Free Cash Flow vs EPS: Finding the FCF Number That Matters

Earnings per share is the number companies want you to anchor on. Free cash flow is the number you should actually anchor on. The difference is not academic — it routinely flips the investment case on names that look cheap on EPS and bleeds cash, or look expensive on EPS and gushes it.

This is a framework piece. The goal: by the end, you can pull a 10-K, find the FCF figure that matters, and know when management is showing you a flattering version.

Why EPS Misleads Long-Term Investors

EPS is net income divided by share count. Both inputs are softer than they sound.

Net income runs through GAAP rules that don't track cash. Depreciation is a real expense for a steel mill but a near-fiction for a software company amortizing capitalized R&D. Stock-based compensation (SBC) — the equity a company hands to employees — is a real cost to you as a shareholder (your slice shrinks) but gets added back in many "adjusted" presentations. Restructuring charges, goodwill impairments, and one-time tax items can swing reported EPS by significant amounts with no change in the underlying business.

Share count is the other lever. A company can grow EPS with zero operational improvement just by buying back stock — which is fine, except buybacks consume cash you might have wanted reinvested or returned as dividends. EPS doesn't tell you which is which.

Free cash flow strips most of this away. It's a simpler question: after running the business and paying to maintain it, how much cash is left for shareholders?

The Standard FCF Formula and Its Holes

The textbook definition:

FCF = Cash from Operations − Capital Expenditures

Both numbers sit on the cash flow statement. CFO is near the top, CapEx is in the investing section, usually labeled "purchases of property, plant and equipment."

This works as a starting point. It has three holes worth knowing about.

Hole 1: Stock-based compensation. SBC is a non-cash expense, so it gets added back to net income inside CFO. That inflates CFO. For a mature industrial, SBC is a rounding error. For a large-cap software company, SBC can run significantly as a percentage of revenue. If you're using reported FCF on a name like that, you're overstating economic cash flow by a lot. The fix: subtract SBC from FCF, or at minimum track FCF-per-share including the dilution SBC creates.

Hole 2: Working capital swings. A great quarter for collections or a one-time inventory drawdown can flatter CFO. The reverse is also true — a fast-growing company often shows weak FCF because it's funding receivables and inventory. Look at FCF over a trailing four-quarter or three-year window to smooth this.

Hole 3: Maintenance vs growth CapEx. Companies report total CapEx, not the split. A retailer building new stores and a retailer just keeping the lights on look identical on the cash flow statement. If you want "owner earnings" — Buffett's term for cash a business could distribute without shrinking — you want maintenance CapEx only. Most companies don't disclose the split. Reasonable proxies: depreciation expense, or the historical CapEx-to-sales ratio during a no-growth period.

How to Find the FCF Number That Matters

Here's a workable process for any S&P 500 name:

1. Pull the cash flow statement from the latest 10-K (annual) or 10-Q (quarterly). Use annual data first — quarterly FCF is noisy.
2. Start with Cash from Operations. This is line one of your calculation.
3. Subtract total CapEx. That's reported FCF.
4. Subtract stock-based compensation. Found inside the CFO reconciliation as an add-back. This gives you what some analysts call "FCF ex-SBC" or shareholder FCF.
5. Check the lease line. Since ASC 842 (2019), operating lease payments sit inside CFO, but finance lease principal payments sit in financing activities. For asset-heavy retailers, restaurants, or airlines, add back finance lease principal to get a comparable number across periods.
6. Divide by diluted share count (fully diluted, including unvested RSUs) to get FCF per share.
7. Compare FCF yield to the 10-year Treasury. FCF yield = FCF per share / stock price. A 4% FCF yield against a 4.3% Treasury means you're being paid nothing for equity risk unless growth is real.

Do this for three years of history. The shape matters as much as the level — is FCF growing, flat, or being propped up by working capital benefits that won't repeat?

When EPS and FCF Diverge — and What It Tells You

Persistent gaps between EPS and FCF are signals, not noise.

• EPS > FCF for years: aggressive revenue recognition, ballooning receivables, or capitalizing costs that should be expensed. Worth a closer look at accounts receivable days and the accrual ratio.
• FCF > EPS for years: often a sign of a high-quality business. Depreciation exceeds maintenance CapEx, working capital is a source (think subscription businesses collecting cash upfront), or the company is over-earning on a tax basis. Many compounders live here.
• Volatile gap: cyclical business, or a company with lumpy CapEx cycles. Use a multi-year average.

The direction of the gap tells you whether accounting is masking strength or weakness.

What to Watch Next

• Pull one name you own and calculate FCF ex-SBC for the last three fiscal years. Compare to reported EPS growth over the same period.
• Check the SBC line as a percentage of revenue. Anything above a meaningful threshold deserves a dilution-adjusted view.
• For asset-heavy names, separate maintenance from growth CapEx using depreciation as a rough floor.
• Build an FCF yield column in your watchlist and compare it to the 10-year Treasury before you call anything "cheap."