What the number is
Free cash flow is the cash a business produced from operating, less the cash it spent on assets — what is genuinely left over for lenders and owners once the business has been kept running.
Two inputs, and both come off the same page. Cash from operations sits at the top of the cash flow statement. Capital expenditure — spending on plant, equipment, software, buildings — sits below it, under investing.
The definition looks settled and is not. Everyone agrees on the first input; the argument is about which capital spending to subtract. Take all of it and you charge a company for the factory it is building to serve customers it does not yet have. Take only the spending needed to stand still and you are using a number the company does not publish. That tension runs through the rest of this article, and it has no clean resolution.
What free cash flow is not is a verdict. It is one measurement of one period, and a period has two ends.
From profit to cash, line by line
Profit and cash differ for reasons that are not mysterious once you see where they live. Revenue is recorded when the sale is made, when the obligation to pay arises. The money arrives when the customer actually pays. The gap between those two dates sits on the balance sheet as a receivable, and the cash flow statement's job is to take it back out of profit.
The same logic runs the other way for anything the company has received and not paid for. The general rule, and it is worth holding onto because every line in the working capital section follows it: an asset going up uses cash, a liability going up supplies it.
Here is the bridge for an illustrative company — figures invented to make the arithmetic visible, not drawn from any real business, and simplified in that a filed statement starts from profit before tax and shows tax paid as its own line.
| Line | ₹ crore | Why it moves |
|---|---|---|
| Profit after tax | 60 | The reported figure, arrived at after accruals and estimates |
| Add: depreciation and amortisation | +35 | Charged against profit; no cash left the company this year |
| Less: increase in receivables | −45 | Sales counted in profit that customers have not yet paid |
| Less: increase in inventory | −20 | Cash converted into goods still sitting in a warehouse |
| Add: increase in payables | +15 | Costs counted in profit that suppliers are still financing |
| Cash from operations | 45 | What operating actually produced in cash |
| Less: capital expenditure | −70 | Cash spent on assets, whether to maintain or to expand |
| Free cash flow | −25 | What was left for lenders and owners |
Reported profit of ₹60 crore, free cash flow of minus ₹25 crore, a gap of ₹85 crore. Not a single line in that bridge is an accusation. Every one of them is ordinary, and between them they explain the whole difference — which is the useful thing about reading the statement in this order rather than looking at the two end numbers and wondering.
The depreciation line is the one people misread most often. Adding it back does not mean the wear on the assets was imaginary; it means the cash for those assets left in an earlier year and is being charged against profit in instalments. The cash version of that wear is the capex line further down, and the two rarely match.
Why it is harder to dress up
Because most of the room in accounting is room about timing, and cash settles the timing for you. Accrual accounting exists to answer a real question: how much did this period earn, given that money and activity do not arrive on the same day. Answering it requires estimates — how long an asset lasts, how much of a receivable will be collected, what a warranty will cost. Estimates are where the room is.
Cash has less of it, for a structural reason. The estimates decide which year a number lands in, and years end. A receivable is eventually collected or written off; inventory is eventually sold or written down. Over a long enough span the accruals reverse and the two measures converge, which makes most accounting choices timing choices rather than size choices.
The sharpest illustration is the switch that flatters profit most reliably. Suppose an illustrative company spends ₹10 crore on something it could either expense this year or record as an asset with a five-year life — the figures are invented to make the arithmetic visible. Same cash, same day, same bank account.
- Expensed. Profit falls ₹10 crore. The payment is an operating outflow, so cash from operations falls ₹10 crore too.
- Capitalised. Profit falls only by the first year's depreciation, ₹2 crore. The payment is now reported under investing, so cash from operations does not fall at all — and capital expenditure rises ₹10 crore.
Compare the two columns. Capitalising lifts reported profit by ₹8 crore and lifts reported cash from operations by ₹10 crore. It lifts free cash flow by nothing, because the ₹10 crore it added at the top is the ₹10 crore it added to capex, and the subtraction cancels exactly inside free cash flow.
That is the specific mistake this article exists to name. A reader who compares profit against cash from operations, finds the cash comfortably larger and concludes the earnings are real has checked the one number the capitalisation switch inflates. Cash from operations is a half-measure; the subtraction is the part that does the work.
One honest qualification. The cancellation is exact only if the cash tax bill is the same under both treatments. Where tax depreciation is computed on its own schedule rather than following the books, it is; where it follows the accounting treatment, the offset is close rather than perfect.
What still moves it
Harder to dress up is not impossible to dress up. Four routes do it, all of them ordinary enough that none is evidence of anything on its own.
Timing at the year-end is the simplest. Paying a large supplier on 2 April rather than 28 March moves cash out of one year and into the next without a single thing changing in the business. Collecting hard from customers in the last fortnight does the same in reverse. Neither is available twice: a payment deferred at one year-end lands in the following one, which is why a single year's figure is much softer than three consecutive years.
The second route is the one that reads as a cash success. A company can sell its receivables to a financier — factoring, bill discounting, whatever it is called locally — and receive most of the money now instead of in ninety days. Receivables fall, cash from operations rises, and the cash arrived from a lender rather than from a customer. The obligation is real; it is simply somewhere else in the accounts.
The mirror image runs through payables. Under a supplier finance arrangement a bank pays the company's suppliers early and the company repays the bank later, so payables stay high and operating cash flow stays flattered while what is in substance borrowing sits under a heading that does not say so. Both of these are legitimate treasury practice used by a great many ordinary businesses. Both also make one year's cash look better than the year's trading did.
Third, classification. Where interest paid, dividends received and lease payments sit inside the cash flow statement is not always fixed — frameworks differ, and where a framework allows a choice, two companies with identical economics can report different operating cash flow. When a standard changes the split of rent between a financing repayment and interest, reported operating cash flow moves for every company that rents anything, with no change to a single rental agreement. Read the policy note before comparing two companies on this line.
Fourth, netting. If capital expenditure is reported net of what was received for selling old assets, a year in which a company sold a building will show artificially light capex — and free cash flow will absorb the difference. Gross capex and disposal proceeds tell a different story from a single net figure, and the notes generally carry both.
| The line | What can move it without the business changing | Where to look |
|---|---|---|
| Depreciation add-back | Asset lives and residual values are estimates; a longer assumed life raises profit and changes no cash at all | The accounting policy note, and any change of estimate disclosed in it |
| Receivables | Selling or discounting receivables converts them into cash today | Notes on factoring or bill discounting; short-term borrowings |
| Payables | A bank paying suppliers early under a supplier finance arrangement keeps payables high | Notes on supplier or vendor finance; other financial liabilities |
| Cash from operations | Spending recorded as an asset rather than a cost leaves this line untouched and raises profit | Intangible assets and capital work in progress; capitalised interest |
| Capital expenditure | Reporting it net of asset disposals makes a year look light | The fixed assets note — gross additions and disposal proceeds separately |
| Both, at the boundary | Paying or collecting a few days either side of the year-end | Three to five consecutive years, where the shifts have to reverse |
The capex line is doing two jobs
Everything above concerned the first input. The second has a problem of its own, and it is the reason two competent analysts can compute different free cash flow from the same filing and both be defensible.
Capital expenditure is reported as one figure, and it contains two different kinds of spending. Replacing a machine that wore out keeps the business where it is. Building a second factory makes it larger. Charging both against this year's cash is right if you want to know what was actually left over; it is wrong if you want to know what the business could produce were it not growing. One line doing two different jobs, and no company publishes the split.
The usual workaround is to treat depreciation as the maintenance figure, on the reasoning that it is the accountant's estimate of assets being used up. It is a reasonable first approximation and it has a known bias: depreciation is charged on what an asset cost, and replacing it costs what the replacement costs today. An asset bought for ₹100 crore with a ten-year life accumulates ₹100 crore of depreciation over that decade — but if the price of the same machine rose at an illustrative 5% a year, the replacement costs about ₹163 crore, so the depreciation charge covers under two-thirds of it. That is inflation doing arithmetic, not a company doing anything, and it means depreciation understates maintenance spending in most long-lived asset bases.
There is also a kind of expansion the capex line never sees. Buying a company is an investing outflow exactly as buying a machine is, but acquisitions are reported on their own line and are not capital expenditure — so a business that buys its growth rather than building it can show light capex and comfortable free cash flow while spending heavily to expand. The cash left the company either way. Which line it left from decides whether this measure counts it.
The consequence cuts both ways, which is what makes the line worth reading rather than scoring. A company that spends far above depreciation for several years may be building something. A company that spends well below depreciation for several years may be deferring maintenance, and deferred maintenance shows up as free cash flow now and as a large capex year later. Positive free cash flow produced by not spending is a different fact from positive free cash flow produced by trading, and the two are indistinguishable if you read only the bottom line.
Growth consumes cash, mechanically
Before treating a negative number as a warning, it is worth seeing how routinely growth produces one on its own — with no aggressive accounting anywhere and every customer paying on time.
Take an illustrative business whose net working capital — receivables plus inventory, less payables — runs at a steady 25% of revenue, and whose net profit margin is 8%. Both inputs are chosen here to make the arithmetic legible; they are not typical of anything, and no sector has a right answer for either.
- Revenue ₹400 crore growing 25%, to ₹500 crore. Profit is 8% of ₹500 crore, so ₹40 crore. Working capital has to rise by 25% of the extra ₹100 crore of revenue, so ₹25 crore goes into the balance sheet. Net: ₹15 crore of cash.
- The same business growing 50%, to ₹600 crore. Profit is ₹48 crore. Working capital absorbs 25% of the extra ₹200 crore, so ₹50 crore. Net: minus ₹2 crore.
Nothing changed except the growth rate, and the cash generation flipped sign. The tipping point is calculable from the two inputs: it is the margin divided by the difference between the working-capital share and the margin, which for 8% and 25% comes to about 47%. Above that growth rate this illustrative business consumes cash while remaining entirely profitable, because working capital grew faster than the profit funding it.
Two things about that 47%. It is a property of the two numbers chosen and moves the moment either does — it is not a threshold, not a screen, and not a fact about any real company. And it deliberately leaves out depreciation and capital spending so the working-capital effect is visible on its own; put both back and the figures change while the mechanism does not.
What this explains is why fast-growing profitable companies borrow. The cash is inside the receivables and the inventory, and it stays there for as long as the business keeps growing — which is also why the gap has to be funded by someone, at whatever borrowing costs at the time.
Three businesses, one minus sign
So a negative figure needs a cause before it means anything. There are three common ones, they look identical from the outside, and the statement itself separates them.
- Building capacity. Cash from operations is positive and roughly tracks profit; capital expenditure runs far above depreciation and the deficit is the difference. The questions that follow are what the spending is buying, whether it is funded by owners or lenders, and whether earlier rounds of it ever showed up as more cash.
- Growth absorbing working capital. Cash from operations sits far below profit, or turns negative while profit stays positive, and the working capital lines explain the gap. The question that follows is whether working capital is a stable share of revenue or a rising one — a stable share is the arithmetic of growth, a rising share means the terms of trade themselves moved, which is a different question with different answers.
- Not generating cash. Cash from operations is small or negative with no revenue growth and no capital programme to point at, and the deficit is covered by borrowing year after year. The question that follows is what has been funding it, and for how many years running.
Same sign, three different situations, and the cash flow statement distinguishes them in about a minute. That is the argument for reading the whole statement rather than lifting the final figure out of it: on its own the number carries the same minus sign for three different businesses.
The reverse case deserves the same suspicion. A positive figure can come from trading well, from cutting capital spending, from collecting harder than usual for a few weeks, or from selling receivables to a bank. All four produce cash. Only one of them is repeatable without consequence.
Reading it over years, not in one
Every distortion above shares a property: it borrows from an adjacent period. A deferred payment lands next year. A discounted receivable is one the company will not collect later. A skipped maintenance year makes the following capex year larger.
Which is why the number behaves quite differently across a run of years than in a single one. Over three to five years the timing shifts have to reverse, the capital cycle has usually been through both a heavy and a light phase, and the cumulative gap between reported profit and cumulative free cash flow becomes a question with an answer rather than a curiosity. If the two have diverged persistently in one direction, something structural is producing it — working capital, capital intensity, or accruals that are not reversing — and each of those is visible somewhere in the notes.
This is also the trade-off, and it should be stated rather than glossed. Free cash flow is harder to manipulate precisely because it ignores the accruals that exist to smooth timing — so it swings from year to year for entirely innocent reasons, in a way profit is designed not to. What you get is honesty bought with volatility. A measure that refuses to smooth will not be smooth.
One last framing that helps. Cash arriving later is worth less than the same cash arriving now, which is the whole content of the time value of money — so a business whose profit converts into cash after a long delay is not merely slower, it is worth a different amount. That is a mechanism, not a valuation method, and it is as far as this article goes.
Where the numbers come from
The cash flow statement is in the company's own annual report and in the filings it makes to the exchanges, alongside the balance sheet and the profit and loss account. Read the consolidated statement rather than the standalone one where a group has subsidiaries; the standalone version describes the parent company alone, which is rarely the business you had in mind.
Two practical cautions on second-hand figures. Data providers and screeners restate, and they do not agree on what free cash flow means — some subtract gross capex, some net, some deduct lease payments, some do not. Comparing a figure from one source against a figure from another compares two definitions as much as two companies, so check the formula before the comparison rather than after it.
The second is that a screen is a starting list and not a reading. FNOTrader's Market Pulse scanner works on price, volume and market breadth across NSE, and the discipline that applies to any filter — knowing exactly what each field is computed from before ranking anything on it — is set out in our screening guide. Nothing on a screen replaces the statement it was derived from.
Common questions
What is free cash flow?
The cash a business produced from operating, less the cash it spent on assets — what is left for lenders and owners once the business has been kept running. Cash from operations sits at the top of the cash flow statement and capital expenditure below it, under investing.
How is free cash flow different from profit?
Profit counts a sale when it is made; cash counts it when the customer pays. The difference between the two is the change in the balance sheet — receivables, inventory and payables — plus the gap between depreciation charged this year and cash actually spent on assets this year. The cash flow statement shows every one of those adjustments explicitly.
Why is free cash flow harder to manipulate than profit?
Because most accounting choices decide which year a number lands in rather than how large it is, and cash eventually settles. The clearest case is spending recorded as an asset instead of a cost: it raises reported profit and raises reported cash from operations, and it changes free cash flow by nothing, because whatever it adds at the top it also adds to capital expenditure.
Is operating cash flow the same as free cash flow?
No, and the difference matters. Operating cash flow is the figure that rises when spending is capitalised rather than expensed. Free cash flow subtracts capital expenditure, which is precisely the subtraction that cancels the effect. Comparing profit against operating cash flow alone checks the one number that switch inflates.
Is negative free cash flow a bad sign?
It is not a sign at all until you know what caused it. Three different situations produce the same minus sign: a company building capacity, a growing company whose receivables and inventory are absorbing cash, and a company that simply does not generate cash. The statement separates them — look at whether capital expenditure is far above depreciation, and whether working capital is a stable or a rising share of revenue.
Can a profitable company run out of cash?
Yes, and it does not require anything going wrong. If working capital is a fixed share of revenue, growth increases it in proportion, so above a certain growth rate the increase exceeds the profit funding it. With an illustrative 8% net margin and working capital at 25% of revenue, that crossover is around 47% growth — a figure that moves the moment either input does, and which is arithmetic rather than a threshold to screen on.
Should free cash flow subtract all capital expenditure or only maintenance capex?
Both conventions are defensible and they answer different questions. Subtracting all of it tells you what was actually left over this year. Subtracting only maintenance tells you what the business could produce if it stopped expanding — but no company publishes that split, so the figure has to be estimated, usually with depreciation, which understates replacement cost when prices rise.
How many years of free cash flow should be looked at?
More than one. Every way of moving the figure borrows from an adjacent period: a payment deferred past the year-end lands in the next year, a discounted receivable is one that will not be collected later, a skipped maintenance year makes the following capex year larger. Over three to five years those shifts reverse and a persistent gap between profit and cash becomes a question with an answer.
Continue reading
More in Company Analysis · App: Market Pulse Stocks Scanner · Definitions: glossary · Free tools: calculators · All: every article