Chan–Jegadeesh–Lakonishok: Earnings Momentum

#The Paper at a Glance

Title: Momentum Strategies Authors: Louis K. C. Chan, Narasimhan Jegadeesh, Josef Lakonishok Published: Journal of Finance, 1996

Three years after Jegadeesh and Titman's headline momentum paper, the next natural question was: which kind of momentum matters most? Price-based continuation could reflect slow diffusion of fundamental news, noise-trader feedback loops, or risk compensation. Earnings-based continuation — the idea that big upside surprises predict further positive earnings news and further price appreciation — had been documented separately since the 1960s (Ball and Brown; Bernard and Thomas). Chan, Jegadeesh, and Lakonishok combined the literatures. They ranked stocks on price momentum and on earnings momentum simultaneously, examining whether the two sources of continuation were redundant, complementary, or overlapping.

#What the Paper Found

Price Momentum Reconfirmed

Consistent with Jegadeesh and Titman, cross-sectional price momentum at 6 and 12 months produced significant long-short spreads over holding periods of 3 to 12 months. The winners continued, the losers continued to lag, and the spread was robust to standard controls.

Earnings Momentum Signals

The authors constructed multiple earnings-based momentum signals:

1. 1Standardized Unexpected Earnings (SUE): The most recent earnings surprise scaled by its historical standard deviation.
2. 2Cumulative Abnormal Return around earnings announcement: The market-adjusted return in a tight window around earnings.
3. 3Analyst revisions: The change in consensus analyst forecasts relative to pre-revision levels.

Each signal independently predicted future returns over horizons of 6 to 12 months.

Independent Information

When price momentum and earnings momentum signals were used together, the combined long-short spread was larger than either component alone. Regressions showed that both carried incremental information — neither subsumed the other.

Speed of Correction

Price momentum drift was more gradual; earnings momentum (especially PEAD) corrected somewhat more quickly but still over several months rather than days. The slow absorption of public earnings news is one of the most persistent and best-documented anomalies in equity markets.

Underreaction, Not Risk

The authors argued that the most plausible explanation was investor underreaction to news — particularly to the implication that good earnings tend to be followed by more good earnings. Standard risk controls (size, book-to-market, market beta) did not explain the spread. The evidence weighed toward a behavioral or information-diffusion story rather than time-varying risk premia.

#The Math (Lite)

The Standardized Unexpected Earnings (SUE) signal, central to the paper, is defined as:

SUE_i,t = (EPS_i,t - Expected_EPS_i,t) / sigma(EPS surprises for firm i)

where Expected_EPS_i,t is typically the consensus analyst forecast or a time-series model forecast, and the denominator is the historical standard deviation of the firm's earnings surprises. A large positive SUE means the firm beat expectations by an abnormally large margin relative to its own historical variance.

Analyst revision momentum uses:

revision_i,t = (consensus EPS_t - consensus EPS_{t-k}) / price_i,t

where k is a revision window, often one to three months.

To combine price and earnings momentum in a regression:

r_i,t+1 = alpha + b_1 * price_momentum_i,t + b_2 * SUE_i,t + b_3 * revision_i,t + controls + epsilon

Chan, Jegadeesh, and Lakonishok reported that both b_1 and b_2 (and b_3) were positive and statistically significant, with the joint portfolio return spread considerably larger than any single-signal version.

#How Blank Capital Research Uses This

Earnings momentum is a direct input to the 25% Momentum weight in our composite:

Within Momentum, we blend price momentum (6- and 12-month), nearness-to-52-week-high, residual momentum, and earnings momentum. Earnings momentum enters through two channels: standardized earnings surprises and the direction and magnitude of analyst revisions over the trailing one-to-three months. Combining these with price-based signals lifts the information ratio of the momentum sleeve because the signals correlate imperfectly — the diversification benefit is real.

Quality (30%) plays a complementary role. Companies with high, stable profitability are more likely to produce positive earnings surprises again; companies with low or volatile earnings quality are vulnerable to downward revisions. The composite therefore tends to favor firms where fundamental durability and short-horizon earnings momentum align.

#Practitioner Watch-Outs

• Data quality on earnings surprises. Consensus series, point-in-time adjustments, and treatment of pro forma versus GAAP numbers all matter. Poor point-in-time discipline introduces look-ahead bias that flatters backtests.
• PEAD implementation is costly. The best returns concentrate in small and mid-cap names where spreads are wider. Be skeptical of institutional capacity estimates.
• Revisions are stale in fast markets. Analyst consensus moves with a lag. When real-time news has already priced in a revision, the headline signal is already discounted.
• Crowding. Earnings momentum is a foundational quant signal and is heavily used. Expect modest alpha decay in the most liquid part of the market.
• Sample period sensitivity. The magnitude of the effect varies across decades. Keep expectations anchored to out-of-sample, not peak in-sample, performance.
• Interaction with quality. Earnings momentum works best when paired with a quality filter that removes firms with poor accrual quality and unsustainable earnings.

#See It in Action

• Momentum screener →
• Top momentum stocks →
• Momentum factor page →

#Further Reading

• Chan, L. K. C., Jegadeesh, N., and Lakonishok, J. (1996). "Momentum Strategies." Journal of Finance.
• Ball, R., and Brown, P. (1968). "An Empirical Evaluation of Accounting Income Numbers." Journal of Accounting Research.
• Bernard, V. L., and Thomas, J. K. (1989). "Post-Earnings-Announcement Drift: Delayed Price Response or Risk Premium?" Journal of Accounting Research.
• Jegadeesh, N., and Titman, S. (1993). "Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency." Journal of Finance.

#Related Factor Explainers

• Jegadeesh & Titman (1993) — Momentum
• Asness: Value-Momentum Everywhere
• Novy-Marx — Gross Profitability
• Carhart Four-Factor