Accounting Valuation, Market Expectation, and Cross-Sectional Stock Returns

Source: Frankel & Lee (1998) · Journal of Accounting and Economics · DOI: 10.1016/s0165-4101(98)00026-3

TL;DR

Frankel & Lee estimate each firm's fundamental value (V) from analyst forecasts via a residual-income model, then show the value-to-price (V/P) ratio predicts long-run cross-sectional returns: firms cheap relative to their analyst-based intrinsic value outperform. The effect survives controls for beta, book-to-price, and size.

What anomaly it documents

Predictor: value-to-price ratio (V/P), with V from a residual-income model on I/B/E/S forecasts.

Direction: positive — high V/P (undervalued) firms earn higher long-run returns.

Shape: monotone; predictive at 1–3 year horizons; not explained by B/P or size.

OSAP predictor: PredictedFE.

How to construct it

Sorting variable: V/P, where V is residual-income intrinsic value using consensus forecasts.

Universe: firms with analyst coverage.

Portfolio formation: rank into V/P deciles.

Long / short: long high V/P, short low V/P.

Weighting: equal-weighted.

Rebalancing: annual.

Evidence and replication

Period	Notes	Source
IS (1975–1993)	V/P predicts long-run returns beyond B/P, beta, size	this paper
OOS (post-1998)	intrinsic-value investing signal; persists, decays	post-publication
OSAP (PredictedFE)	replicates	Chen & Zimmermann 2022

Why it might work

Mispricing: the market underweights forecast-implied fundamentals.

Better value metric: V/P conditions cheapness on expected profitability, unlike raw B/P.

Limitations and risks

Forecast dependence: requires analyst estimates and model assumptions.

Coverage: limited to followed firms.

Value overlap: correlated with book-to-price.

Key references

Frankel, R. & Lee, C. (1998) — Accounting Valuation, Market Expectation, and Cross-Sectional Stock Returns — JAE — DOI: 10.1016/s0165-4101(98)00026-3

Provenance: generated from the paper's abstract and metadata, not full text; sample periods and replication notes are indicative — verify against the source.