Valuation multiples
FCF yield vs earnings yield: 830 differ
What FCF yield says that earnings yield does not: on 2026-09-07, 496 US listings lost money with positive free cash flow and 334 made money while burning it.
By Javier SanzPublished
Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.
Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)
On 2026-09-07, FCF yield and earnings yield pointed opposite ways for at least 830 US listings: 496 reported a loss alongside a positive free cash flow yield, and 334 reported a profit while free cash flow was negative.
Both answer the same question with two different numerators. One divides last year's accounting profit by the share price. The other divides the cash left after capital spending by the same price. When they point the same way, neither tells you much the other did not. When they point opposite ways, one of them is carrying something.
The two counts
| Condition | Listings |
|---|---|
| Earnings yield below zero, free cash flow yield at or above zero | 496 |
| Earnings yield at or above zero, free cash flow yield below zero | 334 |
Both are exact counts of exactly those conditions. The denominator is harder to state precisely: on 2026-09-07 the earnings yield was held for 4,961 listings and the free cash flow yield for 4,158, out of a pool of 5,257 that day, so no more than 4,158 listings could hold both. That makes 830 at least 20.0 percent of the testable population, and the true share is higher than that rather than lower.
One listing in five, at minimum, has an accounting result and a cash result that disagree about whether the year was positive.
The 496 that lost money and produced cash
This is the larger group and the more interesting one, because it is where a purely accounting screen throws away companies for a reason that did not happen in the bank account.
The free cash flow yields involved are not marginal. Distribution of the 496:
| Free cash flow yield | Listings |
|---|---|
| 0 to 1% | 85 |
| 1 to 2% | 47 |
| 2 to 3% | 44 |
| 3 to 5% | 62 |
| 5 to 8% | 66 |
| 8 to 12% | 59 |
| 12 to 20% | 52 |
| 20 to 50% | 46 |
| Above 50% | 35 |
One hundred and ninety-two of the 496 carried a free cash flow yield of 8 percent or more while reporting a loss.
The mechanism is not mysterious. A charge that reduces reported profit without moving cash produces exactly this signature: impairments, write-downs of assets carried above their worth, and provisions for costs not yet paid. All three reduce the earnings yield to a negative number and leave the cash flow statement alone.
Among the largest listings in this group by market capitalisation on 2026-09-07, sampled from those with a free cash flow yield of 8 percent or more: Ford Motor Company at a free cash flow yield of 22.54 percent against an earnings yield of minus 14.76; Centene at 13.38 against minus 20.67; Kraft Heinz at 12.44 against minus 19.87; Omnicom at 15.48 against minus 0.30; Viatris at 9.87 against minus 17.91; Molson Coors at 12.74 against minus 25.53.
These are not small companies and this is not an edge case. A screen with a positive earnings condition on it removes all of them, silently, and returns a result that looks complete.
The 334 that reported a profit and burned cash
The reverse group is smaller and the reason it exists is different. Profit can be reported while cash goes out when working capital absorbs it, when capital spending exceeds operating cash flow, or when revenue is recognised before it is collected.
Distribution by earnings yield, so you can see where in the profitable range these sit:
| Earnings yield | Listings |
|---|---|
| 0 to 1% | 43 |
| 1 to 2% | 47 |
| 2 to 3% | 21 |
| 3 to 5% | 51 |
| 5 to 8% | 49 |
| 8 to 12% | 35 |
| 12 to 20% | 22 |
| 20 to 50% | 22 |
| Above 50% | 44 |
Forty-four of the 334 report an earnings yield above 50 percent, which is a price to earnings ratio below two, alongside negative free cash flow. That combination is the one worth looking at hardest, because a very high earnings yield is the shape a screen sorted on cheapness puts at the top of the page, and the cash statement is saying something the income statement is not.
Which one to believe
Neither, on its own, and that is the point of holding both.
A negative earnings yield with a strong cash yield is a question about the charge. Find it in the filing. If it is an impairment of something bought years ago, the cash left long ago and the accounts have caught up. If it is a provision, the cash has not left yet and will.
A positive earnings yield with a negative cash yield is a question about working capital and capital spending. Growth absorbs cash, which is normal and reversible. Receivables growing faster than revenue is not.
A single year decides nothing either way. Free cash flow is lumpy: one heavy investment year makes a good business look like a cash sink. Both figures here are one year against one price.
What this leaves out
Every count is one reading of the public screener on 2026-09-07, on prices stamped 2026-08-26. Both yields divide by a price that was twelve days old when the counts were taken.
The denominator of the paired test is not measured. The count of listings holding both figures cannot be obtained from the public endpoint, so 4,158 is used as an upper bound on it and every share stated here is a lower bound on the true share.
Both yields are clamped to a range of minus 200 to 200 percent, so a listing with a near-zero price cannot produce an unbounded figure. The counts in the tails include clamped values.
Naming a company is a report of two figures on one date. It is not an assessment of the business, and the charge behind any particular negative earnings yield is not identified here: finding it means opening the filing.
Free cash flow is operating cash flow less capital expenditure as filed. It does not separate maintenance spending from expansion, so a company investing heavily to grow reads the same as one that cannot fund itself.