Skip to main content

Earnings & Cash-Flow Yield

Updated Jun 24, 2026 at 2:35pm

Research Draft High 1,188 words

Earnings yield and cash-flow yield are the two most-used flow-based value signals: they ask "how much profit, or how much spendable cash, do I get per dollar I pay for this business?" Both are simply valuation ratios inverted — yields rather than multiples — so a high reading means a stock is cheap relative to what it earns or generates in cash. The core tension is that earnings are an accounting construct (subject to accruals, depreciation policy, and management discretion), while free cash flow is closer to real money but noisier year-to-year and distorted by lumpy capital spending. Within the value factor, these flow metrics are the empirical workhorses that have largely displaced book-to-market as the preferred cheapness signal among practitioners.

How it's calculated / formed

Earnings yield (E/P) is the reciprocal of the P/E ratio: earnings per share divided by price (or aggregate net income divided by market cap). It is most commonly built on trailing-twelve-month earnings, and is the metric used in the "Fed model," which compares the S&P 500 earnings yield to the 10-year Treasury yield — a heuristic, not a validated forecasting tool.

Cash-flow yield comes in two forms (per Wall Street Prep):

  • Levered / equity FCF yield = Free Cash Flow to Equity ÷ Equity Value (or FCF per share ÷ price). Equity FCFE is operating cash flow minus capex minus mandatory debt service.
  • Unlevered / enterprise FCF yield = Free Cash Flow to Firm ÷ Enterprise Value. FCFF ≈ NOPAT + D&A − capex − change in net working capital.

The enterprise-value denominator matters: EV = market cap + debt − cash, so it prices the whole business and is not distorted by leverage. This is why quant value researchers increasingly normalize all cheapness metrics by EV — yielding the closely related family EBIT/EV, EBITDA/EV, and FCF/EV. Common defaults: trailing-12-month numerator, EV denominator, sector-relative ranking.

How it's used in practice

In a factor framework, you rank a universe by yield (high = cheap), then go long the cheapest decile/quintile and sometimes short the most expensive, rebalancing quarterly or annually. In discretionary work, FCF yield serves as a sanity check — a 7–8% FCF yield on a stable business signals the market is paying little for its cash generation, while a high earnings yield paired with a low FCF yield is a red flag that reported profit isn't converting to cash (heavy accruals, aggressive revenue recognition, or capex sinkholes).

Practitioners favor FCF yield for two reasons cited consistently across sources: cash is harder to manipulate than earnings, and it directly measures capacity to pay dividends, buy back stock, and service debt. The reported-FCF average for the S&P 500 has historically sat around 4–5%; value indices typically screen for 6–8% (Quant Investing). Many shops combine yield with a quality gate (Piotroski F-score, FCF stability, or accruals screen) to avoid value traps — a structurally declining business can show a seductive yield right before earnings collapse.

Adoption, debate & evidence

Flow-based value is the dominant value implementation in quant practice, but the evidence has real nuance:

  • EV-normalized metrics tend to win head-to-head. Loughran & Wellman (2011, JFQA) documented a return premium of roughly 5.3% per year for the cheap-enterprise-multiple (EBITDA/EV) portfolio and argued it is the most profitable single price ratio for stock selection. Gray & Vogel's Alpha Architect study Analysing Valuation Measures: A Performance Horse-Race over the past 40 Years (1971–2010, as reported by Quant Investing) ranked EBITDA/EV first at ≈ 17.7% annualized and FCF/EV second at ≈ 16.6% — both beating raw E/P and book-to-market; EBIT/EV performed near-identically to EBITDA/EV in their work.
  • FCF yield's standalone edge is real but modest. A figure of roughly 3–5% annual top-decile outperformance is commonly cited and attributed to O'Shaughnessy's What Works on Wall Street (this exact range comes via Quant Investing's paraphrase rather than a verified table in the book). A Merrill Lynch US Quantitative Strategy study (~30-year window), also cited by Quant Investing, found FCF yield delivered the highest return and the fewest periods of negative returns among the metrics it tested. These are practitioner backtests, not peer-reviewed, and are subject to the usual data-mining caveats.
  • Yield ≠ profitability, and the distinction matters. Recent work (Abacus FCF, 2025) separates FCF yield (a value signal) from FCF profitability (a quality signal). During the stressed 2022–2025 window that paper reports FCF-yield's long-short Sharpe deteriorated ~41% versus history while FCF-profitability's improved — a caution that cheap-on-cash-flow is not robust through all regimes.
  • The broader value factor has been contested. Fama-French codified value via book-to-market (HML), but HML delivered a long, painful drawdown from roughly 2007–2020, prompting debate about whether value is "broken" or whether book value is simply a poor denominator for asset-light, intangible-heavy firms — which is precisely the argument for using earnings and cash-flow yields instead.

Honest summary: the value premium itself is one of the better-documented factors, and flow-based yields (especially EV-normalized) are the most defensible way to express it — but the precise return numbers above are backtest folklore-grade, not settled fact, and the signal is regime-dependent.

Strengths & limitations

Works best for mature, cash-generative businesses with stable capex, applied as a relative (sector-neutral) rank within a diversified basket and paired with a quality filter.

Fails in predictable ways: (1) value traps — cheapness that reflects genuine secular decline; (2) financials and capital-light tech, where FCF and EV are ill-defined or where book/cash-flow metrics misfire; (3) single-year FCF noise — one big capex year or working-capital swing can hide a strong business, which is why some practitioners normalize FCF over a cycle; (4) buyback/SBC distortion in equity FCF yield. The #1 misuse is treating a high earnings yield as a buy signal without checking cash conversion — earnings that don't become cash are the classic precursor to a value trap.

Sources

Dispute flags: the specific annualized backtest returns are practitioner (non-peer-reviewed) figures and are regime-dependent; EV/EBITDA vs FCF-yield ranking varies by study and period; the Fama-French value premium has been actively contested post-2007.