The Scheme of Arrangement: Court-Sanctioned Restructuring in India

A scheme of arrangement is a court-sanctioned way of changing what a company owes or how it is put together, and its defining property is that it can bind a group rather than needing every last agreement. Binding a group is also why the argument inside a scheme is an argument about ranking. Between strict ranking and pro rata, Rs 65.3 crore moves between the two groups of lenders at Meghdoot Coated Products Limited.

Three things are being taken as settled underneath that answer, and it is worth naming them before anything is built on top. The first is that a ranking rule changes who receives what. The ranking arithmetic is set out under insolvency and quoted here rather than derived a second time. The second is that separating a business into parts is a structure. A scheme is frequently the route by which a separation is carried out, and the route is a different thing from the separation itself. The third is the arithmetic of what a heavily borrowed business can actually carry. Carrying capacity fixes the Rs 210 crore distributed below. Everything new here sits on top of those three.

What is a scheme of arrangement, in plain terms?

Strip away every procedural detail and a scheme of arrangement is an agreement, sanctioned by a court, between a company and the people it owes money to, or between a company and its members. The agreement can do a great deal. A scheme can reduce what is owed. A scheme can push repayment further out. A scheme can turn one kind of claim into another. A scheme can split the company into separate parts or fold separate parts together. In substance it is a bargain, and with the court taken out of it the whole thing is recognisable as an ordinary commercial negotiation.

So what does the court add? Exactly one thing, and it is the thing worth the trouble. An ordinary contract cannot touch a party who never signed it, and a sanctioned arrangement reaches that party anyway. A contract signed with four lenders is a bargain with four lenders. The fifth, who declined to sign, carries on holding the claim they always held, unchanged and undisturbed. A sanction changes that geometry: it makes the arrangement effective across a group, and the party who said no is inside it along with everybody else.

Little of that is procedural. Who has to agree, how many of them, how they are grouped, what gets filed and how long a court takes are all real questions with real answers. Every one of those answers belongs to the Companies Act, named and routed below. A scheme's purpose decides how the money ends up divided, and that purpose carries into any market.

Why does binding a group matter more than anything else here?

Take the position at Meghdoot Coated Products Limited, an invented maker of coated packaging. Rs 620 crore is owed to securedA lender is secured when a named asset stands behind the loan, so that lender has a claim against that particular asset ahead of lenders who have none. lenders and Rs 280 crore to unsecuredA lender with nothing specific standing behind the loan. The lender holds a promise from the business as a whole and no claim on any named asset. ones, and Rs 900 crore of borrowings in total. A year of trading underneath all of it throws off EBITDAShort for earnings before interest, tax, depreciation and amortisation. Treat it as a rough figure for what a year of trading produces, taken above the interest bill and above the charges that write asset values down over time. of Rs 60 crore, meaning earnings before interest, tax, depreciation and amortisation. Stack the borrowings on one year of that and the load is 15.0 times over. Even if every rupee of those earnings went nowhere else, repaying the borrowings would take fifteen years. No schedule services a load like that. Something has to change.

Under a rule requiring everybody to agree, the question is what that leaves the smallest lender in the room. Give that lender an arithmetic rather than an adjective: a claim of Rs 9 crore, one per cent of the Rs 900 crore owed. The Rs 9 crore claim carries 1.0 per cent of the money. The plan does not happen without that claim, so under unanimity it carries the whole of the say. Requiring everyone to agree hands the smallest position in the room the largest lever, and the lever is worth what stopping the plan is worth to everybody else.

There is a second, quieter reason the lever is so valuable, and it comes straight out of the ranking arithmetic. Suppose the Rs 9 crore claim is unsecured. Under strict ranking it receives nothing at all, and under pro rata it receives 23.3 per cent, or Rs 2.1 crore. So the holder of the smallest claim is also the holder with the least to lose from refusing and the most to gain from being paid to stop refusing. A rule requiring unanimity does not merely slow a restructuring down. A unanimity rule systematically rewards the position with the weakest claim on the merits, and it does so with money that would otherwise have gone to somebody else in the room.

Here is the everyday version, and it is worth holding on to because the finance version is the same shape at a larger size. Ten shop units share one mall roof. The roof needs a repair nobody can pay for alone. Nine of the ten agree a way of splitting the bill and the tenth refuses. Until there is some way of binding the tenth to what the nine agreed, the nine are not negotiating about the roof at all. The nine are negotiating with the one, and the price of that negotiation comes out of the roof budget.

What can a scheme change, and what can it never change?

The list of things a scheme can reach is genuinely long, and that is what makes the route attractive. A scheme can change the terms of what is owed, so a repayment due next year falls due in five years instead. A scheme can change the amount, so a claim of Rs 100 crore becomes a claim of Rs 40 crore and the rest is written downCutting the recorded amount of a debt down, with the part taken off no longer expected back by the lender and no longer carried as a liability by the borrower.. A scheme can change the form, so what was a loan becomes shares instead. A scheme can even change the shape of the company itself, and that is the usual route by which a demergerSplitting a business into separate companies, so that what was one holding becomes two, each with its own shares and its own accounts. is carried out. Four levers, all of them real, all of them movable by agreement plus a sanction.

Then there is the one thing on the other side of the line. A scheme rearranges claims against a pot and creates nothing, so a scheme cannot make the Rs 210 crore larger. Where does the Rs 210 crore come from? From what the earnings support. Rs 60 crore of EBITDA at a judged sustainable 3.5 timesA borrowing measured against a year of earnings. Rs 900 crore of debt against Rs 60 crore of earnings is fifteen of them, and lenders use that shorthand for how heavy a load is. is Rs 210 crore of debt the business can actually carry. Change the judgement and the figure moves. The range is worked through in full below. But whatever the figure turns out to be, no sanction touches it. A court order does not sell a single additional carton or reduce a single rupee of operating cost.

Why does that turn the whole thing into a distribution argument?

Put the two halves together and the consequence follows without any further assumption. The pot is fixed by the earnings. The claims are fixed by what was borrowed. The only thing left for anybody to argue about is how the fixed pot is divided across the fixed claims, and division is a closed system. Nothing in the room is adding to the total, so every rupee one group gains inside a scheme is a rupee some other group loses.

Division inside a closed system explains almost everything about how these negotiations feel from the inside. Both groups arrive with an argument, both arguments are internally coherent, and both cannot be satisfied. The secured lenders will argue that ranking is the whole point of having taken security in the first place, and that a rule which ignores it makes security meaningless. The unsecured lenders will argue that the business is still trading, that the value being shared out came from the business rather than from any particular asset, and that a rule handing everything to one group and nothing to the other is a poor way to keep a supplier in place.

Neither argument is bad faith, and neither is a negotiating trick. A lender arguing for the ranking rule that pays it more is doing exactly what its own position requires, and would be failing in its own duty by doing anything else. Two groups arguing for the rule that pays them are behaving as their positions require, and reading that as obstruction is the fastest way to misunderstand the entire process. The useful question is never who is being reasonable, but how much money the disagreement is actually about.

How much actually moves between the two groups here?

The arithmetic follows. Two ranking rules are put against the same Rs 210 crore. Under strict ranking, the secured group is paid before the unsecured group receives anything, and since Rs 620 crore of secured claims comfortably exceeds the pot, the secured group takes the whole Rs 210 crore and the unsecured group takes nothing. Under pro rataIn proportion to size. Each share comes out at the same fraction of what that holder was owed, so a claim twice as large receives twice as much., every claim is treated identically as a fraction of itself, so the secured group takes Rs 144.7 crore and the unsecured group takes Rs 65.3 crore.

Now subtract along the top row rather than reading down the columns. Rs 210 crore with Rs 144.7 crore taken out of it leaves Rs 65.3 crore, and that is what the secured group gives up in moving from one rule to the other. Read the second row and the unsecured group receives Rs 65.3 crore in making exactly the same move. The two figures of Rs 65.3 crore do not merely happen to be close. One figure is being seen from opposite sides of the table, and that figure is the substance of the negotiation once everything else has been stripped away.

One number generalises, and it is worth keeping alongside the Rs 65.3 crore. Of the Rs 900 crore owed altogether, Rs 280 crore sits with the unsecured group, giving it 31.1 per cent of the borrowings. Under pro rata that group receives 31.1 per cent of whatever the pot happens to be, and under strict ranking it receives none of it. So the amount that moves when the rule changes is always 31.1 per cent of the pot, whatever the pot turns out to be. The 31.1 per cent share is why the range work below falls out so cleanly, and that share depends on the split between the two groups rather than on the size of the pot at all.

What is each group actually choosing between?

Now for the part that gets misquoted more than anything else in a restructuring. Working the secured group's outcome under each rule produces two recovery percentages, 33.9 per cent and 23.3 per cent, and summarising the whole argument as one against the other is tempting. The summary is worse than useless. Both percentages describe one group and not two, and a lender who is not secured will read the 33.9 per cent as though it might one day be its own.

Set it out with the base attached to every figure and the confusion cannot survive. The secured group is choosing between 33.9 per cent and 23.3 per cent, and both of those are measured against its own Rs 620 crore. The unsecured group is choosing between nothing and 23.3 per cent, and both of those are measured against its own Rs 280 crore. Neither group is choosing between the same pair of numbers as the other, and a percentage with no base attached to it is not a fact about anything.

The bottom row settles the argument about whether anybody is better off overall, so look hard at it. Taken across the whole Rs 900 crore, the recovery is 23.3 per cent under strict ranking and 23.3 per cent under pro rata. Nothing else was possible: the pot was Rs 210 crore before anybody argued and Rs 210 crore afterwards, and 210 divided by 900 does not care which rule was used to hand it out. The aggregate is identical under both rules because the pot never moved, so every point one group gains has come out of the other and none of it has come from anywhere new.

A small stretch of forced arithmetic hides in that drawing, and it is worth pointing at. The distance between the two secured fills is exactly the height of the filled unsecured column under pro rata. Both heights represent the same Rs 65.3 crore drawn on the same scale, so no other outcome was available. Seeing the two heights match is a faster check than following the arithmetic again.

Does the assumed multiple change how much is argued over?

Everything so far has treated the Rs 210 crore as settled, and it is not. The Rs 210 crore came from a judgement that this business can sustainably carry 3.5 times its earnings, and the honest position is that the sustainable level is a range rather than a point. At 3.0 times the earnings support Rs 180 crore. At 4.0 times they support Rs 240 crore. Nothing about the business settles which of the three is right. A sustainable level is a judgement, and three defensible judgements can sit side by side.

Work the swing at each of the three levels and the answer is plain. At a pot of Rs 180 crore the secured group takes Rs 124.0 crore under pro rata, so Rs 56.0 crore moves. At Rs 210 crore, Rs 65.3 crore moves. At Rs 240 crore the secured group takes Rs 165.3 crore under pro rata, so Rs 74.7 crore moves. The amount being fought over is not a fixed sum at all, and across the three judged levels it travels from Rs 56.0 crore to Rs 74.7 crore, a spread of Rs 18.7 crore.

The Rs 18.7 crore spread is forced by the way the levels were set, and forced arithmetic is worth unpacking. Take the outer two multiples: 3.0 times and 4.0 times sit one turn apart, so the pots they produce differ by a single turn of earnings, and a turn of earnings here is Rs 60 crore. The unsecured group takes 31.1 per cent of any pot under pro rata and none of it under strict ranking, so the swing must differ by 31.1 per cent of that Rs 60 crore, and 31.1 per cent of Rs 60 crore is Rs 18.7 crore. Nothing there was measured. The Rs 18.7 crore follows from the choice of levels, and two figures that match by construction are not independent confirmation of each other.

The practical consequence is uncomfortable. A scheme negotiated on the assumption of 3.5 times has quietly settled the multiple for everybody in the room, and the parties who spent their energy arguing about the ranking rule may have spent it on the smaller of the two questions. Rs 65.3 crore was in dispute between the two groups. Rs 60 crore of the pot itself sat between the outer judgements about what the business can carry, and nobody was arguing about that at all.

Why does a scheme that only redistributes leave the problem where it was?

Plans go wrong at exactly this point, and they go wrong quietly. A scheme that redistributes claims and does nothing else has moved money between two groups of lenders and has not touched the relationship between the borrowings and the earnings. Rearranging who is paid does not change what the earnings support. A scheme that leaves the debt and the earnings where they were has redistributed a problem instead of resolving one, and the paperwork will look identical either way.

The distinction that matters is between the two kinds of change a scheme can carry. One kind alters the size of the claim: a write-down, a conversion into shares, a genuine reduction in what is owed. The other kind alters only the shares of a fixed pot: a change in the ranking rule, a reallocation between groups, a rearrangement of who stands where. The first kind changes the burden the business carries afterwards. The second kind changes only who bears it. A plan can contain a great deal of the second and none of the first, and still be described in its own summary as a restructuring.

Watch for it in how a plan describes itself. Where the sentence explaining what the arrangement achieves is about who receives what, the arrangement is distributive and nobody has yet asked what the business can carry. If the sentence names a smaller figure that the Rs 900 crore has been brought down to, then the size question has been engaged. Both sentences can appear in the same document, and only one of them determines whether the business can service what is left afterwards.

What is settled in the room, and what is set in law?

Two columns run through this whole subject and they never swap places. In one column sit the things settled by arithmetic and negotiation: what the earnings support, how a pot divides, which ranking rule the parties accept, what each group recovers on its own claim. Arithmetic does not have a jurisdiction, so every answer in that first column is the same everywhere.

In the other column sit the things set in law. Who has to consent to a scheme. In what proportions. How groups of creditors are constituted for the purpose of consenting. Which documents must be filed, and with whom. Which considerations a court weighs before it sanctions anything. How long any of it takes. Not one figure, proportion, period or requirement from that second column is settled by arithmetic; each of them is fixed in law.

What does a scheme cost, apart from the terms it sets?

The distribution table settles who receives what, and it is silent about what getting there costs. Three costs sit outside it. The first is time, and a business that cannot service its borrowings is a business where months are expensive. The second is professional cost, which is real money leaving a company that has very little of it, and which is spent before anybody knows whether the arrangement will be sanctioned at all.

The third is the one that gets underweighted, and it is peculiar to this route. A scheme requires the position to be laid out for a whole group at once. Everybody in that room learns what the business owes, what it earns, what it can carry, and what its lenders are being asked to give up. For a business still trading, telling every counterparty at once exactly where it stands is a genuine cost, and it appears in no recovery percentage anywhere. Suppliers reconsider terms. Customers ask questions about continuity. Nothing in the arithmetic above captures any of that, and the arithmetic is what gets circulated.

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

An adviser opening a restructuring is walking into a room where every party wants one number and the honest answer is a range. The first job is not to pick a ranking rule. Everything downstream is a division of the pot, so the job is to get its size agreed, or at least to get the range on the table and the judgement behind it stated openly. An adviser who lets the ranking argument start before the size argument has finished has allowed the room to spend its energy on the smaller question, and on the arithmetic above the smaller question was Rs 65.3 crore while the range on the pot itself spanned Rs 60 crore.

A lender reads it in the opposite order, from its own seat outwards. Before any recovery percentage means anything, the question is which base it was struck on, and that follows from where the claim sits. A lender holding unsecured paper looks at 33.9 per cent in a circulated plan and has to ask, immediately, whose 33.9 per cent that is. On the arithmetic above the answer is somebody else's, and the figure for this lender is nothing under one rule and 23.3 per cent under the other. The protective habit is small and unglamorous: a recovery percentage is never read without finding the rupee amount it came from and the claim it was divided by.

An analyst looking at a business that has come through a scheme asks a different question again, and it is the one the failure block is built around. Not what the arrangement was called, and not how the recoveries were split, but what the borrowings are against the earnings on the far side of it. If the answer is a multiple the business can carry, the scheme did something. If the answer is a multiple that still cannot be serviced, the scheme moved money between lenders and the same conversation is coming back.

And a household reading about any of this in the ordinary way can hold on to the one transferable idea. When a fixed amount of money has to be divided among people who all have a claim on it, the argument about who gets what will always feel like the important argument, and it is usually the second most important one. The first is how much there is, and whether anybody in the room has actually checked.