The Cash Flow Statement: The One Statement That Is Harder to Dress Up

Why cash is harder to dress up, and exactly where the hardness runs out

Reported profit is assembled from judgements. When a sale has been earned, how long a machine will last, whether a debt is doubtful, whether a cost belongs to this year or the next: each of those is a decision inside a permitted range, and each of them moves the bottom line. None of that is dishonest. It is what accrual accounting is for, because a business that recognised revenue only when money arrived would report nonsense about its own performance.

Cash is different in three specific ways, and it is worth being precise about them rather than waving at the slogan.

The closing balance is confirmed by somebody with no stake in the answer. A receivable is worth what management believes it is worth. A bank balance is what a bank says it is. That single asymmetry is the whole basis of the claim, and it is narrower than it sounds, because it constrains the total rather than any particular line inside the statement.

The statement is derived, so it has to tie. The cash flow statement is not an independent report. It is a reconciliation between the profit and loss account and two consecutive balance sheets, which means every rupee added to operating cash has to come out of a balance-sheet account somewhere, and balance-sheet accounts persist into the following year. Profit can be flattered by a judgement that leaves no trace. Cash cannot be flattered without leaving a footprint on a balance you will have to publish again in twelve months.

Timing shifts reverse, and reverse visibly. An estimate can be held at the same optimistic level indefinitely. A payment postponed from March is a payment made in April, and it takes a larger postponement each year to keep the trick running. The reversal is arithmetic, not a matter of anyone changing their mind.

What survives all three is the interesting part, and it is where the second half of this page goes. Three things remain available to a company that wants a better looking operating cash number without saying anything untrue. It can choose which section of the statement a genuine cash movement is reported in. It can change when a genuine payment happens. And it can make an accounting judgement that lifts profit and operating cash together, which is the dangerous case, because the cash flow statement is normally treated as the check on profit and in that case it is not checking anything.

The structure below assumes you already know what the three statements are and how they connect. If that map is unfamiliar, our guide to reading an Indian annual report like an analyst lays out the whole report, including how net profit reaches equity as retained earnings and why the cash line ties back to the balance sheet. This page takes that as read and goes into one statement in detail.

The entity, and why it was constructed rather than typed in

Everything on this page belongs to constructed entities. No real company appears anywhere, as an example or a reference. That is partly caution and partly method: a constructed entity lets the drivers be stated openly, which means the arithmetic can be reproduced or disputed, and it lets the answer be known in advance, which is the only way to demonstrate what a diagnostic actually detects.

The construction order matters and is worth stating, because it is the reason the statement on this page ties rather than merely appearing to. A set of drivers was written first: a revenue path, a margin path, working-capital days, a capital-spending schedule, a borrowing schedule and a dividend. From those, the profit and loss account was computed. From the profit and loss account and the days, the balance sheet was computed. Only then was the cash flow statement derived, entirely from the movements between two consecutive balance sheets, with the closing cash balance left as the number the statement produces rather than a number chosen in advance.

That ordering turns the accounting identity into a real test. The balance sheet balances only if the profit and loss account, the retained-earnings roll and the cash flow statement are all mutually consistent, so the code asserts four things on every year and refuses to produce a page if any of them fails. Assets must equal liabilities plus equity. The net movement in cash must equal the three sections added together. The closing cash balance must equal the opening balance plus that movement. And cash generated from operations, computed by the indirect method, must equal the same figure computed independently by the direct method, from receipts collected from customers less payments made to suppliers and everyone else. That last check is the useful one, because the two expressions are algebraically equivalent but share no code, so it catches the sign errors and transposed lines that an eye reading a column will not.

One presentation decision is worth flagging. Every balance is stated to the nearest tenth of a crore, and the statement is derived from those stated balances rather than from unrounded figures. Real statements are rounded before they are printed, and a reader adding a column expects the column to add. Doing the rounding at construction rather than at printing means every column on this page adds exactly.

The entity, called Entity One throughout, is an ordinary capital-heavy manufacturer of no particular sector. Revenue grows from 940.0 crore rupees to 1,260.0 crore over three years, the margin improves slightly, working-capital days drift up by a handful, and the company spends heavily on capacity. Interest is charged at an illustrative 9 percent on average borrowings and tax at an illustrative 25 percent, a round number chosen for legibility that is not a statement of any statutory rate. All figures are illustrative and simulated.

Three sections, and what the pattern of signs is telling you

The split into operating, investing and financing is not a filing convention. It is the statement answering three different questions that most readers collapse into one, and the answers are only interesting next to each other.

Section A, operating. Cash thrown off by running the business: collecting from customers, paying suppliers and staff, paying tax. This is the section that gets quoted, and quoting it alone is how the most common misreading happens.

Section B, investing. What the company spent on capacity and long-term assets, and what it got back from selling any. For most industrial companies this is dominated by one line, the purchase of property, plant and equipment.

Section C, financing. Where the difference came from. New borrowing, repayment, share issues, dividends and, under the Indian standard, interest paid. This section names the party who funded whatever the first two sections could not.

Eight sign combinations are possible and four of them describe most companies. Positive operating cash, negative investing and negative financing is a mature business paying down debt and paying its owners. Positive, negative and positive is a growth business borrowing to expand. Negative operating cash with positive financing is a business being kept alive by its funders, which may be perfectly reasonable early on and is a serious question later. Positive operating cash with positive investing is often a company selling assets, which is worth understanding before anything else in the report.

Entity One is the second pattern, and the numbers show why the pattern is more informative than any single line inside it.

Read only the operating section and Entity One is a company whose cash generation improved by 35 percent in two years. Read all three and it is a company whose expansion costs more than it earns, financed by borrowings that rose from 380.0 crore to 665.0 crore while the finance cost rose from 36.0 crore to 54.0 crore. Both descriptions are accurate. Only the second one tells you what happens if the lenders stop.

There is a habit worth forming here, and it costs nothing. Before reading any figure in the statement, read the three subtotals as a triple and say out loud what they describe. It takes ten seconds and it prevents the single most common error in retail fundamental analysis, which is treating a healthy operating line as though it settled the question.

The reconciliation, line by line, with the account that produced each one

Under the indirect method the operating section does not start with cash at all. It starts with profit and then removes everything about profit that was not cash, which is a strange way to present a cash figure and an extremely informative one, because every adjustment names a specific gap between the two. The direct method, which lists receipts and payments, gives the same answer and teaches less, since it never shows you why the two numbers differ.

Here is year three, walked.

Depreciation, add back 100.0 crore. Depreciation is a real cost and it took no cash out this year, because the cash left when the asset was bought, in some earlier year, and appeared in the investing section then. Adding it back is not a claim that the cost is fictional. It is a correction for the fact that this statement is about this year's cash. Notice how large the add-back is relative to profit before tax of 81.0 crore: on a capital-heavy balance sheet the single biggest driver of operating cash flow is often not the trading performance at all.

Finance cost, add back 54.0 crore. The interest was paid in cash, so this is not a non-cash adjustment. It is a classification adjustment. Under the Indian standard the interest paid by an ordinary company belongs in the financing section, so it has to be taken out of the operating figure here and shown again lower down. It appears twice in the statement and is deducted once. This is the line that most often confuses a reader who has learned the statement from a source written for another jurisdiction, and it is covered again further down because it has a consequence.

Interest income, strip out 3.2 crore. The mirror image. Interest received on investments is real cash and it belongs in the investing section, so it comes out of the operating subtotal.

Those three, plus the gain on the plant disposal in year two, produce the subtotal called operating profit before working-capital changes, which for year three is 231.8 crore. It is worth pausing on that number because it is the closest thing in the statement to the cash the trading operation would have generated if the balance sheet had stood still. For Entity One it is identical to the year's EBITDA, and that is not a coincidence: the adjustments above are exactly the items that sit between EBITDA and profit before tax.

The second half of the reconciliation is where the balance sheet arrives.

Trade receivables rose by 40.0 crore, so deduct 40.0 crore. That much revenue was recognised in the profit and loss account and has not been collected. The customer has the goods and the company has an entry in an asset account.

Inventories rose by 20.9 crore, so deduct 20.9 crore. Cash was spent on materials that have not yet been sold. It is on the shelf, not in the bank.

Other current assets rose by 2.0 crore, so deduct 2.0 crore. Prepayments and advances: money already gone for something not yet consumed.

Trade payables rose by 16.7 crore, so add 16.7 crore. The company is holding suppliers' money. It has taken delivery and not yet paid, and until it does, the supplier is providing working capital free of charge.

Other current liabilities rose by 5.0 crore, so add 5.0 crore. Accrued costs recognised as expenses and not yet paid out.

The five movements net to 41.2 crore of cash absorbed, and the subtotal becomes cash generated from operations of 190.6 crore. There is one more line.

Income taxes paid, deduct 17.7 crore. Not the tax charge in the profit and loss account, which was 20.2 crore. The charge includes a deferred component of 4.0 crore that involves no cash this year, and the cash actually handed over also depends on the movement in the current tax liability. Confusing the charge with the payment is a common slip, and on Entity One in year three it is a 2.5 crore slip.

What is left is net cash from operating activities of 172.9 crore. Below is the whole thing, all three years, with the underlying accounts above it so the derivation can be followed rather than believed.

The last three rows are the reason the statement is worth trusting more than a single number pulled from a database. The closing cash figure in that final row is also an entry on the balance sheet above, arrived at from a completely different direction. If a company's cash flow statement did not reconcile, the auditor would have found it long before you did, which is precisely why the statement is a harder surface to work on than the profit and loss account.

Two companies, one reported profit

This is the demonstration the page exists for, and it needs no drift, no deterioration and no suggestion of wrongdoing. Two constructed entities, Twin One and Twin Two, trade for one year. They have the same revenue of 1,000.0 crore rupees, the same gross margin, the same EBITDA of 200.0 crore, the same depreciation of 60.0 crore, the same finance cost of 24.0 crore and the same tax. They therefore report the same profit before tax of 116.0 crore and the same profit after tax of 87.0 crore. Not similar. Identical, to the rupee.

They differ in exactly three numbers. Twin One holds its working-capital cycle where it started the year: 58 days of receivables, 62 days of inventory, 52 days of payables. Twin Two lets the cycle stretch to 79, 84 and 45 days, so it collects later, holds more stock and pays its suppliers sooner.

Twin One absorbs 21.0 crore into working capital over the year and finishes with net cash from operations of 150.0 crore. Twin Two absorbs 126.2 crore and finishes with 44.8 crore. The gap is 105.2 crore, and it comes apart cleanly: 57.5 crore of it is the extra 21 days of receivables, 36.2 crore is the extra 22 days of inventory, and 11.5 crore is the 7 days of supplier credit Twin Two gave up. Those three add to 105.2 crore exactly, because there is nothing else to account for.

The consequence lands one line further down. Both companies spend 70.0 crore on capital equipment. Twin One pays for that out of its own operating cash and has 80.0 crore left over, which it can use to reduce debt, pay a dividend or build the next plant. Twin Two is 25.2 crore short and has to raise the difference from somewhere: a loan, an overdraft, or a delay of the capital spending it just committed to. Same profit. Entirely different year.

Now the part that is usually left out, because it complicates a clean story. Nothing here establishes that Twin Two has done anything wrong. Extending credit is how a company enters a market where the incumbents already offer terms. Building stock is how a company prepares for a launch or protects itself from a supply disruption. Paying suppliers faster is sometimes bought with a discount that is worth more than the financing cost. Each of those is a real, defensible decision, and each produces exactly the pattern above.

What the statement does establish is that a question exists, and that the question cannot be answered from the profit and loss account, because on that document the two companies are the same company. It also establishes that the question is expensive: 105.2 crore rupees is 121 percent of the reported profit, and if it recurs, it is the difference between a business that funds its own growth and one that must keep going back to a lender.

Two further points follow from this pair. The first is a warning about valuation. Any multiple built on earnings values these two identically, because the earnings are the same number. Our explainer on what a price to earnings ratio actually measures sets out that arithmetic; the observation here is narrower, which is that the denominator of such a ratio is silent on whether the earnings arrived in cash. The second is that a single year cannot distinguish an investment from a deterioration. Whether Twin Two is entering a market or losing control of its collections is a question about the next three years, and it is the question our companion page on accrual and cash red flags in Indian filings is built to answer, with a multi-year screen and an honest account of how often such screens are wrong. This page stops at the mechanism.

The three ratios, their formulas, and what each one refuses to tell you

Three ratios are computed from the cash flow statement often enough to be worth stating precisely, and each has a specific blind spot that is not a subtlety but the main thing about it.

The intuitive one, and the most misread. On Entity One in year three it is 2.84, which looks superb and means very little. Operating cash sits above depreciation and, under the Indian standard, above interest paid, while profit after tax sits below both. Entity One charged 100.0 crore of depreciation against profit after tax of 60.8 crore and paid 54.0 crore of interest, so a ratio near three is a statement about how capital-heavy and how levered the company is, not about how good its earnings are. The same capital intensity that produces the flattering ratio is what makes its free cash flow negative. The ratio is informative against a company's own history and close to meaningless as a league table across companies.

The most useful of the three and the one most often skipped, because it requires reading the investing section. It asks what is left after keeping the asset base intact, and it is the number Entity One does badly on: 7.7 crore in year one, then minus 67.0 and minus 72.1. Its blind spot is that it makes no distinction between capital spending that maintains capacity and capital spending that adds it, and those are completely different economically. A crude but honest proxy is to treat depreciation as the maintenance portion, which for Entity One in year three means 100.0 crore of the 245.0 crore. On that basis the same company generated 72.9 crore rather than minus 72.1 crore. Both numbers are defensible and they are 145.0 crore apart, which is exactly why the phrase free cash flow should never be used without saying which version is meant.

The best of the three for comparing companies, because numerator and denominator both sit above interest and depreciation, so capital structure stops distorting the answer. Entity One runs at 0.78, 0.74 and 0.75. Twin One runs at 0.75 and Twin Two at 0.22, which is the divergence stated as a ratio. It is also cleaner in India than in some other jurisdictions for a reason given in the next section but one: interest paid is already outside the numerator by rule rather than by choice.

Its blind spot is tax. The numerator is after cash taxes paid and the denominator is before tax entirely, so a company that pays less cash tax scores better without operating any better. Hold Twin One's business completely constant and change only its cash tax bill from 29.0 crore to 10.0 crore, and cash conversion moves from 0.75 to 0.84, a gain of 9 percentage points bought with no operational improvement at all. If that matters for a comparison you are making, use cash generated from operations, the subtotal above the tax line, over EBITDA. For Twin One and Twin Two that reads 0.90 against 0.37.

A word on where these belong in a wider read. They are three inputs among many, and our page on building a fundamental scorecard shows how ratios of this kind get combined with balance-sheet and governance measures into a single view. The narrower claim here is that a cash ratio quoted without its formula and without its blind spot is decoration.

The limits, published: three ways the headline moves without a lie

A page that argued cash cannot be manipulated would be easier to write and would be wrong. Operating cash flow is harder to shape than profit and it is not beyond shaping, and the honest thing to do is to compute the size of the effect rather than to gesture at it. Take Twin Two exactly as it stands, with net cash from operating activities of 44.8 crore rupees, and change nothing about what the business actually did.

One. Pay the suppliers later. Twin Two ends the year on 45 days of payables. Suppose it takes 30 more days, ending on 75. Trade payables rise by 49.3 crore, and because an increase in payables is an operating inflow, net cash from operating activities rises to 94.1 crore, which is 110 percent above the base. Not one rupee of trading has changed. The company simply held onto money it owed across the year end, which is a decision that can be taken in the last fortnight of March.

It is a loan from next year and the repayment schedule is arithmetic. Assume the following year's revenue grows ten percent to 1,100.0 crore and payable days return to 45. Trade payables fall by 41.9 crore, and that is a direct deduction from the next year's operating cash. The mechanism only continues if the stretch is extended again, and it can be extended only so far before suppliers change their terms or stop supplying, which is why a payables line that grows faster than the cost of goods for several consecutive years is the thing to look at rather than any one year.

Two. Record an operating cost as an asset. Suppose 38.0 crore of internally generated development work, which could defensibly have been charged to the profit and loss account, is instead capitalised and amortised over 5 years. The accounting standard on intangible assets permits capitalisation only where defined criteria are met, and whether they are met in a particular case is a judgement rather than an arithmetic test. We did not read the text of that standard from an official source while preparing this page, so it is described here rather than quoted, and the criteria should be checked at source before being relied on.

The consequences are worth following carefully because they are counterintuitive. EBITDA rises from 200.0 to 238.0 crore. Amortisation of 7.6 crore appears, so profit before tax rises less than EBITDA does, and profit after tax rises from 87.0 to 109.8 crore, a gain of 22.8 crore. Net cash from operating activities rises from 44.8 to 75.2 crore, a gain of 30.4 crore, or 68 percent. Both headline numbers improved, which is the point: the cash flow statement is usually the check on profit, and here it moved the same way profit did, so it checked nothing.

Now the third number. The 38.0 crore did not vanish; it moved into the investing section, so capital spending rises from 70.0 to 108.0 crore. Free cash flow therefore goes from minus 25.2 to minus 32.8 crore. The company is 7.6 crore worse off in cash, and that figure is exactly the extra tax payable on the higher reported profit, because the 38.0 crore itself cancels between the two sections. Reported profit up 22.8, operating cash up 30.4, actual cash down 7.6. Free cash flow was the only one of the three that told the truth, and that is the strongest practical argument for computing it.

Three. The choice the standard already made for you. This one is not a manoeuvre and no company decides it. Under the Indian standard, interest paid by an entity other than a financial institution is classified in financing rather than operating, so Twin Two's 24.0 crore of interest never touches the operating figure. In jurisdictions where the equivalent standard leaves the classification open, the same company could show that interest inside operating, in which case its net cash from operating activities would read 20.8 crore rather than 44.8 crore, 54 percent lower. Nothing about the business differs. It is a rule, and it is stated in the next section with the source.

The practical consequence is that an Indian company's operating cash flow figure is structurally higher than an otherwise identical company reporting under a regime that allows the operating choice, and the gap widens with leverage. Comparing that headline across regimes without adjusting for it is a straightforward error, and it is invisible unless you know the rule exists.

What the Indian rules actually say, and what could not be verified

Four things about the Indian treatment are specific enough to be worth citing rather than paraphrasing, and one widely repeated claim turned out not to be verifiable, which is reported here alongside the rest.

The standard. Cash flow statements in India are prepared under Indian Accounting Standard 7, Statement of Cash Flows, notified under the Companies (Indian Accounting Standards) Rules, 2015. Paragraph 1 requires an entity to prepare a statement of cash flows and to present it as an integral part of its financial statements for every period presented. Paragraph 10 requires those cash flows to be classified by operating, investing and financing activities, which is where the three sections on this page come from.

Either method is permitted. Paragraph 18 allows an entity to report operating cash flows using the direct method, disclosing major classes of gross receipts and payments, or the indirect method, adjusting profit or loss for non-cash items, for deferrals and accruals, and for items of income or expense associated with investing or financing cash flows. The indirect method is what is presented on this page because it is what Indian readers encounter.

The classification carve-out, which is genuinely Indian. Paragraph 31 requires interest and dividends received and paid to be disclosed separately and then states the classification. For a financial institution, interest paid and interest and dividends received are operating. For every other entity, interest paid is classified as financing, interest and dividends received are classified as investing, and dividends paid are classified as financing. This is a hard requirement rather than a choice. The international standard on which Ind AS 7 is based leaves those classifications open for non-financial entities, and India removed the option. That single difference is why the third lever above exists, and it is the reason a levered Indian company's operating cash line is not directly comparable with a foreign peer's. One caveat on the citation rather than the substance: the reproduction of the standard consulted places this wording at paragraph 31, while one secondary source places it at paragraph 33, so the paragraph number should be confirmed against the notified text.

Who has to publish one. Section 2(40) of the Companies Act, 2013 defines a financial statement to include a cash flow statement, with a proviso that the financial statement of a one person company, a small company, a dormant company and a private company that qualifies as a start-up need not include one. A listed company falls outside every one of those exemptions, so the statement is mandatory for any company whose shares you can buy on an exchange. The filing timetable and the half-yearly requirement under the listing regulations are set out on our page on what the market regulator actually does and in the companion red-flags page, and are not repeated here.

What could not be verified. It is frequently said that Indian listed companies must use the indirect method. The old Clause 32 of the equity listing agreement is reported to have required it, though that wording could not be read from the regulator's own site, and in any case that agreement was superseded by the listing regulations in 2015. A current instrument compelling the indirect method for equity-listed companies could not be located from a primary source during preparation of this page, and the requirement that was found in a master circular applies to entities with listed non-convertible securities rather than to listed equity. Nor could any published count be found of how many Indian companies use each method. So the honest statement is this: the standard permits both, the indirect method is what Indian annual reports overwhelmingly present, the historical reason is the old listing agreement, and a reader who needs the current position for a specific class of issuer should check the applicable regulation directly. Separately, the criteria for capitalising development cost are described rather than quoted, for the reason given in the section above.

Reading one, in the order that works

The reading order below is not a checklist for its own sake. It is arranged so that each step tells you what to look for in the next, which is the difference between reading a statement and scanning it.

Read the three subtotals as a triple, before any line inside them. Say what the pattern describes. A business funding itself and returning cash, a business borrowing to expand, a business being kept alive, a business selling assets. Ten seconds, and it frames everything else.

Compare the operating subtotal with profit before tax, then find the largest adjustment. On a capital-heavy company it will be depreciation and the comparison is telling you about the asset base rather than the trading. On a working-capital-heavy company it will be receivables or inventories and the comparison is telling you about collection and stock.

Subtract capital spending. This is the step most retail readers never take, and it is one line in the investing section. Then look at depreciation in the same year, because the gap between the two is the rough size of the expansion, and expansion is a choice while maintenance is not.

Read the financing section as the answer to a question the first two sections asked. If operating cash did not cover investing, something in financing made up the difference, and the composition matters: new equity, new debt, or a reduction in the cash balance are three different futures.

Check the payables movement against the cost of goods. Payables growing faster than the cost they finance, for more than one year, is the fingerprint of the first lever above. It is two subtractions.

Read the accounting policy note on capitalisation. This is where the second lever has to be disclosed, and it is the only one of the three that the cash flow statement itself will not reveal, because it moves profit and operating cash in the same direction.

Then do all of that for five or six years at once. A single statement is a photograph of a year, and almost every conclusion worth reaching is about a direction. The subtotals, the days and the gap between capital spending and depreciation all mean something as a sequence and very little as a snapshot.

None of this needs paid data, and none of it needs mathematics beyond addition. What it needs is the willingness to read the second and third sections of a statement most people stop reading after the first, and to keep a column for what the accounts said next to a column for what the bank account did. Those are habits rather than techniques, which is why they are rare, and they are a fair description of the analytical discipline that the method we teach is built around.

Every entity on this page is constructed. None corresponds to any real company, in India or elsewhere, and no real company is named or alluded to anywhere in the text. Each entity is generated from a written set of drivers: a revenue path, an EBITDA margin path, receivable, inventory and payable days, a capital-spending and disposal schedule, a borrowing and repayment schedule, and a dividend. From those the profit and loss account is computed, then the balance sheet, and only then the cash flow statement, which is derived entirely from the movements between consecutive balance sheets with the closing cash balance left as the figure the statement produces.

Four assertions run on every year and the page is not produced if any of them fails. Assets must equal liabilities plus equity. The three sections must add to the net movement in cash. Closing cash must equal opening cash plus that movement. And cash generated from operations, computed by the indirect method, must equal the same subtotal computed independently by the direct method as receipts from customers less payments to suppliers and less payments for other operating costs. The two expressions are algebraically equivalent and share no code, so the check catches implementation errors rather than restating a definition. A further assertion confirms that the figures as printed, at one decimal place, also add, so that a reader adding a column on the page gets the stated total.

Interest is charged at an illustrative nine percent on average borrowings and tax at an illustrative twenty five percent of profit before tax, of which a fifth is treated as deferred. Those are round numbers chosen for legibility. Neither is a statement of any market rate or any statutory rate, and current rates should be verified at source. Depreciation is a fixed proportion of the opening gross block, which keeps the three years comparable and is a simplification rather than a model of any asset register.

Every regulatory provision cited was checked against the text of the instrument or a reproduction of it, with two exceptions stated in the body where the primary document could not be reached or the claim could not be substantiated. All results are illustrative and simulated. They are not a track record, not a forecast, and not an assessment of any company. Nothing on this page is investment advice, and the presence or absence of any pattern described here does not indicate that any company has done anything improper.

• Reading a results announcement →
• Where a dividend is actually funded from →
• What the blue-chip label does and does not mean →
• How Indian and American market structure differ →