Earnings Yield and Return on Capital: Greenblatt's Two Metrics

Joel Greenblatt’s Magic Formula ranks stocks on two metrics: earnings yield (operating earnings divided by enterprise value) and return on capital (operating earnings divided by invested capital). The formula treats these as independent signals: earnings yield finds cheap prices relative to current earnings, while return on capital identifies sustainable competitive advantages. A stock must score well on both to signal both value and quality.

Earnings Yield: The Valuation Filter

Earnings yield inverts the familiar price-to-earnings (P/E) ratio. If a stock trades at a P/E of 10, its earnings yield is 10%. If a stock trades at a P/E of 25, its earnings yield is 4%. Greenblatt uses enterprise value instead of market cap (enterprise value = market cap + debt − cash), and operating earnings instead of net earnings, to avoid distortions from capital structure and one-time items.

This metric answers one question: How much current cash generation do I own for the price I am paying?

An earnings yield of 10% means that the company is generating 10% of enterprise value in cash earnings each year. If earnings are stable, this suggests either a good bargain (the market is pricing in future decline) or a boring, mature business (limited growth, appropriately priced). An earnings yield of 3% means the company is generating only 3% of enterprise value, implying the market is pricing in significant growth. If that growth fails to materialize, the stock will fall.

Earnings yield screens out the most obvious overvaluations. A tech startup trading at 100 times earnings has an earnings yield near 1%—the market is pricing in enormous growth. If the company merely maintains stable earnings, a large valuation reset awaits. Greenblatt’s formula eliminates these extreme cases by screening for stocks with above-median earnings yields.

Return on Capital: The Quality Filter

Return on capital (or return on invested capital, ROIC) divides operating earnings by the total capital employed (debt plus equity). This metric asks: How much profit does the company generate on every dollar of capital it uses?

A company with 20% ROIC generates $0.20 of profit for each dollar invested. A company with 5% ROIC generates only $0.05. Over time, high-ROIC businesses compound wealth far faster than low-ROIC businesses, all else equal.

Why does ROIC matter? Because it measures the durability of the business. A company with high earnings yield but low ROIC is likely in a commoditized industry with high competition and weak pricing power. Its current earnings are strong, but competition will erode margins, and the company will need to invest capital to merely sustain those earnings. It is a value trap: cheap on current earnings but not cheap if earnings fall.

Conversely, a company with high ROIC has a competitive advantage—a moat. It can invest additional capital and expect to earn high returns on that new investment. It can raise prices without losing customers. Its current earnings are less likely to deteriorate. High ROIC, therefore, signals that current earnings are sustainable and likely to grow.

Why Two Independent Metrics?

Greenblatt deliberately chose two metrics that measure different dimensions and can be independent. A cheap stock is not automatically high-quality. A high-quality business is not automatically cheap. By combining them, he aims to capture both:

• Value: Earnings yield ensures you pay a reasonable price for current cash generation.
• Quality: Return on capital ensures the cash generation is durable and likely to persist or grow.

A stock with high earnings yield and low ROIC is a value play in a competitive industry. It trades cheap because the market sees limited growth or competitive pressure. Buying requires conviction that earnings will stabilize or that the market has overshot to the downside.

A stock with low earnings yield and high ROIC is a quality play. The market is pricing in growth because the company has demonstrated the ability to reinvest capital at high returns. Buying requires conviction that the market has rightly priced in growth or that growth will exceed expectations.

A stock with high earnings yield and high ROIC is the Magic Formula’s ideal target: a business that is both cheap (good entry price) and durable (strong competitive position). These stocks are rare and typically represent either temporary dislocations (market pessimism despite good fundamentals) or newly discovered or neglected opportunities.

A stock with low earnings yield and low ROIC is typically avoided: you are paying up for mediocre returns. This is the classic overvalued commodity business.

Separating the Signals: A Worked Example

Consider two hypothetical retailers:

Company A: Price-to-EBITDA multiple of 6 (earnings yield 17%), ROIC 8%. Company B: Price-to-EBITDA multiple of 15 (earnings yield 7%), ROIC 25%.

Company A is cheaper on earnings yield, but its low ROIC suggests thin margins and high asset requirements—a race-to-the-bottom competitive environment. Its cheap valuation may reflect justified pessimism.

Company B is expensive on earnings yield, but its high ROIC suggests brand strength or operational excellence. The market is betting that this company can reinvest earnings at 25% returns, justifying the premium. If that ROIC persists, reinvestment will drive earnings growth, eventually justifying the price.

The Magic Formula does not make a binary choice. It ranks stocks on both metrics and gives higher scores to stocks strong on both. Company B might still make the cut if it ranks in the top quartile on ROIC despite mediocre ranking on earnings yield. Conversely, Company A might be excluded if its ROIC is too low to suggest a durable advantage.

The Operational Earnings Adjustment

Greenblatt uses operating earnings (EBIT or EBITDA), not net income. This avoids distortions from:

• Interest expense: Different capital structures make net income incomparable.
• Taxes: Tax rates vary by jurisdiction and change over time.
• One-time charges: Write-downs or restructuring costs distort net income.

Operating earnings normalize across these differences and focus on the core cash-generation capability of the business.

The enterprise-value denominator (not market cap) ensures that debt-heavy and debt-light companies are compared fairly. A company financed entirely with equity has the same enterprise value as one financed half with debt if they have the same asset base. This prevents bias toward leveraged businesses that inflate returns through financing, rather than operational excellence.

Criticisms and Limitations

ROIC measurement: Return on capital can be gamed. Write-downs, asset revaluations, and aggressive depreciation policies affect the denominator. A company that has written down assets appears to have higher ROIC than an identical company that has not. Greenblatt advocated checking the calculation and using consistent methodologies across the universe of stocks.

Earnings quality: High earnings yield assumes current earnings are representative and sustainable. For cyclical businesses (automakers, semiconductor companies), earnings can be near peak during good years. A steel mill with high earnings yield at peak cycle is not a bargain—earnings will fall sharply as the cycle turns.

Growth assumption: Low earnings yield assumes the market is correctly pricing in growth. But growth forecasts are notoriously unreliable. Many high-flying stocks with depressed earnings yields fail to achieve expected growth, disappointing investors. The formula does not distinguish between justified and unjustified growth expectations.

Market efficiency: If the Magic Formula identifies value, why does the market allow it to persist? One explanation is that institutional constraints (index funds, passive allocation) prevent efficient pricing. Another is that transaction costs and behavioral biases offset the advantage. A third is that the advantage, if real, has decayed as the strategy has become widely known.

Empirical Results

Greenblatt published research showing that portfolios constructed by ranking stocks on the Magic Formula and equal-weighting the top performers beat the market in backtests and in a real-money portfolio during a covered period. However, backtests can be subject to overfitting, survivor bias, and luck. Performance in real-money portfolios has been more mixed.

The formula’s greatest benefit may be disciplinary: it imposes a systematic, quantitative approach to screening that avoids emotional decision-making and forces consideration of both valuation and quality simultaneously.

See also