FoxAlgo → Research → Academic Audit → Post-Earnings Announcement Effect

Academic audit

Failedequity event / PEAD

Post-Earnings Announcement Effect

The three-gate gauntlet · genuine only if it clears all three and survives adversarial refutation
Gate 1
Survivorship-free
free
clean panel
Eliminated here
Gate 2
Placebo ≥ P95
P93.6
outranked ~187 of 200 baskets
Gate 3
Cost-aware net
RF -0.34
net-negative after costs
Failed
Worst 12-month leg (RF)-0.84
−1.00 floor0
Every strategy here — winners included — loses in its worst 12 months. Depth is honest context, not the verdict.
Rejected at the luck gate — its net ranked no better than random baskets (below the P95 skill line).

Post-earnings-announcement drift (PEAD) is the documented tendency for stocks to keep drifting in the direction of an earnings surprise for weeks after the report. This test ranks stocks cross-sectionally on that surprise and holds a long/short book.

What we found

The signal has genuine cross-sectional ranking skill: it places at the 93.6th percentile against random baskets on a survivorship-free panel. But the drift concentrates in small-capitalization names, and once realistic liquidity-aware costs are applied the long/short book turns net-negative (RF -0.34, worst-year RF -0.84). The effect that exists on paper is not tradeable at liquid average-daily-volume scale, so we reject it.

How we tested it
2005–2026 test windowmodelled liquidity-aware costssurvivorship free
  • Data: survivorship-free 1077-name US common-stock panel, 2005-2026 (delisted names retained). Realistic modelled costs.
  • Placebo / robustness test: real result vs random baskets or shuffled signals (real vs the 95th percentile of random)
Source: Bernard & Thomas (1989), "Post-Earnings-Announcement Drift", J. Accounting Research
Read the paper ↗
← The Academic Audit — all 54 studies

Research, not investment advice. “Validated” factor-legs are market-neutral diversifying building blocks with a losing worst year — none is a standalone tradeable strategy. Metrics are cost-aware and modelled (not live fills); the 2005–2026 test window is out-of-sample versus the source paper. Dollar figures are not returns and are omitted by design.