What free cash flow measures, and why there are three of them
Free cash flow answers a question net income cannot: after paying suppliers, employees, tax and the cost of keeping the asset base intact, how much cash is genuinely available to the people who financed the business? Net income is an accrual figure full of judgement — revenue recognised before collection, depreciation on decade-old decisions, provisions that may reverse. Free cash flow strips most of that out and measures money.
There are three versions because there are three audiences, and mixing them up is the most common error in valuation work.
Simple FCF is operating cash flow less capital expenditure — what a screener and most press releases mean. Because US GAAP puts interest paid inside operating cash flow, this figure is already net of the cost of debt: it is a levered measure whether or not anyone says so.
FCFF, free cash flow to the firm, is built as though the company had no debt at all. You start from operating profit, tax it, add back non-cash depreciation, and subtract the investment in working capital and fixed assets. Because financing has been removed, FCFF belongs with a weighted average cost of capital and produces an enterprise value.
FCFE, free cash flow to equity, takes FCFF and puts debt back: it deducts after-tax interest and adds net new borrowing. It is the cash a shareholder could in principle be paid, and it belongs with the cost of equity.
The two routes to FCFF, and why they must agree
You can build FCFF from the income statement or from the cash flow statement, and under US GAAP the two are algebraically identical. Understanding why is the single most useful thing on this page.
Start from operating cash flow under the indirect method: net income, plus non-cash charges, less the increase in working capital. Net income itself is (EBIT − Interest) × (1 − t), so operating cash flow becomes EBIT(1 − t) − Interest(1 − t) + D&A − ΔNWC. Add after-tax interest back and subtract capital expenditure, and the interest terms cancel: what remains is EBIT(1 − t) + D&A − ΔNWC − CapEx, exactly FCFF. The identity holds because the only difference between the starting points is the financing cost, and both routes remove it.
That makes the difference between the routes a diagnostic rather than a nuisance. If they disagree, operating cash flow contains something your income-statement inputs do not: stock-based compensation, deferred tax movements, an impairment, a gain on disposal, or a provision. This calculator computes both and reports the gap so you can name the missing item instead of picking whichever answer you prefer.
Two conventions decide the sign of every term. ΔNWC is positive when working capital grows, because building receivables and inventory consumes cash; a release is negative and adds to free cash flow. And net working capital here excludes cash and short-term debt, since including cash would make the measure circular. If your cash conversion cycle is lengthening, this term is where you will see it.
One caveat for non-US filers: IAS 7 lets a company classify interest paid as operating or financing. If it sits in financing, operating cash flow is already pre-interest and the cross-check double-counts it.
Worked example: a manufacturer with $8 billion of revenue
Take a company reporting $8,000M of revenue, $1,250M of EBIT, a 25% effective tax rate, $450M of depreciation and amortisation, a $120M build in working capital, $520M of capital expenditure, $90M of interest expense and $40M of net new borrowing. Net income is $870M and operating cash flow is $1,200M.
- Tax the operating profit. NOPAT = $1,250M × (1 − 0.25) = $937.5M. This is what operating profit would be after tax if the company carried no debt.
- Add back depreciation. $937.5M + $450M = $1,387.5M. Depreciation reduced EBIT but moved no cash.
- Subtract the working capital build. $1,387.5M − $120M = $1,267.5M.
- Subtract capital expenditure. $1,267.5M − $520M = $747.5M of FCFF.
- Deduct after-tax interest. $90M × 0.75 = $67.5M, leaving $680M.
- Add net new borrowing. $680M + $40M = $720M of FCFE.
- Compute simple FCF. $1,200M − $520M = $680M.
- Run the cross-check. $1,200M + $67.5M − $520M = $747.5M, which matches the FCFF from step 4 exactly. The gap is zero, so nothing unexplained is hiding in operating cash flow.
Read the three side by side. FCFF of $747.5M is the biggest because it ignores debt. FCFE of $720M is next, flattered by $40M of fresh borrowing. Simple FCF of $680M is the strictest of the three, and it is the number to quote if someone asks what the business actually threw off. The margin is $680M ÷ $8,000M = 8.5%, and cash conversion is $680M ÷ $870M = 78%.
How to read the result: margin, conversion and reinvestment
Three ratios turn the dollar figure into something comparable.
Free cash flow margin is FCF divided by revenue. It varies enormously by business model, so compare a company only with itself over time and with direct competitors. Asset-light businesses that collect before they spend — software, franchising, licensing — convert a far larger share of revenue than distribution, construction or airlines, where working capital or fixed assets absorb the cash first. The worked example above lands at 8.5%. What matters is the trend and whether the margin is stable enough to fund a dividend.
Cash conversion is FCF divided by net income. Sustained conversion near or above 100% means reported profit is genuinely turning into cash. Persistently well below that is worth chasing: the usual causes are receivables growing faster than sales, inventory building ahead of demand, capitalised costs inflating profit, or capex permanently above depreciation. Any can be legitimate; none should be assumed to be.
Reinvestment ratio is capital expenditure over depreciation. Above 1.0 the asset base is growing; below 1.0 it is shrinking in book terms. A company can lift free cash flow for several years simply by cutting capital spending below depreciation, and that is one of the easiest manipulations to spot — free cash flow rises while revenue stalls.
Negative free cash flow is not automatically bad. A company building a plant should consume cash. The question is whether the spending is discretionary growth capital or unavoidable maintenance, and whether the shortfall is funded by borrowing that future free cash flow must repay.
Which cash flow measure to use for which job
| Measure | Built from | Deducts capex | Treatment of debt | Discount at | Gives you |
|---|---|---|---|---|---|
| FCF (CFO − capex) | Cash flow statement | Yes | Interest already deducted inside CFO | Not used directly | A quick read on cash generation |
| FCFF (unlevered) | EBIT, or CFO plus after-tax interest | Yes | Removed entirely | WACC | Enterprise value |
| FCFE (levered) | FCFF, or CFO less capex plus net borrowing | Yes | Interest and principal flows included | Cost of equity | Equity value directly |
| EBITDA | Income statement | No | Removed entirely | Never — it is a multiple base | A proxy for operating scale, not cash |
EBITDA appears here because it is so often used as though it were free cash flow. It ignores tax, working capital and capex — the three things that decide whether operating profit becomes money.
Free cash flow is not a GAAP measure
No accounting standard defines free cash flow. ASC 230 defines the operating, investing and financing sections you draw the inputs from, but the subtraction is yours. A company presenting free cash flow in an SEC filing or earnings release is presenting a non-GAAP financial measure, and Regulation G and Item 10(e) of Regulation S-K require it to be reconciled to the most directly comparable GAAP figure — normally net cash provided by operating activities. Read that reconciliation: definitions differ on capitalised software, finance-lease principal, intangibles and pension contributions, and each choice can move the number by double digits.
Adjustments that quietly change your free cash flow
- Finance lease principal payments. Under ASC 842 the principal portion is a financing outflow, so it never reduces operating cash flow. A lease-heavy company can show strong FCF while its committed obligations grow.
- Stock-based compensation. Added back as non-cash, which lifts FCF. It is a real cost to shareholders through dilution and the usual culprit when the two FCFF routes disagree.
- Capitalised software and development costs. Capitalising rather than expensing raises operating cash flow and capex by the same amount, leaving FCF unchanged — unless you omit the capitalised amount from capex, in which case FCF is overstated.
- Acquisitions. Buying capacity instead of building it moves the spend out of capex into a separate investing line, flattering FCF. For a serial acquirer, treat acquisition spend as reinvestment.
- Receivables factoring and supply-chain finance. Selling receivables or stretching payables through a bank programme pulls cash forward and appears as a working-capital release. The benefit reverses if the programme stops.
- Cash taxes versus book taxes. The FCFF formula taxes EBIT at an effective rate. Where accelerated depreciation makes cash taxes much lower, use the cash rate or model the deferral.
How free cash flow feeds a valuation, and when to use something else
Free cash flow is the input to almost every intrinsic valuation. Project FCFF, discount it at WACC, add a terminal value, and you have enterprise value; subtract net debt and you have equity value. That is the architecture of a discounted cash flow model. Working with FCFE instead lets you discount at the cost of equity and skip the net-debt bridge — convenient when capital structure is stable, awkward when it is not. See the FCFE calculator for that route.
Free cash flow is the wrong tool in three situations. For a bank or insurer, capital expenditure and working capital are not meaningful and lending itself looks like an investing outflow; use dividend discount or residual income methods. For a pre-revenue company it is negative by construction and carries no information. And where capital spending is violently lumpy, a single year tells you nothing — average capex over a full cycle first.
Two adjacent figures are worth computing alongside this one. NOPAT starts the FCFF build-up and is the numerator of return on invested capital. EBITDA is where most people begin and should never be where they stop. To see how the cash flow statement itself is assembled from net income, work through the indirect-method operating cash flow calculator — every input here comes from that statement.
Key terms
- NOPAT
- Net operating profit after tax: EBIT multiplied by one minus the tax rate. Operating profit as if the company were all-equity financed.
- Unlevered vs levered
- Unlevered measures (FCFF) are computed before any financing cost. Levered measures (FCFE, and simple FCF under US GAAP) are computed after it.
- Net working capital
- Non-cash current assets less non-debt current liabilities. Cash and short-term borrowings are excluded so the measure reflects operating investment only.
- Maintenance vs growth capex
- Maintenance capex keeps existing capacity running; growth capex adds capacity. Companies rarely split them, so depreciation is the usual proxy.