The Debt Maturity Profile: Reading the Refinancing Wall

A debt maturity profile is a table of when borrowings fall due, and the refinancing wall is the year that dwarfs the others. At Nirjhar Industries Limited, invented, Rs 2,160 crore falls due in year 3, being 50.0 per cent of all debt and 1.67 times a full year of earnings before interest, tax, depreciation and amortisation. Rs 960 crore of that one year is owed to a single lender.

Everything else in this sequence has been standing inside a bank, looking at the bank's own funding. A wall of maturing debt is a liquidity event for a company in exactly the way a wave of deposit withdrawals is for a bank. So stand on the other side of the table for a while, inside the borrower. The money has to be there on a stated morning, and if it is not, wanting it does not produce it.

What is a debt maturity profile, and what does the total not reveal?

A debt maturity profileA table showing how much borrowing falls due in each future period, which is the only place the timing of a debt book is visible. is about as simple an artefact as finance produces. One row per future period, one figure per row, and the figures add to the total borrowing. There is no model in it, no assumption, nothing to calibrate. The profile is a list of dates and amounts that somebody has already agreed to.

And yet no other document in the accounts carries the information a profile carries, and that is why the profile is printed separately. Nirjhar Industries Limited, a steel and alloys maker, has total borrowings of Rs 4,320 crore. Take that total and ask what it establishes about the company's next twelve months. A total is a size and a profile is a timing, and the two are unrelated pieces of information, so the honest answer is nothing whatsoever. Rs 4,320 crore falling due evenly over six years is one company. The same Rs 4,320 crore with half of it landing in a single year is a completely different company, and the balance sheet prints them identically.

The shape of a profile comes from how the individual borrowings were written. A borrowing repaid in instalments across its life is amortisingA borrowing repaid in instalments across its life, which spreads a profile rather than concentrating it., and it spreads itself across several rows on its own. A borrowing repaid in one amount at the end is a bullet maturityA borrowing repaid in one amount at the end rather than in instalments, which is what builds a wall in the first place., and it lands entirely in one row. Nobody sets out to build a wall. A wall is what appears when three or four bullets written in different years by different lenders happen to come due in the same one, and nobody was looking at the combined picture when the third one was signed.

What do the shares and the cumulative column add to the picture?

Two columns turn a list of amounts into something comparable across companies. The first is each row as a share of the whole book. The second is the running total of those shares, and the running total shows how fast the whole book has to be dealt with. No individual row answers that question.

Read down the last column and the cumulative share does the work. Ten per cent of everything this company has borrowed falls due inside one year, a quarter inside two, and then it jumps. Three quarters of every rupee the company has borrowed has to be repaid or replaced inside three years, and that single sentence is worth more than the whole balance sheet for judging whether the next few years are comfortable. After that the pace collapses: only a quarter of the book survives past year three, and only 5.0 per cent of it is out beyond year five.

One number, four meanings, and why the object is named every time

Before going further, a habit worth building. Round rupee figures repeat inside any single case, and this company is a good example of how badly that can trip a reader. Rs 432 crore is four different objects here: the amount falling due in year 1, the interest paid for the year, the depreciation charge, and the profit before tax. Rs 540 crore is both the year 4 maturity and the cash the company holds. Rs 324 crore is both the year 5 maturity and the profit after tax. Rs 3,780 crore is both the cumulative amount due through year 4 and the net debt. None of these is an error and none of them is a coincidence worth chasing; they are simply what happens when one set of figures is built to be readable. The rule that prevents the error is to name the object rather than the bare number, every single time. Write the year 1 maturity of Rs 432 crore, never just Rs 432 crore.

What makes one bucket a refinancing wall rather than merely the largest?

Every profile has a largest row. Being the largest is arithmetic, not a finding. A refinancing wallA period in which an unusually large share of the debt book falls due at once, so that the borrower depends on lenders agreeing rather than on trading well. is something else, and three tests separate the two.

The first is the row against its neighbours. Year 3 carries Rs 2,160 crore. The year before carries Rs 648 crore and the year after Rs 540 crore, so the wall is three and a third times the first neighbour and four times the second. A largest row that was a fifth bigger than its neighbours would just be a lumpy schedule. The second test is the row against the whole book, and 50.0 per cent of everything borrowed inside one year is the definition of concentration. The third test is the row against what the business actually generates in a year, and that third test is worked in full below. A wall is a row that is large relative to its neighbours, to the book and to the business at the same time, and a row that fails any one of those three tests is just the biggest year.

The everyday version sits in a lot of households. A household with a home loan, a vehicle loan and a small personal loan is not in difficulty because the total is large. The household is in difficulty in the one month when the insurance premium, the school fee and the property tax all land together. The crowded month asks for cash the salary does not produce, and no amount of being comfortable in the other eleven months helps. The company version is the same event with a longer clock. Notice what makes both cases uncomfortable. The money was always owed, and the discomfort comes from its being owed all at once.

The part that matters most is the part the borrower controls. Trading well is inside the business. RefinancingReplacing a borrowing that is falling due with a new one, which needs somebody to agree and is therefore not within the borrower's control. is not: it needs a lender to agree, on the day, at a price. A wall converts a company's position from something it can influence with its own effort into something that depends on other people saying yes, and it does that on a date fixed years earlier.

How big is the wall against a year of earnings and a year of cash?

A rupee figure means nothing until it is put beside something the business produces. Three yardsticks are worth using. All three are used precisely because they give three different answers.

Against earnings first. Nirjhar earns Rs 1,296 crore before interest, tax, depreciation and amortisation, on revenue of Rs 10,800 crore, a 12.0 per cent margin. The wall of Rs 2,160 crore is 1.67 times a full year of that. Against cash actually in hand next. The company holds Rs 540 crore, enough to cover the year 1 maturity of Rs 432 crore 1.25 times over and enough to cover 25.0 per cent of the wall. Third, and this is the one that decides things, against cash before capital spendingEarnings before interest, tax, depreciation and amortisation less interest and tax, which is what is left before anything is spent on assets.: Rs 1,296 crore of earnings less Rs 432 crore of interest less Rs 108 crore of tax leaves Rs 756 crore a year, and the wall is 2.86 years of that.

The wall is 1.67 years of earnings and 2.86 years of cash, and the gap between those two numbers is the whole reason the earnings measure flatters every borrower who quotes it. Earnings before interest and tax is a figure computed before the two payments the company has no choice about. Cash before capital spending is computed after them, and it is a smaller number for every borrowed rupee on the balance sheet.

What share of the wall has to be refinanced whatever else happens?

One question decides whether the wall is an inconvenience or a problem, and it has an exact answer rather than a judgement. Suppose the company did everything it possibly could. Suppose it put every rupee of cash it holds into the wall, and suppose it also put an entire year of cash generation into it and spent nothing at all on the business for that year. How much would still be left?

Cash of Rs 540 crore plus one year of cash before capital spending of Rs 756 crore is Rs 1,296 crore. Against a wall of Rs 2,160 crore, that is 60.0 per cent covered and Rs 864 crore left over. At least 40.0 per cent of this wall must be refinanced whatever else happens, and that figure is solved rather than estimated. The 40.0 per cent floor is not a stress case or a pessimistic view. It is what remains after the most extreme self-help the company could possibly attempt. A year of spending nothing on plant is not a plan a business survives, so no operating steel maker could attempt even that.

What does repricing the wall do to interest cover?

The profile says when the money is due. The profile says nothing whatever about the price at which the money will be replaced, and that silence is where most of the damage hides. The company's Rs 4,320 crore of borrowing carries an average cost of 10.0 per cent, and that rate on that book is where the Rs 432 crore of interest comes from. Interest coverEarnings divided by interest, being how many times over the profit covers the cost of the debt, and always stated with the earnings measure named. is therefore 2.00 times on earnings before interest and tax of Rs 864 crore, and 3.00 times on the Rs 1,296 crore measured before depreciation.

Now suppose the Rs 2,160 crore wall gets refinanced at a higher rate while the rest of the book stays where it is. Every percentage point on Rs 2,160 crore adds Rs 21.6 crore of interest, so the arithmetic is one multiplication. Two points more costs Rs 43.2 crore and takes cover from 2.00 to 1.82 times, and ten points more costs Rs 216 crore and halves profit before tax from Rs 432 crore to exactly Rs 216 crore. The curve is flat at first and then steepens. A small repricing therefore looks harmless and a moderate one does not.

Three of those rows repay a second look. At 6.67 points, six and two thirds exactly, cover lands on 1.50 times, and 1.50 times is the level a great many loan agreements name as the one a borrower has promised not to go below. At 10.0 points profit before tax has halved, and yet the company is still profitable and still covers its interest twice over on the measure before depreciation, so nothing in the accounts screams. At 20.0 points profit before tax is nil and cover is exactly 1.00, meaning the whole of the year's operating profit goes to the lenders. The shareholders get the rest, and the rest is nothing.

Why is the borrower's wall the lender's concentration as well?

Turn the table around. Of the Rs 2,160 crore falling due in year 3, Rs 960 crore is owed to Vindhya Commercial Bank Limited, an invented lender, and Vindhya is the bank the rest of this sequence has been standing inside. The same Rs 960 crore is 44.4 per cent of the borrower's wall and 57.1 per cent of the bank's Rs 1,680 crore of funded exposure to this name, and this name is the largest single exposure the bank carries.

The borrower's worst year and the lender's largest single decision are the same rupees, and neither party's own report shows the other half of the picture. The company's profile shows a year with a big number in it and does not say who holds it. The bank's exposure report shows a large position against one name and does not say that most of it comes back for a decision in the same twelve months. Put the two documents on one table and a fact appears that is on neither of them alone.

A refusal is expensive on both sides for the same reason. If the bank declines to refinance, the company has a liquidity event. If the bank agrees, it has just renewed 57.1 per cent of its largest single position for another term. Renewing that much of one name is a concentration decision rather than a routine rollover. Neither party gets to treat the day as administrative.

The Rs 960 crore and the Rs 1,680 crore are facts about who holds what, and nothing further is read out of them. The internal grade on this name, the probability of default, the exposure at default and the expected loss are each covered separately.

What can a profile never reveal?

A profile has hard limits, and they are worth stating plainly. Three things, and each is a hard limit rather than a caution.

First, there is no weighted average lifeThe average time to repayment weighted by amount, which cannot be computed when the last bucket has no end date. to be had from this profile. The method is to multiply each amount by its year, add, and divide by the total, and it needs a year on every row. The last row reads beyond year 5 and carries Rs 216 crore with no year attached at all. A floor can be put under the answer by giving that row the earliest year it could possibly carry, and that floor is 3.03 years. There is no ceiling whatsoever. Anybody who quotes a single average life for this book has silently assigned a date to a row that has none. Note carefully that the 3.4 year figure the lender reports is a weighted average remaining life on the lender's own Rs 1,680 crore of funded exposure to this name. The lender's figure measures a different population, and the two must never be swapped.

Second, no free cash figure. The Rs 756 crore is stated throughout as cash before capital spending precisely because this case carries no capital expenditure figure at all, and a steel maker that spends nothing on its plant is not a going concern. The moment an unavoidable capital number is subtracted, the coverage gets worse, never better, and the 40.0 per cent floor computed above rises.

Third, no asset sale can be sized. Total assets are Rs 9,180 crore and nothing below that total is given, so a sentence beginning with the phrase they could always sell something has no arithmetic underneath it here. The honest position is that the measurement was not made, rather than making one up.

Who actually reads a maturity profile, and what do they do with it?

Four people read this table, and they read it for four different things. Anybody building one is better off knowing all four.

The credit officer at the lending bank reads it for the year that concentrates. The officer's real question is whether the money the bank is putting out comes back on a date when everybody else is asking for theirs at the same time. On that day the borrower's options are worst and the negotiation is hardest. The same day turns the bank's own position into a decision rather than a routine renewal. The profile therefore sits in the credit file next to the exposure summary and not somewhere at the back.

The analyst reads it for what happens to the earnings if the concentrated year reprices. The whole of the ladder above is one multiplication, and an analyst who has done it knows the level of repricing at which cover reaches 1.50 times before anybody has to have a conversation about it. The practical value of a profile is that it converts a vague worry about a leveraged company into a specific question with a date and a number attached.

The company's own treasurer reads it as a work list. The useful time to refinance is when nothing is urgent and the borrower still has options, so a wall in year 3 is a task that starts in year 1. Waiting until the wall is six months away means negotiating from the one position no borrower wants, and that position is having to.

The mechanism does not change with the size of the number, so the household version is the same story. A person with a home loan that has a large payment falling due when the fixed period ends, a vehicle loan and a personal loan is not in trouble because of the total. The household is in trouble if all three land in one year and the salary produces enough for two of them. The fix in both cases is identical and it is not clever: look at the schedule early, and move something out of the crowded year while somebody is still willing to move it.