Debt to equity, and the two questions it never asks

What the ratio actually divides

Debt to equity divides what a company has borrowed by what its shareholders own on the books. Borrowings of ₹60 crore against shareholders' funds of ₹40 crore give a ratio of 1.5. Both figures here are illustrative, and both are accounting numbers rather than measurements.

The numerator is reasonably intuitive: money the company owes to lenders, on which it pays interest and which it has to give back. The denominator is less so. Shareholders' funds — share capital plus accumulated reserves, the figure also called net worth — is not cash sitting anywhere. It is a residual: what would be left for shareholders if every asset fetched exactly its carrying value and every liability were settled at face. The idea is the same one set out in net worth, applied to a company instead of a household.

So the ratio answers a narrow question. For every rupee the owners have in the business at book value, how many rupees have been borrowed alongside it? That is worth knowing.

What it cannot tell you is whether the borrowing is a problem, and the rest of this article is about why. A ratio is only as good as its denominator and the accounting policy behind it, and in this particular ratio both halves move for reasons that have nothing to do with how much risk the company is running.

The denominator is the soft half

Start with the part almost nobody interrogates, because the mechanism is the most surprising thing on this page.

Take the illustrative company above: ₹60 crore borrowed, ₹40 crore of shareholders' funds, ratio 1.5. Now it buys back ₹15 crore of its own shares, or pays the same amount out as a special dividend. Borrowings have not moved. Shareholders' funds fall to ₹25 crore, because the cash left and the reserves went with it. The ratio is now 2.4.

The company's leverage rose by 60% without it borrowing a single rupee — every bit of the movement came from the denominator. Read the ratio alone across those two dates and you would describe a company that had taken on debt. It had returned capital.

The same works in reverse, and this is the version that flatters. A rights issue, a preferential allotment, or a year of retained profit all raise shareholders' funds, so the ratio falls while the borrowings sit exactly where they were. A falling debt-to-equity ratio is not evidence that debt was repaid. It is evidence that the fraction got smaller.

Four more things move book equity without anything happening in the business:

• Revaluation — where a policy permits carrying an asset at a revised value, the uplift lands in reserves. The land is the same land.
• Goodwill — an acquisition paid for above book value parks the excess on the balance sheet. Write it off later and equity drops in one step, pushing the ratio up on a day when no debt was raised.
• Accumulated losses — these eat reserves. Enough of them and shareholders' funds go negative, at which point the ratio turns negative too. A negative debt-to-equity is not a low one; it is the arithmetic telling you the denominator has stopped meaning anything.
• Depreciation policy — how fast assets are written down feeds straight through retained profit into reserves. Two companies with identical plants and identical borrowings can carry different equity, and therefore different ratios, because they chose different useful lives.

None of that is manipulation. It is what a residual figure does. But it means the denominator carries the accumulated effect of every accounting choice the company has ever made, which is a strange thing to divide by and then treat as a risk measure.

And the numerator is not one line either

The borrowings figure looks harder to argue with. It is not, because what counts as debt has changed and because some of it does not sit under a heading with the word debt in it.

The largest shift of the last decade is leases. Under the current accounting standard a lessee brings the future rentals onto the balance sheet as a liability, with a matching right-of-use asset, instead of leaving them in a note. For a business that owns its factories, this changed little. For a retailer that leases every store, it added a liability the size of a decade of rent to a balance sheet that had shown almost no borrowings the year before.

Add an illustrative ₹30 crore of lease liabilities to our company and its ratio goes from 1.5 to 2.25. Nothing about the shops, the sales or the rent cheques changed. Only the presentation did. Any comparison spanning that transition, or spanning two companies where one owns and one leases, is comparing conventions rather than risk.

Four other places borrowing hides:

• Current maturities — the slice of long-term debt due within twelve months. Presentation rules have moved this line around over the years, so add the pieces up yourself rather than trusting a heading to catch them all.
• Acceptances and bill discounting — a bank pays the supplier now and the company repays the bank later. Economically that is borrowing; it can sit inside trade payables, where a ratio built from the borrowings line will miss it.
• Guarantees to subsidiaries and associates — not a liability until called, so they appear as a contingent-liability note rather than in the total. They are still an obligation, and they belong to the parent when they crystallise.
• Standalone against consolidated — a holding company can look almost unborrowed on its own accounts while the operating subsidiaries carry everything. The consolidated statements are where the group's real position sits.

Then there is the choice of gross or net. Netting cash off borrowings gives a tidier number and is often the fairer one, but it assumes the cash is actually available — not pledged against a facility, not trapped in a subsidiary or a jurisdiction the money cannot easily leave, not already committed to a project. Net debt is a useful figure and a set of assumptions, and the assumptions are not printed next to it.

Where the risk actually sits: the maturity profile

Here is the part that decides outcomes, and the ratio does not contain it.

Two illustrative companies. Both have borrowed ₹60 crore against shareholders' funds of ₹40 crore. Both show a debt-to-equity of 1.5. Both earn operating profit of ₹18 crore and pay interest of ₹6 crore, so both show interest cover of 3×. On every ratio a screen can compute, they are the same company.

Company B has a repayment problem it can plan around: find ₹6 crore a year from operations, ten times. Company A has a refinancing problem, and refinancing is not a task it controls. It depends on a lender's appetite on a particular date — and lenders' appetite shrinks in exactly the conditions that also shrink Company A's profits.

Debt rarely kills a company by being large. It kills by being due — and being due at a moment when the money to refinance it has become expensive or unavailable. Call it the refinancing wall. The distinguishing feature of the failure is that nothing on the balance sheet deteriorated first; the date simply arrived.

The rate basis compounds it. Company A's floating rate resets, so its interest bill is set by conditions rather than by contract, and it rises in the same environment that makes rolling the loan harder. Two exposures that look separate are one exposure. How a benchmark reset actually works is in interest rates explained.

The maturity profile is disclosed in the notes to the accounts, not on the face of the balance sheet, which is precisely why it is missing from every screening column in existence. It is also the single most decision-relevant thing about a company's borrowings.

The flow question: can the profit carry it?

Debt to equity sets one balance-sheet total against another, and both are levels measured on a single day. Interest is paid out of a flow.

The measure that closes the gap divides operating profit by the interest bill — interest coverage. On our illustrative figures, ₹18 crore of operating profit against ₹6 crore of interest is a cover of 3×: profit could fall by two-thirds before interest stopped being covered from trading.

That single number does more work than the leverage ratio, because it is denominated in the thing that actually pays lenders. And it moves faster. Knock 40% off operating profit — an ordinary year for a cyclical business — and ₹10.8 crore against ₹6 crore is a cover of 1.8×. The balance sheet has not changed at all. The debt-to-equity ratio still reads 1.5.

Four things to hold against interest coverage, in the spirit of not swapping one lazy number for another:

• It ignores principal — the largest payment most borrowers make. Add Company B's ₹6 crore of annual repayment and the honest denominator is ₹12 crore, so ₹18 crore of profit covers total debt service 1.5× rather than 3×. That is a different-looking company, from the same accounts.
• Which profit line — cover struck on profit before depreciation is a larger number than the same cover struck after it, and one that folds in other income is measuring something other than trading. Two people can publish different coverage for the same company and both be arithmetically right, so read what went into the numerator before comparing one figure with another.
• Operating profit is not cash — it is struck after accruals and before working capital. A company whose profit is genuine but whose receivables are lengthening is servicing debt out of a number it has not collected.
• Capitalised interest is not in the interest line — borrowing costs attributable to an asset under construction get added to the asset instead of charged to profit. The company is paying it. The coverage ratio cannot see it.

So the pair to read together is the level and the flow: how much is owed, and how comfortably the trading profit carries what it costs. Neither alone is informative, which is the general shape of the problem with single-ratio analysis.

Why the same ratio means different things

Because debt capacity is a function of how volatile the cash flow is, not of the industry the company files itself under. Two companies can carry the identical ratio and be describing two different arrangements.

A business whose revenue arrives under long contracts — a regulated utility, an operator paid an availability charge whether or not anyone consumes anything — can service a fixed schedule of repayments from a flow that is itself close to fixed. The asset is long-lived, the cash is contracted, and the debt is matched to both. What a high ratio describes there is the match between the financing and the asset — which is a statement about structure, not a verdict on the company.

A cyclical manufacturer has the same fixed schedule sitting on top of a flow that halves. Its revenue depends on a commodity price or a capital-expenditure cycle it does not control, and the years when it most needs to make repayments are the years it earns least. The identical ratio describes something genuinely more fragile, because the two sides of the arrangement are mismatched in time.

The mechanism is the mismatch, not the sector. Which is why sector labels are a poor proxy: an infrastructure operator with a contracted offtake and one selling into a merchant market sit in the same industry bucket and carry entirely different capacity for the same debt.

Lenders and financial companies are a further category error. For a bank or a non-banking finance company, borrowing is the raw material — it takes money in at one price and lends it out at another, so a leverage ratio that would be alarming in a factory is simply what the business is. Their solvency is governed by capital adequacy rules built for exactly that reason, and comparing a lender's debt to equity with a manufacturer's is not a comparison at all. Where the risk in a lending book actually sits is the subject of credit risk.

The specific mistake worth recognising: sorting a mixed list of companies on a single debt-to-equity column, top to bottom, and reading the top as risky and the bottom as safe. The column is comparable within a set of companies that finance similar assets against similar cash flows. Across a full market it ranks accounting conventions and business models, and presents the result as a risk order. A screen can only narrow a universe on the fields it holds; the interpretation is not in the field.

Why some debt is not a defect

The nuance most coverage skips: a company with no borrowings is not thereby a better financed company. Debt is cheaper than equity, for two structural reasons, and refusing it has a cost that never appears anywhere.

The first reason is priority. A lender is paid before shareholders and can enforce; a shareholder is paid last and cannot. Because the lender's claim is safer, it is satisfied with less. Equity's cost is not a number on any statement — it is the whole residual claim on every future rupee of profit, uncapped, handed over permanently. Interest is a defined, finite, contractual amount. That is a real difference in price, and it is invisible precisely because the expensive one is never invoiced.

The second is tax. Interest is deducted before profit is taxed; dividends are paid out of profit already taxed. Assume a tax rate of 25% purely to do the arithmetic — the point is the mechanism, not the rate, which a Finance Act can move. An illustrative 9% coupon then costs the company 6.75% after the tax it saves. The same 9% promised to a shareholder costs 9%. Working through what the same borrowing does to a household, where no such deduction applies to most of it, is good debt against bad debt.

The consequence is that a moderate amount of borrowing raises the return on the shareholders' money, and here is the arithmetic. Illustrative company, ₹100 crore of assets earning 12% before interest, so ₹12 crore. Financed entirely by shareholders, they earn 12% on their ₹100 crore. Financed half by debt at 9%, interest is ₹4.5 crore, ₹7.5 crore is left, and the shareholders earn 15% on their ₹50 crore. The assets did not become more productive. The owners simply kept everything the assets earned above what the lender was owed.

Which is the trade-off, stated properly, because the arithmetic runs both ways. Leverage amplifies in both directions and the schedule is fixed in both. Let the same assets earn 6% instead of 12%: ₹6 crore, less the same ₹4.5 crore of interest, leaves ₹1.5 crore — 3% on ₹50 crore of equity, against the 6% the unlevered version would have earned. A halving of the underlying return has more than halved the owners' return, because the interest bill did not halve with it.

And there is a second cost that is not in any ratio. Borrowing sells flexibility. The lender gets dates, and usually covenants — conditions on leverage, on coverage, on what can be sold or pledged — and a breach can accelerate the loan in a bad year rather than a good one. Equity never demands a payment on a date. The cheaper instrument is cheaper because it is stricter, and the strictness is billed in the one year the company can least afford it.

So the honest reading is that debt to equity measures how much of that trade has been made. It does not say whether the trade was a good one, and it cannot, because the answer depends on the stability of the cash flow, the shape of the repayment schedule and the price paid — none of which is in the ratio.

Reading it without turning it into a rule

What the ratio is genuinely good for is narrow and worth stating plainly, because the alternative to a threshold is not vagueness — it is a better question.

1. As a change over time in one company — provided you first check whether the numerator or the denominator moved, and why. Those are different events with the same arithmetic signature.
2. As a comparison within a genuinely comparable set — companies financing similar assets against similar cash flows on the same accounting basis, all consolidated or all standalone.
3. As a prompt, not a verdict — a number that has moved is a reason to open the notes to the accounts, which is where the maturity profile, the rate basis, the covenants and the contingent liabilities live.

The three questions that follow the ratio, in order: when does it fall due, what does the operating profit cover it by, and how stable is that profit? Answer those and the ratio has done its job, which was to make you ask them.

None of this reduces to a level. Two companies at the same ratio can be running completely different risks, and a rule expressed as a single number would be a decision dressed as arithmetic — which is the failure this whole article is about.

Working it against real statements

Everything above is checkable against a company's own filings, and the tedious part is assembling the series rather than doing the sums.

FNOTrader's Stocks app carries balance sheet, profit and loss and cash flow series per listed company on its fundamentals page, and exposes debt to equity as one of the fields the custom-query screener can filter and sort on, alongside return on capital employed, operating margin and the growth series. Sorting a universe on that field narrows a list; it does not rank risk, for the reasons set out above.

The limitation is worth stating rather than glossing. No screening field carries the maturity profile — when the borrowings fall due, whether the rate is fixed or floating, what the covenants say. That is in the notes to the annual report, and reading it is the step the screen exists to reduce the number of, not to replace.

Common questions

Continue reading

More in Company Analysis · App: Stocks · Definitions: glossary · Free tools: calculators · All: every article