What is the earnings yield gap and how is it interpreted?

The short answer

The earnings yield gap, sometimes called the Fed Model after a 1997 internal Federal Reserve note that referenced the comparison, computes the difference between the S&P 500’s earnings yield (the inverse of the P/E ratio) and the ten-year Treasury yield. The intuition is that when equity earnings yields exceed bond yields by a wide margin, equities are attractively priced relative to fixed income.

The complication is methodological. Earnings yield is a real (inflation-adjusted) measure — corporate earnings tend to grow with inflation over time, so an earnings yield of 5% retains its purchasing power. The Treasury yield is a nominal measure — its purchasing power depends on the inflation rate over the holding period. Comparing them directly mixes apples and oranges.

This problem was crystallised by Cliff Asness in his 2003 paper Fight the Fed Model, where he showed that the apparent predictive power of the EYG in the 1970s and 1980s came from coincidental co-movement of earnings yields and bond yields driven by the inflation regime, not from an underlying valuation relationship.

→ New to valuation logic? Financial education framework

What the data shows

As of May 2026, the S&P 500’s forward earnings yield stands at approximately 4.5% (the inverse of a forward P/E around 22), while the ten-year Treasury yield ranges around 4.2-4.5%. The earnings yield gap is therefore close to zero — neither historically wide nor narrow.

Three regime points clarify the interpretation. In 1980, with ten-year yields above 11% and equity earnings yields around 13%, the EYG was modestly positive but the entire framework was dominated by inflation expectations. In 2000, with ten-year yields around 6% and equity earnings yields around 4% (forward P/E of 25), the EYG was strongly negative — and indeed forward equity returns over the next decade were poor. In 2020-2021, with ten-year yields under 1.5% and earnings yields around 4%, the EYG was very wide — yet forward returns from those levels have been mixed once inflation returned.

The current near-zero reading is itself notable. It is the first time in two decades that earnings yields and bond yields have aligned closely, reflecting both the rise in long rates from post-2022 monetary tightening and the elevation of equity multiples.

→ Shiller CAPE: monthly history (dataset)

Why it happens — the macro mechanism

The Asness critique runs as follows. Bond yields contain inflation expectations: a nominal yield of 5% might reflect 2% real rate plus 3% inflation, or 1% real rate plus 4% inflation. Earnings yields do not move mechanically with inflation in the short run, but corporate earnings do — companies pass through input costs imperfectly but progressively. Comparing nominal bond yield to nominal-but-real-anchored earnings yield therefore creates a spurious relationship that varies with the inflation regime.

In a high-inflation regime (1970s-early 1980s), bond yields are pushed up by inflation expectations while equity multiples are pushed down by the same expectations, creating a wide spread that does not reflect equity attractiveness — both assets are simply being repriced for inflation. In a low-inflation regime (post-2010), the same logic operates in reverse: low bond yields make equity earnings yields look attractive even when equity multiples are stretched.

Three regime-specific patterns illustrate the point. In the 1980 regime, the EYG was a reasonable proxy for valuation because high inflation was the dominant driver of both yields. In the 2000 regime, the EYG correctly flagged equity overvaluation because the inflation environment was stable. In the 2020 regime, the EYG flagged equity attractiveness during a period when both inflation and rates were artificially compressed, leading to misleading conclusions when conditions normalised.

The signature of the current 2026 regime is the near-elimination of the gap. This does not necessarily mean equities are correctly priced — it means that one historically used valuation framework no longer indicates a clear direction.

What it means for different economic actors

For institutional asset managers, the lesson from Asness is that the EYG should not be used as a stand-alone valuation signal. It can be useful in conjunction with absolute valuation measures (CAPE, price-to-sales) and with explicit inflation regime adjustments.

For retail investors, the practical consequence is that reasoning of the form “stocks yield more than bonds so they must be cheap” is unreliable. The framework worked imperfectly even in its strongest period and breaks under regime shifts.

For pension funds and insurers, the joint movement of bond and equity valuations creates real challenges. When both yields compress together, both asset classes price expensively, leaving few options for required-return allocation.

For households accumulating wealth, the relevant observation is that valuation signals are most reliable when they account for the macro regime in which they are observed. Why valuations matter over the long term develops the framework.

Frequently asked questions

Did the Federal Reserve actually endorse the Fed Model?

No. The “Fed Model” name comes from a single chart in a July 1997 Humphrey-Hawkins report that compared earnings yields and bond yields. The Fed has never endorsed it as a valuation framework, and senior researchers have published critiques. The popular branding is informal and overstates institutional endorsement.

Is there a corrected version of the EYG that handles inflation properly?

The cleanest correction compares earnings yields to real (inflation-adjusted) bond yields, typically using TIPS yields. This produces a real-on-real comparison that avoids the money illusion problem. With ten-year TIPS yields around 2% in 2026, the gap between earnings yield (~4.5%) and real yield (~2%) is approximately 2.5 percentage points — historically modest but interpretable as a real risk premium.

What’s the relationship between EYG and the equity risk premium?

The equity risk premium is conceptually the expected excess return on equities over the risk-free rate. The EYG is sometimes used as a proxy for it, but the inflation problem and the assumption that earnings yields equal expected returns both weaken this connection. Damodaran maintains a more rigorous monthly estimate that adjusts for buybacks, expected growth, and term structure — generally producing implied equity risk premiums in the 4-5% range in recent years.

Last updated — 24 May 2026