Restructuring: Changing the Shape of a Business or Its Debt

The two numbers that produced the trouble are the debt and the earnings that have to service it, so a restructuring changes either what a business consists of or what it owes. Meghdoot Coated Products Limited, an invented maker of coated packaging, carries Rs 900 crore of borrowings against Rs 60 crore of annual operating earnings. The ratio is 15.0 times. Every route in this guide moves one of those two numbers.

The answer hides a difficulty that surfaces within a week of working on one of these. The word restructuring is used for two quite different operations, and nobody says which they mean. In one conversation it means the business is being taken apart and put back together in a different shape: a division sold, a division separated out, the whole thing handed to somebody else. In the other it means the business is left exactly as it is and what it owes is rewritten: less of it, later, or turned into shares. Both are called a restructuring. A reader who does not ask which one is meant will sit through two different conversations believing they are one.

Everything below rests on three things that were settled elsewhere and are used here rather than rebuilt. The first is EBITDAEarnings before interest, tax, depreciation and amortisation. A rough stand-in for the cash a business throws off from trading, struck before anything is paid to lenders or to the tax authorities., or earnings before interest, tax, depreciation and amortisation, as a measure of what a business earns from trading, and a multiple of it as a way of describing how much is owed. The second is what it means for a lender to hold a specific asset behind a loan. Security of that kind is what makes the split between two groups of lenders worth arguing about at all. The third is the locked position of the borrower worked throughout: Rs 900 crore of borrowings, Rs 60 crore of EBITDA, of which Rs 620 crore is secured and Rs 280 crore is not.

So what does a restructuring actually change?

Take the two branches separately. The vocabulary blurs them and the arithmetic does not. On the first branch, the business changes shape. The business might be split into two companies, leaving a healthy part no longer carrying a struggling one. The business might sell a division, turning part of itself into cash. The business might be sold whole, and then whatever price it fetches becomes the only money there is. On the second branch, the business is untouched and the obligation changes shape. Part of the amount owed is given up. The dates are pushed back. Some of it stops being debt and becomes shares.

The two branches are not alternatives so much as different levers on the same machine, and the machine only has two moving parts. There is an amount owed, and there is an amount earned each year out of which the amount owed has to be serviced. Anything done to a business in trouble either makes the first number smaller, makes the second number bigger, or does neither. There is no third possibility, and a proposal that moves neither number is not a restructuring at all, however much paper it generates.

Why do only two numbers matter here?

Because the trouble is defined by the relationship between the debt and the earnings, and by nothing else. Meghdoot Coated Products Limited owes Rs 900 crore. The company earns Rs 60 crore a year before interest, tax and the non-cash charges. The first divided by the second is 15.0 times. The ratio of 15.0 times is the whole diagnosis, and once it is in hand there is no cleverer number to look for.

Here is what 15.0 times means in practice, and it is worth being blunt about it. Suppose every rupee of the Rs 60 crore went to lenders, with nothing kept back for interest, nothing for maintenance and nothing for tax. Fifteen years of that would clear the borrowings and only just. Since interest has to be paid out of the same Rs 60 crore, and since a business that spends nothing on itself for fifteen years does not survive to the fifteenth, the honest statement is simpler. At 15.0 times there is no repayment scheduleThe agreed calendar of dates and instalments by which a borrowing is meant to be paid down. Changing it moves when money is due, never how much is owed. that the earnings reach, so the question is not how to reschedule the amount but how much of it has to go.

A rescheduling conversation is usually the first one anybody proposes, and at 15.0 times it dies quickly. Pushing an instalment from March to September moves a date. Moving the date does not move the Rs 900 crore and it does not move the Rs 60 crore. The two numbers that produced the problem stand exactly where they stood, and so does the problem.

Restructuring vs Recapitalisation: which of the two is being talked about?

The distinction between the two saves the most time, and it is worth fixing before anything else. A recapitalisation rearranges the funding of a business and leaves the business itself alone. New shares are issued and the money comes in. Some of what is owed stops being debt and becomes shares. The order in which lenders would be paid is rewritten. Through all of that, the plant still makes what it made, the customers are the same customers, and the Rs 60 crore of earnings is still Rs 60 crore.

A restructuring is the wider word. The wider word covers every recapitalisation, and it also covers the operations that change what the business consists of: a division sold, the company split in two, the whole thing handed over. So a recapitalisation is always a restructuring, and a restructuring is not always a recapitalisation. The one-way relationship is exactly what produces crossed conversations when nobody says which word they mean.

One question separates them cleanly: does the enterpriseThe business itself. The plant, the people, the customers and the operations that produce the earnings, considered apart from the question of who has a claim on them. change, or only the claims on it? If the answer is that the plant, the customers and the earnings are where they were and only the paper has moved, the operation is a recapitalisation. If the answer is that the business itself will be a different business on the other side, the operation is the wider one, and the arithmetic in this guide is only half of what it takes.

What does the recapitalisation look like when it is actually worked?

Verbal distinctions are cheap, so work the variation and watch what does and does not move. Take Rs 690 crore of the Rs 900 crore owed by Meghdoot Coated Products Limited and turn it into shares in the company. The converted Rs 690 crore stops being an obligation. Rs 210 crore of borrowings is left, and against Rs 60 crore of earnings that is 3.5 times.

Now write down the same business with a write-down instead: Rs 690 crore is simply given up by the lenders and not replaced with anything. Rs 210 crore of borrowings is left, and against Rs 60 crore of earnings that is 3.5 times. The two lines are word for word identical. The debt arithmetic of a conversion and the debt arithmetic of a write-down of the same size are not merely similar, they are the same arithmetic, and any measure built only on the ratio cannot show which one happened.

The difference between the two routes sits outside the ratio entirely. Under the write-down, the lenders who gave up Rs 690 crore hold nothing at all in exchange. Under the conversion, the lenders who gave up Rs 690 crore hold shares, so if the business recovers and is worth something in five years, they are the ones who receive that. One thing did not happen in either case: no cash entered the business. A conversion does not fund anything. A conversion rearranges who has a claim and what kind of claim it is, and the bank account is exactly where it was the day before.

Who decides how much debt this business can carry?

Now the central question, and the honest answer is unsatisfying on first reading. Nobody measures it. There is no instrument that reads out how much debt a business can support, no test that settles it, and no authority that publishes a figure. A judgement exists instead, formed by people looking at a business and forming a view about what its earnings will bear over years that have not happened yet. The judgement is written down as a multiple of EBITDA.

Say the judgement is three and a half turns. The sentence is doing enormous work. The multiple of three and a half is a claim about how stable the earnings are, about whether packaging demand holds up, about how much has to be spent each year keeping machines running, about how patient lenders will be, and about whether the earnings could fall again before they recover. None of that is in the record. All of it is compressed into one number that then gets used as though it were measured. The multiple is the assumption, and every figure derived downstream of it inherits that assumption whether or not anybody says so.

There is a household version that makes this concrete. A bank asked how large a home loan a household can carry does not measure anything either. The bank takes the monthly income, applies a rule about what share of income an instalment may be, and produces a figure. Change the share from forty per cent to fifty and the loan the household can supposedly carry jumps, without a rupee of income changing. The figure looked like a measurement and was a policy. Exactly the same thing is happening when somebody says Meghdoot Coated Products Limited can carry three and a half turns.

So how much of the Rs 900 crore has to go?

Because the multiple is a judgement, there is no single answer, and the only responsible way to present the arithmetic is as a range. Work three levels on the locked figures and lay all four columns out at each one. The supported debt is the multiple times the Rs 60 crore of earnings. The write-down is whatever is left of the Rs 900 crore after that. The percentage is that amount over the Rs 900 crore, and the Rs 900 crore is the base that has to be said aloud every single time.

The bottom row is where the table pays off. Read it slowly. Between the most conservative of the three assumptions and the most generous, Rs 60 crore of recovery changes hands. The Rs 60 crore is 6.7 percentage points on the whole Rs 900 crore, and it is decided by a judgement that nobody can measure and that most summaries never state. The record also states these three write-downs as whole numbers, 80, 77 and 73 per cent. A quoted figure of 77 per cent is 76.7 rounded, and it belongs to the middle row and to no other.

One thing in that table looks like a coincidence and is not. The Rs 60 crore of spread is the same figure as the Rs 60 crore of earnings, and the match can make a reader think something has been copied across. It has not. The two outer assumptions are exactly one turn apart, so the distance between them is by construction one turn of earnings, or Rs 60 crore. The match is forced arithmetic rather than corroboration, and if the outer assumptions had been 3.0 and 4.5 times the spread would have been Rs 90 crore and the resemblance would have vanished.

The shape the control produces is the thing to watch. No setting of the assumption creates a rupee or destroys one, so the bar never gets longer and never gets shorter. All that happens is that a line inside a fixed quantity slides, and everything on one side of it is called surviving debt while everything on the other side is called a write-down. The control produces nothing; it only decides where a boundary is drawn, and an assumed sustainable level does exactly that in a real negotiation.

Once the amount is fixed, who is actually paid out of it?

The arithmetic stops here, and the stopping point is itself worth understanding. Everything so far has answered one question: how much. There is a completely separate question, and it is the harder of the two: who receives what is left. Answering the first settles nothing whatever about the second.

Look again at the Rs 900 crore owed by Meghdoot Coated Products Limited and notice that it is not one claim. Rs 620 crore of it is held by lenders with security over an assetA lender's registered hold on a named thing the borrower has, so that if repayment stops that lender can look to that thing before anyone else can., meaning a specific machine or building or receivable stands behind the loan. The other Rs 280 crore is held by parties with nothing but an agreement. Both are owed real money. Both will present an invoice for it. But when the pot is short, those two positions behave nothing alike.

Take the Rs 210 crore that the earnings support at 3.5 times and ask who receives it. Who receives the Rs 210 crore is settled by rankingThe order in which claims are paid out of an amount too small to satisfy them all. Where the amount is enough for everybody, ranking never comes up., and ranking can produce wildly different outcomes from the identical Rs 210 crore. Under one rule the secured lenders take the whole of it and the others take nothing. Under a pro rataIn proportion. Each claim takes the same fraction of what it is owed, rather than one being satisfied in full before the next begins. rule, every claim takes the same fraction of itself and both groups end up part paid. Same pot, same arithmetic upstream, entirely different people disappointed. Working those two rules out in full is what the guides that follow do, and it is the reason this subject exists at all.

Which routes exist, and what does each one move?

Five routes exist, and naming what each one actually does to the two numbers is the test of whether it is a fix or a rearrangement. Reduce the debt or convert it into shares, and the amount owed falls while the earnings sit still. Sell the whole business, and the amount owed is no longer serviced out of annual earnings at all; it is settled out of whatever price the sale produces, so both numbers are replaced by a single pot. Sell a part, and both numbers move together: the division that leaves takes its earnings with it, and the cash it fetches pays down debt. Rebuild the earnings, and the amount owed sits still while the denominator climbs. Rebuilding is the only route that makes the business bigger rather than smaller.

Then there is the fifth, and it is the one that catches people out. Separate the business into two, whether by a demergerSplitting one company into two, each holding part of what the single company held before, so that shareholders end up holding two things instead of one. or by some other division, and on the day it happens neither number has moved. The same total earnings exist; the same total debt exists. The change is in which company each sits in, and therefore which lenders can reach which earnings. A separation is a way of allocating the two numbers between two containers, not a way of changing either one, and treating it as a fix is how a restructuring produces a great deal of activity and no improvement.

None of that makes separation useless. If a healthy operation is being dragged down because it is stapled to a struggling one, separating them can let the healthy one refinance on its own strength. Refinancing changes what the two halves can each carry, even though the totals on the day are unchanged. But that benefit comes from what happens afterwards, and any account of it that skips the afterwards is describing a rearrangement as though it were a repair.

What does this look like away from the balance sheet?

Stripped of the crore, the machinery is familiar. A household runs on one salary of Rs 40,000 a month. The household has taken on obligations whose combined instalment is Rs 52,000 a month. Everybody involved is reasonable and nobody is careless. Rs 52,000 still does not come out of Rs 40,000.

Now watch what does and does not help. Rearranging which lender is paid first changes who is disappointed, and it changes it completely, but the household is still short by Rs 12,000 a month afterwards. Consolidating the obligations into one paper produces a tidier statement and the same shortfall. Extending the tenor lowers the monthly figure and genuinely helps at the margin, and it does not touch the total owed at all. There are only two things that end the problem: the instalment gets smaller, or the salary gets bigger.

A smaller instalment and a bigger salary are the household version of the two numbers, and every proposal in a restructuring is one of them, a rearrangement of who is disappointed, or a delay. Holding that picture in mind is the fastest way to read a restructuring proposal. One question can be put to every part of it: which of the two numbers does this move, and by how much? A proposal that cannot answer that question in one sentence has usually not moved either.

How is this actually used, and by whom?

Four people look at the same Rs 900 crore and want different things from it. The figure means something different to each of them, so all four are worth seeing.

A lender looks at the supported level first and works backwards from it. The amount owed is already known, so a lender is occupied instead by what the business will still be able to service after everything is agreed. Only the serviceable amount will actually be repaid. So a lender argues hard about the multiple, and about whether the Rs 60 crore of earnings is a fair picture of a normal year or a flattering one. A lender is also watching every covenantA promise written into a loan agreement, usually a test the borrower must keep passing. A breach hands the lender rights it did not have the day before, without any money having changed hands. in the existing documents. A breach changes what can be demanded and when, long before any restructuring is agreed.

An analyst wants the range and refuses the point. Confronted with a summary saying a business needs a write-down of a certain size, an analyst who is doing the job asks which multiple produced it and what the figure becomes half a turn either side. The question is not scepticism for its own sake. Asking it is the only way to know whether a conclusion survives the assumption or depends on it, and a conclusion that only holds at one setting is a conclusion about the setting.

An investor considering putting money into a business in this position is really asking a different question again: after the restructuring, what will this company owe, what will it earn, and what share of it will the new money hold? Notice that the write-down percentage does not appear anywhere in that sentence. The shape on the other side is what matters, and the write-down is only the route by which the shape is reached.

And the household described above is doing the same arithmetic without the vocabulary. The household is asking what instalment it can genuinely carry, and that is the same act of judgement that produces a sustainable multiple. The household is also discovering the same uncomfortable fact: the answer depends on what is assumed about the years ahead, and nobody can check the assumption today.

What here is arithmetic, and what is law?

The two halves have completely different rules of evidence, so the division between them is worth being strict about. How much the earnings support is arithmetic, and every step of it is visible above. Whether a particular reduction can be imposed, on whom it can be imposed, by what route and in what order the resulting money is applied is not arithmetic at all. All of that is set in law, and the law changes.

The proportion of lenders that must agree to anything, how long any process runs, what classes exist, and in what order claims are satisfied when a formal process takes over are all settled in law rather than by arithmetic. Every one of those is published, none of them is safely carried in anybody's head, and the two ranking rules used here are illustrations chosen because they are the extremes with visible arithmetic between them. Neither is a statement of what Indian law provides, and which of them applies to any real situation is a question for the law itself.