Glossary term
Earnings Yield
A cheapness measure expressed as profit over price. Joel Greenblatt's variant — EBIT divided by enterprise value — neutralizes differences in leverage that plain earnings-over-price (E/P) cannot.
EV-Based Yield vs. E/P
Earnings yield inverts a valuation multiple into profit-per-dollar-of-price terms: the classic version is E/P, the reciprocal of the P/E ratio. Greenblatt's variant, used in the Magic Formula, instead divides EBIT — operating profit before interest and tax — by enterprise value, which adds debt to and subtracts cash from market capitalization. The distinction matters because two companies with identical operating profitability can carry very different P/E ratios purely from how much debt sits on the balance sheet: a highly levered company's equity looks cheaper on E/P for reasons that have nothing to do with the underlying business. Dividing by enterprise value instead puts every company on the same footing regardless of financing, which is why the EV-based version is the standard comparability tool in cross-sectional ranking.
Role in Composite Scores
Earnings yield's job in a composite score is narrow and specific: it is the cheapness leg, paired against a separate quality measure rather than blended into it. Cheap alone says nothing about whether a business is any good — a high earnings yield can reflect a genuine bargain or a business the market has correctly priced for decline, and the ratio by itself cannot tell the two apart. That is why Greenblatt's formula, and every serious descendant of it, always reports earnings yield alongside a capital-efficiency measure like ROIC rather than ranking on yield in isolation. A universe ranked on earnings yield alone tends to surface exactly the names a quality screen would separately flag as deteriorating, which is the trap the paired ranking exists to avoid.
How Closelook Uses It
On stock pages, earnings yield sits in the established-ratios layer next to FCF yield and EBITDA multiple — reported with its universe percentile, not folded into a single Closelook verdict. Presenting the ratio on its own, rather than as one input already blended into a composite grade, lets a reader apply Greenblatt's pairing logic directly: check the earnings yield percentile against ROIC before concluding a name is cheap in any useful sense. The same layer carries enough context — EV components, EBIT, and the peer-percentile rank — that the comparison Greenblatt's book asks for can be run from the page itself, without exporting anything to a separate screener. A high earnings-yield percentile paired with a weak ROIC percentile is the specific pattern the book warns against buying on cheapness alone.