Anomaly Replication Lab
  • The zoo
  • Screener
  • Multiple testing
  • Correlations
  • Composite
  • Methodology
  • Guides
  • Resources
Sign in
Anomaly Replication LabSign in
  • The zoo
  • Screener
  • Multiple testing
  • Correlations
  • Composite
  • Methodology
  • Guides
  • Resources

Educational replication, not advice. Long-short returns are decile spreads or published factor series from the Kenneth R. French Data Library and AQR, value-weighted, US equities, and gross of transaction and shorting costs — which for high-turnover signals like short-term reversal is the whole story. Reproducing a premium in-sample is not evidence it will persist. Past performance does not predict future results.

© 2026 Austin Levenberg. All rights reserved. This site’s code, design, written analysis, and computed results are not licensed for reuse or redistribution — terms of use · privacy.

Data: Kenneth R. French Data Library, AQR Capital Management public data sets.

Part of austinlev.dev — four research tools, and a note on where this is headed.

← the zoo
ValueGone

Dividend yield

Higher-dividend-yield stocks have earned higher returns.

Litzenberger & Ramaswamy · Journal of Financial Economics · published 1979 · original sample 1936–1977 (reported ≈ +4.0%/yr, t ≈ 2.4)

How our version differs from the paper. Zero-dividend firms (now most of the market) sit in a separate bucket; this is effectively a slow value sort on the dividend-payers. The decay figure compares our replication in-sample to our replication after publication, so it isn’t affected by this — but the level gap between “reported” and “our replication” partly is.
Our replication
-0.6%
1927–2026, per year
Full-sample t
0.6
Sharpe 0.06
Post-publication
+0.5%
since 1980, t = 0.2
Decay
74%
of the premium is gone

The whole history

Growth of $1 in the long-short strategy, gross of trading and shorting costs. The dashed line marks the year the paper was published — everything to its right is out-of-sample.

$1 in the long-short strategy, gross of costs. Grey = the sample the paper had; blue = everything since.

Before and after publication

The comparison the site is built around.

WindowReturn/yrVolSharpet-statMax DD
Paper's sample (1936–1977)+1.0%13.7%0.140.9-64%
After publication (1980–2026)-1.4%19.3%0.030.2-76%
Full sample (1927–2026)-0.6%18.8%0.060.6-86%

Against the factor models

Is there a risk-adjusted alpha left once you control for the standard factors? Alpha is annualized; t-stats use Newey–West standard errors.

ModelAlpha/yrAlpha tR²Biggest loadings
CAPM (market only)+3.9%1.70.19Mkt-RF -0.52
Fama–French 3-factor-0.9%−0.50.52HML +1.06, Mkt-RF -0.38, SMB +0.01
Fama–French 5-factor-1.5%−0.90.53HML +0.87, CMA +0.43, Mkt-RF -0.34
FF5 + momentum+0.3%0.10.55HML +0.78, CMA +0.49, Mkt-RF -0.37

Rolling performance

Trailing 3-year Sharpe ratio. Long stretches below zero are common.

By decade

1920s-13.5%/yrSharpe -0.72
1930s+7.0%/yrSharpe 0.23
1940s+7.7%/yrSharpe 0.71
1950s-2.4%/yrSharpe -0.20
1960s-3.8%/yrSharpe -0.28
1970s+3.8%/yrSharpe 0.21
1980s+3.6%/yrSharpe 0.19
1990s-10.6%/yrSharpe -0.59
2000s+11.2%/yrSharpe 0.51
2010s+1.9%/yrSharpe 0.13
2020s-5.8%/yrSharpe -0.25

The risk explanation

Originally a tax story — dividends were taxed harder than capital gains, so high-yield stocks had to offer a pre-tax premium.

The behavioral explanation

A preference for cash income ('bird in the hand'); retail demand for yield can push it either way.

Why it might have decayed

Tax rates on dividends fell sharply (2003 US cut), removing the original mechanism. Zero-dividend firms — now a huge share of the market — have to be handled separately, and the sort is really a slow value sort.

Long-short = top-decile minus bottom-decile of stocks univariately sorted on the characteristic. Value-weighted, US equities, monthly, gross of transaction and shorting costs. The original-paper figures are approximate and annualized for comparison.