited brings in Rs 60 crore in a year. For the comparison to have anything to bite on, that Rs 60 crore has to be attributable to two sides, and the split assumed under separation structures is a larger side earning Rs 45 crore and a smaller side earning Rs 15 crore. The split into Rs 45 crore and Rs 15 crore is an assumption made so the arithmetic has something to work on, not a division the record contains, and every figure below inherits that caveat. The record holds one earnings figure for the business and no breakdown of it at all.

The demerger comes first. The company is divided, so the Rs 900 crore has to be allocated between the two resulting entities. Allocated in the same proportion as the earnings, the larger side takes Rs 675 crore while the smaller takes Rs 225 crore. An allocation is a placement rather than a payment, so the two amounts set beside each other come back to Rs 900 crore, as they must.

So what does the allocation do to each side? Rs 675 crore divides by Rs 45 crore and lands on 15.0 times. Rs 225 crore divides by Rs 15 crore and also lands on 15.0 times. Both resulting entities stand exactly where the undivided business stood. The matching answer is not a happy coincidence, and it is not evidence that the allocation was well judged. The match is forced arithmetic: allocating a total in the same proportion as the earnings makes each part carry the ratio of the whole by construction. Any other proportion gives a different pair, and an even split gives one.

Now the spin-off. Nothing is divided, so nothing is allocated. The borrowings stay wherever they already sit, and in a group structure that is usually with the parent, where the borrowing was raised. Put the whole Rs 900 crore there and the parent stands at Rs 900 crore against Rs 45 crore, or 20.0 times. The separated side stands at nil against its Rs 15 crore. One entity now looks worse than the whole ever did and the other looks spotless, and the group total has not moved by a single rupee.

The record places the Rs 900 crore nowhere inside a structure, so where the borrowings sit before a distribution is itself a construction here, and the mirror is worth working as well. Put the whole Rs 900 crore in the entity that is being handed out instead, and the picture inverts: the separated side carries Rs 900 crore against Rs 15 crore of earnings, putting that side at 60.0 times, and the parent stands at nil on its Rs 45 crore. Nobody would design a group that way on purpose. The mirror is worth working anyway: a spin-off improves nothing by itself. A spin-off reveals where the borrowings already were, and the answer can be flattering or brutal depending on a structural fact that predates the decision.

The unchanged total is the point to hold on to. Rs 675 crore beside Rs 225 crore returns to Rs 900 crore. Rs 900 crore beside nil comes to the same place. Two different routes, two very different pictures of each side, one identical total. Anyone reading only the parent after a spin-off would say the position got worse; anyone reading only the separated side would say it got better; both would be describing the same unchanged Rs 900 crore from opposite ends.

Is the allocation a choice, or does it fall out of the arithmetic?

The allocation is a choice, and the block above may have made it look otherwise. Allocating Rs 900 crore in the same proportion as Rs 45 crore and Rs 15 crore is one option among many, and it happens to be the one option under which both sides reproduce the ratio of the whole. A reader who saw 15.0 times land on both sides could reasonably conclude that a division preserves leverage, so the choice is worth saying out loud. It does not. A division preserves the total; the proportion in which the total is placed is decided by people.

Take the same Rs 900 crore and split it evenly instead: Rs 450 crore on each side. The larger side now carries Rs 450 crore against Rs 45 crore of earnings, or 10.0 times. The smaller side carries Rs 450 crore against Rs 15 crore, or 30.0 times. The two sides are now 20.0 turns apart, and the only thing that changed was a decision about where to place rupees that were already owed. The total is still Rs 900 crore. The group still earns Rs 60 crore. Nobody has paid anybody anything.

A demerger has a dial inside it, and the dial has consequences for two different sets of lenders who may not both be in the room. A lender to the smaller side, if that side ends up carrying Rs 450 crore, is now facing an entity at 30.0 times where they were previously lending to a business at 15.0 times. Nothing about their claim changed. The business standing behind it did.

The dial is also the honest reason why an allocation is negotiated rather than calculated. There is no published level of borrowing that either resulting entity ought to carry, and the record here holds none: no net worth, no assets, no description of what secures what, not even an interest rate. So there is nothing to measure a proposed allocation against and call it right or wrong. The work that can be done is arithmetic. Take each candidate allocation, work out where it leaves each side, and put the numbers where everyone can see them.

What genuinely differs between the two, then?

Four things, and it is worth being precise about them because the list is shorter than most treatments of this subject suggest.

The first is what is being divided. A demerger divides the company, and that is why the allocation exercise exists. A spin-off divides nothing. The entity being separated was already apart.

The second is how the borrowings come to rest. In a demerger they come to rest wherever the allocation puts them, and an allocation is a negotiated outcome. In a spin-off they come to rest wherever they already were, and nothing about that is negotiated unless somebody deliberately moves them first.

The third is what has to be established about each side standing aloneWhat a business looks like once it shares nothing with the other side, from its own accounts down to its own banking arrangements.. In a demerger, two sides have just been created, so two sides have to be shown to work on their own. In a spin-off, the separated entity was already standing on its own in structural terms, so the question narrows to whether the parent still works without it. Fewer things to prove, and a narrower question.

The fourth is which route is available at all, and that one is not a choice the parties make. Availability is settled below.

The size of the debt is conspicuously not on that list, and saying so plainly is the useful part of setting the two routes side by side. The Rs 900 crore appears in every column of every table above at exactly the same size. Both routes leave it there. If the question brought to this comparison is which route helps a business that cannot service what it owes, the answer is that this is the wrong pair to be choosing between.

Who decides which of the two routes is even available?

Not the board, not the lenders and not the adviser. AvailabilityWhether a given route is open at all, which is settled outside the negotiation rather than inside it. is settled outside the negotiation entirely, and the parties choose only among routes that are already open to them. A board can resolve to proceed by a route that the law does not provide for, and the resolution will simply not go anywhere.

Availability is the deepest of the four differences. The other three are outcomes the parties shape, and availability is a constraint the parties walk into. Availability is also the difference most often skipped. A comparison of this kind tends to line up two structures as though they were both sitting on the shelf waiting to be picked.

Does either route change what the people owed money receive?

The arithmetic is already worked above, and the conclusion follows from it.

The answer is no, and here is the arithmetic behind it rather than an assertion. The amount a restructuring can pay out is limited by what the business can sustain, and the sustainable level is a judgement about how many turns of earnings a lender would live with. Three such judgements are set out under the write-down range, and they are repeated here only to place the middle one. At 3.0 times, Rs 180 crore is all the Rs 60 crore of earnings will carry, leaving Rs 720 crore to be struck out, a proportion of 80.0 per cent. At 3.5 times the carrying capacity rises to Rs 210 crore, Rs 690 crore is struck out and the proportion eases to 76.7 per cent. At 4.0 times the capacity is Rs 240 crore, Rs 660 crore is struck out and the proportion is 73.3 per cent. Each of those three proportions is measured on everything owed. The record holds those three proportions as the whole numbers 80, 77 and 73 per cent; 76.7 is that middle whole number carried to one decimal, and one decimal is used throughout so that the tables and the drawings cannot disagree with each other.

Take the middle level, Rs 210 crore, purely because it is the middle. Set beside a claim list totalling Rs 900 crore, that reads as 23.3 per cent, and the base has to be named every single time the figure is used. Carrying capacity follows the earnings, and no word on a board paper adds to those, so neither route changes the Rs 210 crore. A division moves rupees of debt sideways between two entities; a distribution moves a shareholding upward and out. Neither adds a rupee of earnings, and neither brings a rupee in from outside.

How much each group actually receives out of that Rs 210 crore is a question about ranking rather than about either of these routes, and it is worked under the ranking of claims. Strict ranking sends the entire Rs 210 crore in a single direction, to the secured group, and the recovery reads 33.9 per cent once it is struck on the Rs 620 crore standing behind that group. The holders of unsecured paper get nothing at all. A share taken in proportion splits the direction of travel: Rs 144.7 crore towards the secured group, Rs 65.3 crore towards the holders of the Rs 280 crore of unsecured claims. Both groups then recover 23.3 per cent, each measured against what it is owed and against nothing else. The danger of leaving the bases unnamed is worth marking. One of them uses Rs 620 crore underneath it and the other uses Rs 900 crore, so the two figures sit on different denominators and are not a comparison at all, though they read like one. Both rules appear only to show what ranking does to an outcome. Neither describes what the law in India actually provides, and which of the two applies is a matter for that law.

When does the distinction actually matter to a decision?

The distinction matters, and it matters in two specific places. The first is what has to be documented and established: a demerger requires proof that two businesses work standing alone and a spin-off asks a narrower question about one. The second is where the borrowings come to rest, negotiated in one route and inherited in the other.

The two places are not small. Between them they account for most of the time, most of the cost and most of the argument in a separation. For the person who has to run the process, the distinction is the difference between an exercise that takes a long list of decisions and one that takes a shorter list.

The distinction does not touch the size of the debt, and the size of the debt is the question a business at 15.0 times is actually facing. Being willing to say so is the useful part of this comparison. A great deal of writing on this subject stops after the mechanism and leaves the reader with a tidy comparison and no sense of what it is good for. The tidy comparison is real; it just does not answer the question most people arrive with.

So there is a sequencing point here that costs nothing to adopt. Settle what the business owes and what it can sustain first. Only then take up which shape the group should take. Taken the other way round, the work becomes the design of a structure for a set of entities whose debt burden has not yet been settled, and the structure then gets designed twice.

What does this look like on a shop floor rather than a balance sheet?

Two people run one shop on one loan. The shop has a grocery counter and a tailoring counter, and there is also a small godown around the corner held on a separate lease that has always been kept apart from the shop.

Dividing the shop into two counters, each with its own books, its own rent share and its own stock, is the first route. Somebody has to sit down and decide which counter keeps the delivery scooter, which one takes the deposit with the landlord, and how much of the loan each counter is going to be treated as carrying. The loan share is the allocation, and it is a decision, not a calculation. The two of them could say the grocery counter carries three quarters of it because it earns three quarters of the money, and both counters would then be carrying the same burden relative to what they earn. Half and half is equally available, and the tailoring counter would then be left carrying far more than it can service while the grocery counter looks comfortable.

Handing one partner the keys and the lease to the godown is the second route. There is no list to work through. The godown was never mixed in. Only the holder changes, and nothing else about the arrangement moves.

Either way the loan is the same loan, taken for the same amount, owed to the same lender. Nobody walked in and paid it down because the counters were separated or because the godown changed hands. If the two people running the shop cannot service the loan on the money the shop makes, they still cannot service it the next morning. An unchanged loan is the whole of the finance content in this comparison, dressed in rupees in the tables above and in a shop lease here, and it is the same content in both.

Who reaches for this comparison, and what do they do with it?

Four kinds of reader use this, and they use it differently.

A lender to one side of a proposed demerger reads it as a question about the allocation and nothing else. The mechanism is somebody else's problem; what matters is which entity the claim ends up sitting against and what that entity carries relative to what it earns. The practical move is to ask for the allocation in rupees before the structure is approved rather than after, and to ask what the resulting entity stands at, in turns of earnings, under the proposed placement. If the answer is Rs 675 crore against Rs 45 crore, that is 15.0 times and it is the same position as before. If the answer is Rs 450 crore against Rs 15 crore, that is 30.0 times and it is a materially different exposure created by a decision rather than by anything the business did.

An analyst covering a group that has announced a separation reads it as a warning about comparison. The parent after a spin-off looks worse on leverage than the group did, and the separated entity looks pristine, and neither of those observations means anything on its own. The move is to add the two sides back together before saying anything about direction. Rs 900 crore and nil is Rs 900 crore, and the group earns Rs 60 crore, and the combined position is 15.0 times, exactly where it started. Adding the two sides back together takes ten seconds and stops a whole category of wrong conclusion.

An adviser drafting the paper reads it as a drafting discipline. Define the word where it first appears; state what is being divided or handed out; state where the borrowings will rest, or state explicitly that this is open. The cost of doing all three is a paragraph. The cost of not doing them shows up as a renegotiation after approval.

And a household comparing two structures for a small business reads it as the shop and the godown. Whether the counters are split or the godown is handed over, the loan is unchanged, so the loan question comes first. All four readers converge on the same practical instruction: the structure question and the debt question are separate questions, and the debt question is the one with a deadline attached to it.

Is the difference between the two routes a matter of degree?

It is not. Many questions about a restructuring are about a quantity: how far a write-down goes, what a pot is worth, how a range narrows. The choice between a demerger and a spin-off is not one of them.

The difference between these two routes is categorical rather than continuous: a company is divided, or a shareholding is distributed, and there is no quantity sitting between the two for a reader to move. Any dial drawn between the two would manufacture a continuum with no counterpart in the structures themselves, and a reader who dragged it would come away believing there are intermediate structures on a scale between the two. There are not.

One genuinely continuous relationship sits in the neighbourhood, and it is how an allocation moves each side's leverage. Allocation and leverage are covered under separation structures. The figure comparing the proportional allocation with the even split gives the two ends of that dial, with the arithmetic visible at both.