NBER WORKING PAPER SERIES PRICING WITHOUT MISPRICING ...
Using a clean test with decade-old information, the paper asks which asset pricing models assign zero alpha when mispricing should be gone. CAPM passes. Multifactor models like FF3, FF5, FF6, Q4, and Q5 fail, notably for large liquid stocks. The method sorts on 10-year-old FF3 expected returns and tests 1968 to 2024. Persistent betas inflate alphas despite near zero spreads, plus 41 bps for mega caps, t=2.62. Conclusion: multifactor models embed temporary mispricing for investors.
What it examines
This study asks which asset-pricing models could price assets if markets were fully efficient. It builds long--short portfolios using only decade-old FF3-based return forecasts and tests whether models assign zero alpha. Power comes from persistent betas, especially in large caps. CAPM and simple variants are compared to prominent multifactor models.
What it concludes
Results: multifactor models (FF and q) give significant alphas on decade-old sorts and fail as no‑mispricing benchmarks; CAPM and simple two-factor variants pass. Implications: prefer simpler benchmarks. Limits: low power, decade-old assumption. Future work: other no‑mispricing settings. Applications: testing efficiency, model validation, and avoiding distorted signals in portfolio design.
Evidence objects
New paper tests pricing without mispricing by using decade-old data to strip mispricing; shockingly, CAPM assigns zero alpha while FF3/FF5/FF6 and Q4/Q5 fail, most starkly among large, liquid stocks today.
key_findings bullet 1 · key_findings · validation V0
Authors sort stocks on 10-year-old FF3-based expected returns, build value-weighted bottom-versus-top 20% spreads using decade-old weights, and test alphas from 1968--2024 across size; transparent R-squared show persistent large-cap betas power.
key_findings bullet 2 · key_findings · validation V0
Alpha decompositions show lingering MKT/HML or IA exposures inflate alphas despite near-zero spreads (mega caps: +41 bps, t=2.62, FF3). Conclusion: multifactor embed mispricing; simpler pass. Caveats: 10-year assumption, small-cap, horizons.
key_findings bullet 3 · key_findings · validation V0
This paper introduces a pragmatic test for the joint-hypothesis problem: assess alphas from portfolios built solely on decade-old information, presumed efficiently priced. Leveraging persistent multifactor betas for power, it finds $CAPM$ and simple variants pass while multifactor models fail, challenging orthodoxy and elevating simpler equilibrium benchmarks for academia and practice.
key_findings bullet 4 · key_findings · validation V0
Raw abstract and provenance
We empirically confirm that larger stocks have more persistent and more precise multifactor betas. The type of securities affording our test the most power are
Source row: 1422 · abstract type: snippet