can you give me FCF as per my definition - fcf = op income - capex - for mcirosoft - rolliwng ttm window.

I'll calculate Microsoft's Free Cash Flow using your definition (FCF = Operating Income - CapEx) on a rolling TTM basis. Let me load the data first.

Got 20 quarters of data (Q1 FY22 through Q3 FY26). Now let me compute the rolling TTM FCF.

Here's Microsoft's FCF (Op Income − CapEx) on a rolling TTM basis:

Key observations: - FCF TTM peaked at ~$66.9B in Sep 2025 and has been declining since - Op Income TTM has grown strongly, from $82B (Mar 2022) to $149B (Mar 2026) — an ~82% increase - CapEx TTM has exploded, from $23B to $97B — more than 4x — driven by AI/cloud infrastructure investment - The latest reading (Mar 2026) shows FCF TTM of $51.7B, the lowest since 2022, as CapEx is now ~65% of Op Income (vs. ~29% in 2022)

The CapEx ramp is clearly outpacing operating income growth, compressing FCF despite top-line strength.

is this definiton of fcf good to measure the cash generated by the business?

Good question — and the honest answer is: it's an approximation, not the gold standard. Let me break it down:

What your definition captures well

• Core operating earnings power minus reinvestment in PP&E — this is the spirit of FCF
• It's simple and consistent over time
• For a mature, asset-light business, it would be very close to a "true" FCF

What it misses

1. Taxes (usually the biggest gap) Op Income is pre-tax. Real cash available to debt + equity holders comes after taxes. A more standard approach: - EBIT(1−t) − CapEx (e.g., NOPAT-based) - Or just use the cash flow statement's actual OCF − CapEx

For MSFT at a ~17-19% effective tax rate, this would knock ~$25B off your current $149B TTM op income.

2. Non-cash add-backs inside OI Operating income includes D&A as an expense, but D&A isn't a cash outflow. So you're overstating the "cash" side too. Classic FCF formulas use EBITDA − taxes − CapEx to fix this.

3. Working capital changes A sale on 90-day terms generates revenue/op income now but cash later. True FCF nets out ΔNWC. OCF from the cash flow statement already does this.

4. Operating lease obligations Post-ASC 842, lease payments sit in financing/operating cash flow but the "asset" is on the balance sheet. Heavy lease users (retail, airlines) distort this.

5. Stock-based compensation Real cash cost to the company, but usually excluded from "FCF" definitions by sell-side. Worth deciding explicitly.

How it compares to MSFT's actual reported FCF

Let me pull MSFT's cash flow statement so we can compare:

Now you can see how the definitions compare for MSFT (latest TTM, Mar 2026):

Key gaps for MSFT specifically: - D&A is huge and rising fast ($44B TTM, up from $12B in 2021) — CapEx-heavy businesses have big non-cash expenses that your formula double-counts against FCF - Taxes are real cash out ($29B TTM) — your formula ignores them entirely - OCF is well above Op Income because D&A is added back; so standard "OCF − CapEx" actually shows FCF growing, not falling

Bottom line

Your definition is directionally useful for tracking the trend (CapEx pressure is real and visible) but it systematically overstates the cash hit from rising D&A and ignores taxes. For a business like MSFT in heavy CapEx mode, that distortion is large and growing.

If you want a better quick proxy, use: - OCF − CapEx (most common; just pull from the cash flow statement) - Or EBIT(1−t) + D&A − CapEx if you want to stay on the income-statement side

Want me to re-run the TTM trend with one of these cleaner definitions?

Standard: OCF − CapEx - what is this called?

It's commonly called "Free Cash Flow" (FCF) or more specifically "Unlevered Free Cash Flow" (UFCF) — though definitions vary by shop. Here's the terminology landscape:

Common names for OCF − CapEx

• Levered Free Cash Flow (LFCF) — most common name in equity research; "levered" because it's the cash available after the business has met all obligations including debt service
• Free Cash Flow (FCF) — generic, often used loosely to mean this exact thing
• Free Cash Flow to the Firm (FCFF) — sometimes, though this technically should be pre-debt (EBIT-based)
• Cash Flow From Operations minus CapEx — the descriptive name

The naming confusion (it gets messy)

In practice

• Bloomberg FCF field = OCF − CapEx (the standard one)
• Most sell-side reports quote "FCF" as OCF − CapEx when discussing equity stories
• DCF models use the unlevered version (FCFF) because you discount unlevered cash flows at WACC

For MSFT and most US mega-cap coverage, "FCF" = OCF − CapEx is the default and what you're seeing on terminals. Your original Op Income − CapEx version is sometimes called "Operating FCF" or "EBIT-based FCF" but it's less standard.

Want me to chart MSFT's standard FCF (OCF − CapEx) on a TTM basis so you have a cleaner trend to work with?

ok - give me this for msft - opex, capex and (opex - capex)

You already have that data loaded — I computed MSFT's standard FCF = OCF − CapEx on a rolling TTM basis in the previous response. Let me pull that up cleanly.

The table above is trimmed to the last 8 quarters since I built it for the comparison. Let me rebuild with the full TTM history so you get the whole picture from FY22 Q3 onward:

Here's MSFT's standard FCF (OCF − CapEx) on a rolling TTM basis, full history:

Key takeaways:

The story is very different from your original definition: - FCF is still positive and growing, just much more slowly than revenue/op income - OCF nearly doubled — Microsoft's underlying business is generating a lot more cash - CapEx is the absorbing force — it's grown from ~27% of OCF to ~57% - FCF plateaued in the $70-78B range for the last 6 quarters after a steady ramp from 2022 - This is a classic "good business, expensive reinvestment cycle" picture

Compared to your Op Income − CapEx version (which showed FCF falling from ~$66B to $52B), the standard view shows FCF flat-to-slightly-up in the $70-78B band — the difference is largely the rising D&A add-back and timing of working capital.

Want me to layer in FCF margin (FCF / revenue) or FCF as % of OCF to size the CapEx pressure more precisely?