Coverage Ratios: Building Them and Reading Them Right

A coverage ratio divides the cash available in one period by an obligation falling due in that period. Three decisions define it: which cash, then which obligation, then which period. On Tapti Crossing Infrastructure Private Limited, an invented toll road company, Rs 248 crore over Rs 182.70 crore of debt service is 1.36 times, and the same cash over Rs 119.70 crore of interest alone is 2.07 times.

Compute the two ratios from a set of project accounts

What is a coverage ratio actually measuring?

A single month is easier to picture than a project. A household takes home Rs 60,000/- in March and pays a loan instalment of Rs 24,000/- in the same March. Dividing the first by the second gives 2.5 times. The 2.5 times says something narrow and useful: in that month, the money coming in was two and a half times the money that had to go out on the loan. The ratio says nothing about April, and nothing about whether the loan is large.

Changing one half makes the number stop meaning anything. The same Rs 60,000/- divided by the Rs 18,00,000/- still outstanding on the loan gives 0.033. In a household nobody would call that a coverage ratio. A month of pay and a balance owed are obviously different kinds of thing. Both halves of a coverage ratio are quantities belonging to one stretch of time, and the moment one half becomes a balance standing at a date, what is being computed has stopped being cover altogether.

The trap is far less obvious in a project company, where the figures are large and both are printed in rupees crore in the same statement. Tapti Crossing Infrastructure Private Limited is a single-asset toll road company, formed to build and operate one crossing, with no other business and no second source of cash. In the modelled year it collects revenue of Rs 310 crore and spends Rs 62 crore running the crossing, so earnings before interest, tax, depreciation and amortisation, or EBITDAEarnings before interest, tax, depreciation and amortisation. Revenue less the costs of running the business, before any charge for financing, tax or wear on the assets., is Rs 248 crore. Its borrowings stand at Rs 1,260 crore. Both numbers describe the same company and only one of them belongs in a coverage ratioCash available in a period divided by an obligation that falls due in that same period. The answer is written as a number of times..

Revenue, operating cost, the Rs 248 crore that survives, interest of Rs 119.70 crore and scheduled principal of Rs 63 crore are all flows: each accumulated across the twelve months and then stopped being counted. The Rs 1,260 crore of borrowings is not a flow at all. The Rs 1,260 crore is a balance, photographed at one instant, and it will be a different number the instant after a principal payment lands. A ratio needs both halves to be the same kind of thing, and here that means a period quantityA figure that accumulates across a stretch of time, such as a year of revenue, rather than one that stands at a single date, such as a loan balance..

Which three decisions define any coverage ratio?

Two competent people can produce two different coverage ratios for the same company and the same twelve months. Before there is any division there are three choices, and each is a judgement rather than a rule. The first is which cash. The starting point may be EBITDA of Rs 248 crore, or EBITDA after a tax charge, or EBITDA after tax and after whatever the crossing spends that year on resurfacing and repair. Each answers what cash was available, and each is smaller than the one before it. The general name for whatever is chosen is cash available for debt serviceThe figure a coverage ratio starts from: what the business produced in the period and could actually direct at its lenders, after whatever costs the person computing it decides must come first., and the name is no help at all in deciding what goes into it.

The second choice is which obligation. Interest alone. Interest plus scheduled principal, the sum usually called debt serviceInterest plus the principal a schedule requires to be repaid in the same period. It is the total the lenders are due in that stretch of time.. Or every rupee that actually left the account towards borrowings in the year, including anything repaid ahead of schedule. Three different denominators, all honestly describable as what the company owed its lenders.

The third is which period. A year, a quarter, or a rolling twelve months ending at the latest test date. The choice of period matters more than it looks. A road collects tolls unevenly across a year, so a quarter chosen well or badly moves the answer without anything about the crossing changing.

A coverage ratio is not a formula to be looked up: it is three choices, and two people making different choices from the same accounts will get different answers while both of them have divided correctly. The three group headings in the panel above are those three choices.

How is the debt service cover ratio built from this project?

The next step is to build one by hand, writing down each of the three choices in turn. The company is Tapti Crossing Infrastructure Private Limited.

Choice one, the cash. Take EBITDA of Rs 248 crore, being revenue of Rs 310 crore less operating cost of Rs 62 crore. No tax charge and no maintenance spending appear anywhere in this record, so Rs 248 crore is the most it can support. A figure with nothing deducted from it is a ceilingThe highest a figure could be, because every deduction that might apply has been left out. A ratio built on a ceiling is itself a ceiling., every ratio built on it inherits that property, and it has to be said wherever the figure travels.

Choice two, the obligation. Interest on Rs 1,260 crore at the project's own contracted rate of 9.5 per cent is Rs 119.70 crore. Scheduled principal in the modelled year is Rs 63 crore. Together they are Rs 182.70 crore.

Choice three, the period. The modelled year, for both halves, and no other year at all. The record carries no year by year schedule to work from.

The writing out of the three choices is the calculation, and the division is the easy part. Rs 248 crore over Rs 182.70 crore is 1.357417, written 1.36 times. Naming the cash as a ceiling, naming the obligation as interest plus scheduled principal rather than everything paid, and naming the period as one modelled year is what produced a usable figure.

How is interest cover built, and which version can this record not support?

Keep the cash where it is and change only the denominator. Interest alone, at Rs 119.70 crore, over the same Rs 248 crore in the same year, is 2.071846, written 2.07 times, and it is called interest coverCash available in a period divided by the interest falling due in that period, with no principal in the denominator at all.. Nothing about the project changed. Only the question did, from can the year pay everything the schedule asks to can the year pay the rent on the money. A reader shown 2.07 times will feel better about the project than one shown 1.36 times, and neither has been told anything the other has not: the difference is the Rs 63 crore of scheduled principal and nothing else.

The depreciation charge is a rough acknowledgement that the asset is being used up, so interest cover is very often struck on EBITEarnings before interest and tax. EBITDA less the depreciation and amortisation charged for the period, so it is a figure struck after wear on the assets has been recognised., earnings before interest and tax, rather than on EBITDA. For Tapti Crossing Infrastructure Private Limited that version cannot be computed at all: this record carries no depreciation charge for the project company, so there is no numerator, and producing one would mean inventing a figure and then reporting it as though it had been found.

Naming which version was built is part of the figure, and a bare interest cover number with no cash base named is unusable. Two point zero seven times on EBITDA and two point zero seven times on EBIT would describe very different projects, and the second is simply not available here.

What is headroom, and how is it stated usefully?

A ratio of 1.36 times can also be read as a distance. The obligation is Rs 182.70 crore and the cash is Rs 248 crore, so Rs 65.30 crore of that cash can disappear before the year stops covering what it owes. The Rs 65.30 crore gap is the headroomThe amount of cash that could be lost before a coverage ratio reaches 1.00 times, expressed in rupees or as the fall in some line that would cause it., and saying it well is most of the value a tool like this adds.

There are three honest ways to say it and they are not equally useful. As rupees, the headroom is Rs 65.30 crore. As a share of the cash it is 26.3 per cent. Rs 65.30 crore over Rs 248 crore is 26.330645 per cent. As turns of cover it is 0.36, being 1.357417 less 1.00, or 0.357417. All three are the same fact.

The fourth way is the one to use. Hold operating cost at Rs 62 crore and ask how far revenue can fall before EBITDA drops to Rs 182.70 crore. Revenue would have to reach Rs 244.70 crore, so it can fall by Rs 65.30 crore. On Rs 310 crore that is 21.064516 per cent, written 21.1 per cent. Revenue is the line somebody will actually be watching every month, and a percentage of EBITDA means nothing to the person watching it, so the revenue version travels best.

Consider a tea stall whose owner knows the day is fine if forty customers come. Telling that owner the cushion is 26.3 per cent of gross profit says nothing they can act on. Telling them the day goes wrong below thirty-two customers hands them a working instrument. A toll road manager watching daily traffic is in exactly that position.

How does the whole worked year read from end to end?

The whole computation sits in one place below. The panel above opens with these same settings. For Tapti Crossing Infrastructure Private Limited in the modelled year: revenue Rs 310 crore, less operating cost Rs 62 crore, with tax paid and major maintenance both nil in this record, gives cash of Rs 248 crore, and that figure is a ceiling. Interest on borrowings of Rs 1,260 crore at the project's own contracted rate of 9.5 per cent, over a full year, is Rs 119.70 crore. Scheduled principal is Rs 63 crore. Debt service is Rs 182.70 crore.

1. Divide the cash by debt serviceRs 248 crore over Rs 182.70 crore is 1.357417, written 1.36 times.
2. Divide the same cash by interest aloneRs 248 crore over Rs 119.70 crore is 2.071846, written 2.07 times.
3. Take the difference as rupees of headroomRs 248 crore less Rs 182.70 crore is Rs 65.30 crore of cash that could be lost before cover reaches 1.00 times.
4. Translate the headroom into the line somebody watchesHolding operating cost at Rs 62 crore, Rs 65.30 crore is a 21.1 per cent fall in revenue, from Rs 310 crore to Rs 244.70 crore.

On those settings the panel returns 1.36 times, Rs 65.30 crore of headroom, 0.36 turns above 1.00 and a 21.1 per cent fall in revenue to close it, and 2.07 times as soon as the obligation is switched to interest alone. The panel does nothing the four steps above did not already do in words.

The wrong answer teaches more than the right ones. Finish with it. Set the obligation in the panel to the borrowings balance: Rs 248 crore against Rs 1,260 crore returns 0.196825, shown as 0.20, and the reconciliation strip turns red. The 0.20 is not a low coverage ratio. A leverage measure has been computed upside down: the division puts a quantity belonging to one year over a balance standing at one date. The honest way up would be Rs 1,260 crore over Rs 248 crore, or 5.08 times. The three decisions catch it before it happens: the moment the period the denominator belongs to is written down, a balance has nowhere to sit.

What does a coverage ratio of 1.36 times not say?

A coverage ratio is a narrow instrument, and it stays useful only while three absences are held in mind.

The first is every other year. The 1.36 times belongs to the modelled year of Tapti Crossing Infrastructure Private Limited and describes no other twelve months. A traffic forecast and a year by year schedule are what would show whether year two is better or worse, and this project record supplies neither. Estimating around the gap would mean writing a number nobody supplied.

The second is the shape of the whole borrowing. Neither the length of the right to operate the crossing nor the length of the borrowing appears in this record, so cover says nothing about whether the debt retires in time. A project can cover every year handsomely and still reach the last year of its right to collect tolls with a balance outstanding, and cover would never have hinted at it.

The third is any verdict. 1.36 times is not comfortable or thin as a property of itself. A ratio becomes one or the other only against a requirementA level somebody set in a specific agreement for a specific ratio, defined in that agreement's own words. It is a fact about the document, not about ratios in general. that somebody wrote down, and this record states none. The three absences are exactly what a reader most often supplies from imagination, and the supplied version then travels onward looking like part of the analysis.

Why do two people get different answers from the same accounts?

A coverage ratio is rarely built from scratch. One usually arrives already made, in a note or a slide, and the work is deciding what it is worth. The three choices are independent, so two answers can sit far apart with no arithmetic error between them: one person starts from EBITDA where another starts from cash after tax, one counts scheduled principal where another counts every rupee that went to lenders, and one uses the financial year where another uses a rolling twelve months to the latest test date.

A coverage ratio arriving from somebody else calls for three questions: which cash, then which obligation, then which period. Until all three are answered, what has been received is a number rather than a measurement. The three questions take about fifteen seconds, and they are the difference between using a figure and repeating one. If the person cannot answer them, that is the finding, and it is worth more than the ratio was.

How does a project finance desk actually use this?

Three parties around Tapti Crossing Infrastructure Private Limited do this arithmetic for a living, and their uses are not the same.

The project lenders build the ratio forward rather than backward. The lenders are interested in the profile of the ratio across every year the borrowing runs, and the modelled year above is one point on that profile. From a single year they take the headroom, expressed as it is expressed here: how far revenue can fall before the year stops covering. A lender told 21.1 per cent has something to set against their own view of traffic. A lender told 26.3 per cent of EBITDA has been told a fact about a denominator.

The sponsors, who signed the equity cheque of Rs 540 crore, read the same bar from the other end: what is headroom to a lender is cash to them. Rs 248 crore of EBITDA less Rs 182.70 crore of debt service leaves Rs 65.30 crore, and it is the same Rs 65.30 crore in both readings. The lender calls it the buffer before 1.00 times; the sponsor calls it what the year produced for the people who put the equity in. Neither is wrong, and that identity is why lenders and sponsors argue about the same rupees.

An analyst covering a listed sponsor sits outside both, working from what is published rather than from the model. A ratio lifted from a presentation into a spreadsheet without its three choices attached will sit there looking comparable to a ratio built quite differently, and nothing about either figure will flag the problem later. The choices cannot be recovered from the figure afterwards, so the practical discipline is to record all three in the same cell note as the number, every single time.

Set an ordinary corporate borrowing beside this. Harivansh Packaging Limited is a listed packaging manufacturer with many customers and several plants, so a lender looking at its cover sees a business that could sell a plant, cut a product line or borrow elsewhere. Tapti Crossing Infrastructure Private Limited has one crossing, one stream of tolls and no recourse beyond the project. Its coverage ratio therefore carries more of the whole assessment than any single ratio would for a trading company.

Where does a coverage ratio show up in public?

Most coverage ratios never leave a model. Where a listed sponsor says something publicly about a project it holds, what may be said and when is a disclosure question for the Securities and Exchange Board of India (SEBI) rather than for anybody computing ratios; the project company itself, its incorporation, shareholding, charges and filings, sits with the Ministry of Corporate Affairs. Both should be confirmed at source.