EBITDA vs Free Cash Flow: What Each Measure Leaves Out

Earnings before interest, tax, depreciation and amortisation (EBITDA) stops before the working capital funded, the tax paid and the assets bought. Free cash flow stops after all three. For Sarvani Coatings Limited in year three, Rs 446 crore of EBITDA became Rs 304 crore of operating cash flow once Rs 54 crore went into working capital and Rs 88 crore into tax, then Rs 118 crore of free cash flow once Rs 186 crore of capital spending was paid.

Underneath that answer is one published statement and no estimates at all. Sarvani Coatings Limited files a cash flow statement for the year ended 31 March of year three, and every step between the two measures is already sitting in it: the movement in the cycle, the tax the company actually handed over, and the money that went out to buy assets. Which figure a bridge should start from is settled under the two EBITDA adjustment questions, covered separately, and the bridge here starts from the reported Rs 446 crore because the statements themselves carry that figure. No step in it needs a judgement from the reader, and a comparison that asks for none is unusual.

What does EBITDA actually measure, and where does it stop?

Think about a paint shop on a busy street. At the end of the year the owner counts what came in from customers and takes off what she paid for tins, thinner, wages, rent and electricity. The money left over is what the trading itself threw off. She has not yet paid for the extra stock sitting in the back room for the festive season, she has not yet paid the tax on the year, and she has not yet paid for the new mixing machine she ordered in March. EBITDA is that same number for a company: earnings before interest, tax, depreciation and amortisationThe accounting charge that spreads the cost of an asset over the years it is expected to be used. The calculation belongs to the accounting layer of this library. Here the charge is only a scale., which is to say the result of trading before three quite separate claims on the money arrive.

The three claims are not small and they are not optional. A growing business has to fund a growing working capitalStock plus what customers owe the business, less what the business owes its suppliers. The cycle that produces working capital is a subject in its own right, covered separately in this library. position, because stock has to be bought before it is sold and customers pay later than suppliers do. The business has to pay tax in cash, on dates that have nothing to do with when the profit was recognised. And it has to buy and replace the plant that makes the product at all. EBITDA is measured before every one of them.

Stopping early is not a defect in EBITDA, it is the entire reason the measure exists. A figure taken before tax can be compared across two companies with different tax positions. A figure taken before depreciation can be compared across two companies whose plants were built in different decades at different prices. A figure taken before interest describes the business rather than the way somebody chose to finance it. The defect appears at one precise moment, and only then: when somebody reads the number as if it were cash.

What is free cash flow, and whose definition is being used?

Here is where a lot of otherwise careful reading goes wrong, and it goes wrong on a word rather than on an arithmetic step. Free cash flow is not a defined line in a filed statement the way revenue is. Free cash flow is a figure somebody builds, and different people build it differently. Here, and everywhere in this library, free cash flow means operating cash flow less capital spending, and nothing else.

Other definitions are in wide use and none of them is wrong. Some readers deduct interest paid as well, on the argument that servicing debt is not discretionary. Some go further and deduct debt repaid. Some deduct only the portion of capital spending needed to keep the existing plant running, and treat the spending on new capacity as a choice rather than a cost. Each of those is a defensible answer to a slightly different question, and each produces a different number from the identical statements.

The effect on one year at Sarvani Coatings is immediate. On the definition used here, free cash flow is Rs 304 crore less Rs 186 crore, or Rs 118 crore. Deducting the Rs 21 crore of interest paid gives Rs 97 crore. Deducting the Rs 30 crore of borrowing repaid on top of that gives Rs 67 crore. Three figures, one year, one set of statements, nobody being careless: a spread of Rs 51 crore between the widest and the narrowest, on a company whose profit after tax was Rs 278 crore. A free cash flow figure quoted without its definition attached is not a usable figure, and asking which definition is being used is not pedantry but the first step in reading it at all.

What exactly sits between the two?

Three things, and their names are already familiar. The change in working capital, the tax actually paid in cash, and the money spent on assets. Here they are for Sarvani Coatings Limited in year three, in the order the cash flow statement takes them.

Two details in that table are worth slowing down for. The first is that the working capital figure is not an assumption: Sarvani Coatings closed year three with inventory of Rs 402 crore, receivables of Rs 289 crore and trade payablesWhat a business owes its own suppliers for goods already received. Payables sit on the liabilities side of the balance sheet and reduce the cash the cycle ties up. of Rs 356 crore, which is Rs 335 crore of working capital against Rs 281 crore a year earlier. The rise of Rs 54 crore is arithmetic on two balance sheets.

The second is the tax line. The profit ladder shows a tax charge of Rs 93 crore, an effective tax rateThe tax charge in the profit statement divided by profit before tax. The charge differs from the cash actually paid, for reasons that belong to the accounting layer. of 25.1 per cent on profit before tax of Rs 371 crore. The cash flow statement shows Rs 88 crore paid, or 23.7 per cent of the same profit before tax. The charge and the payment are two different quantities, and the Rs 5 crore between them is normal. Why they differ is a matter of accrual accountingRecording income and costs when they are earned or incurred rather than when cash moves. The mechanism is taught in this library's accounting layer and is only named here. and is settled elsewhere in this library. The bridge is about cash, so the bridge uses the cash figure.

The first two steps are already inside the published operating cash flow line, so the bridge asks nothing of the reader: it is arithmetic on figures the company itself filed. That is what makes this comparison unusually cheap to run. There is no model, no assumption, no adjustment and no judgement. Rs 54 crore and then Rs 88 crore come off the Rs 446 crore and give Rs 304 crore, exactly the operating cash flow the company published, so the top half ties. The Rs 186 crore of capital spending comes off that and Rs 118 crore remains. Ten minutes with a filed statement produces the whole thing.

Two ratios fall out of the same table without any further work. Operating cash flow of Rs 304 crore is 68.2 per cent of the EBITDA it started from, and free cash flow of Rs 118 crore is 26.5 per cent of it. Against revenue of Rs 2,415 crore for the year, the capital spending was 7.70 per cent and what remained after it was 4.89 per cent. The four figures are the whole comparison, and every one of them is arithmetic on lines the company filed.

Why can rising EBITDA produce so little cash?

Sarvani Coatings grew EBITDA 31.2 per cent, with Rs 340 crore in year two becoming Rs 446 crore in year three. Growth of that size is a large move by any standard. And 26.5 per cent of the larger figure reached free cash flow. Growth of 31.2 per cent and conversion of 26.5 per cent look like they cannot both describe one year, but they can, and the reason takes a little working through.

A growing business consumes cash in two directions at once. More sales need more stock on the shelf and more money owed by customers before any of it comes back, so a growing business funds a bigger cycle. A paint line ordered this year also makes nothing until it is commissioned, so the business buys assets ahead of the revenue those assets will eventually carry. Both of those are the arithmetic of growing, not symptoms of anything. A household does the same thing when a second child arrives: the outgoings jump the year before the benefit does, and nobody calls that a crisis of household finances.

Here is the uncomfortable part. A company converting badly because it is building capacity and a company converting badly because its cycle is deteriorating produce the same conversion ratio. The ratio cannot tell those two apart and the balance sheet can, so a low conversion reading is a reason to open the balance sheet rather than a reason to reach a conclusion. Where did the money go? Into net blockThe cost of a company's plant, buildings and equipment less the depreciation charged on it so far. Net block sits at the top of the asset side of the balance sheet. and capital work in progress, or into stock and receivables that are turning more slowly than they used to? The two destinations are different stories, and the statements answer the question directly.

Can maintenance spending be told from growth spending?

Mostly, no, and this is the honest limit of the whole comparison. Capital spending is one line. Inside it sit two quite different things: the money that keeps the existing plant capable of making what it already makes, and the money that buys capacity the company does not yet have. The first is a running cost wearing the clothes of an investment. The second is genuinely a choice. Published statements almost never split them, so a reader who claims to know how much of Sarvani Coatings' Rs 186 crore was maintenance has assumed the answer rather than found it.

The usual shortcut is to treat the depreciation charge as the maintenance figure, on the reasoning that depreciation is roughly what the asset base consumes each year. Depreciation is a weak proxy, and the reason it is weak is worth knowing exactly. Depreciation reflects what the assets cost when they were bought and the lives assigned to them in the accounts. Replacing a mixing line today costs what a mixing line costs today, not what the old one cost eleven years ago. The two figures answer different questions and only overlap by accident.

The statements do give a clue, and a good one. Investing activitiesOne of the three headings a cash flow statement is divided into, covering money spent on or received from long term assets. Which heading an item belongs under is settled in the accounting layer. record that Rs 186 crore went out. The balance sheet then shows Rs 118 crore of capital work in progress, described in the accounts as a coatings line not yet commissioned. The work in progress is 14.6 per cent of the Rs 806 crore net block, paid for and earning nothing yet. So it can be said, with evidence rather than with a guess, that a substantial part of this year's spending has bought something that has not started carrying revenue. The balance sheet supports a growth explanation rather than a deteriorating one.

One coincidence needs naming before it misleads anybody. Free cash flow for the year is Rs 118 crore and capital work in progress is also Rs 118 crore. The two figures are equal by coincidence and no relationship whatever connects them. Free cash flow is a flow over twelve months. Work in progress is a balance at one date. A flow and a balance can land on the same number in the same year without either one causing the other.

What does a conversion rate show, and over how long?

A conversion rate describes a period, and the period is shorter than people treat it as. Sarvani Coatings converted 26.5 per cent of EBITDA into free cash flow in year three. Year three's capital programme is the only thing in the calculation that varies much from one year to the next, so the 26.5 per cent belongs to year three and to that programme. Moving the control above produces a completely different conversion rate from an identical business.

Capital programmes arrive in blocks. A coatings line is commissioned once and then runs for a decade. A warehouse is built once. A boiler is replaced once in fifteen years. So the spending line is lumpy by nature, and a lumpy series read at one point shows only where in the lump the reading happened to fall. Averaging a lumpy series over a single observation is not averaging at all, so a conversion reading needs several years before it says anything about the business rather than about the year.

Think of a household with a car. In the year the car is bought, the household saves nothing and might look alarming on paper. In the four years afterwards it saves steadily. Nobody who looked only at the first year would have described that household accurately, and nobody who averaged across all five would have missed anything.

What does a low conversion rate not prove?

Nothing at all. A low conversion rate is not evidence of an accounting problem. It is not evidence of a quality problem. Nor is it evidence of a weak business, a stretched balance sheet or anything else on its own. A low rate is a question about where the money went, and the question has an answer sitting in the same set of statements.

A reader who has spent a week looking for warning signs starts seeing them, and earnings quality reading is a week of exactly that. Low conversion feels like a finding, and it is not one. A low reading is an instruction to look further, in a specific place, at three specific lines: capital work in progress, receivable days and inventory days. Sometimes what turns up raises a further question. Often what turns up is a business building a plant.

How does this fit the rest of the earnings quality reading?

Two comparisons are now in hand that sound similar and are not. One puts profit against operating cash flow. The other puts EBITDA against free cash flow. Sarvani Coatings answers them differently in the same year, the cleanest possible demonstration that they are different questions.

Cash from operations, Rs 304 crore, ran at 1.09 times the Rs 278 crore of profit after tax, so cash was Rs 26 crore ahead of profit. The 1.09 times is the accrual question, and it asks whether the profit the company recognised turned into money at roughly the expected rate. Meanwhile only 26.5 per cent of EBITDA reached free cash flow. The 26.5 per cent is the capital question, and it asks what the business had left after paying for the assets it bought. Both statements are true at once, they do not contradict each other, and a reader who runs one and reports it as the other has answered a question nobody asked.

Who actually runs this comparison, and what do they do with it?

Three people, for three different reasons, and none of them stops at the ratio.

A lender wants to know what is left to service debt after the business has paid for itself. For Sarvani Coatings that is the Rs 118 crore, and the lender then notices the company paid a dividend of Rs 96 crore out of it, leaving Rs 22 crore before interest of Rs 21 crore and repayment of Rs 30 crore. Free cash flow covered the dividend 1.23 times and no more. The cover of 1.23 times is not a warning. The company is also sitting on Rs 312 crore of cash and investments while carrying Rs 240 crore of borrowings, so its net debtWhat a business owes in borrowings, reduced by the cash and liquid investments it already holds. A negative figure means more cash than debt. is minus Rs 72 crore. The cover ratio is a fact that shapes the next conversation.

An analyst uses the comparison to check her own valuation work. If she has built a multiple on enterprise valueThe value of the whole business to all its funders, equity and debt together, before deciding how that value is split between them. The method is taught in this library's valuation layer. and EBITDA, she is using a measure taken before Rs 328 crore of uses of money, which is fine as long as she knows it and does not then describe the same figure as cash. If she is estimating what the business can return to shareholders, EBITDA is the wrong starting point entirely.

An investor in the household sense, someone holding the share and reading the annual report on a Sunday, uses it to ask one question: is the gap between the two numbers being spent on something, and can I see the something? At Sarvani Coatings the answer is visible on one line of the balance sheet, and that is a better Sunday afternoon than any ratio.

Sarvani Coatings Limited and Meghna Iyer are invented.
Educational material. Not advice on any investment, tax, budget or market position.