FoxAlgo → Research → Academic Audit → 12 Month Cycle in Cross-Section of Stocks Returns

Academic audit

Failedequity XS seasonality

12 Month Cycle in Cross-Section of Stocks Returns

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
P10
outranked ~20 of 200 baskets
Gate 3
Cost-aware net
RF -0.72
net-negative after costs
Failed
Worst 12-month leg (RF)-0.94
−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).

The paper documents that a stock's return in a given calendar month tends to echo its return in the same month of prior years, so stocks are ranked by their historical same-month performance and sorted long/short. We tested that annual return-seasonality ranking as a cross-sectional stock factor.

What we found

On the survivorship-free panel, ranking stocks by their annual return-seasonality was net-negative after realistic costs. It landed at the 10th percentile of random baskets, meaning it underperformed most random signals rather than beating them. The risk-adjusted RF was negative and the worst year was deeply negative, so we found no tradeable cross-sectional edge here.

How we tested it
2005–2026 test windowmodelled liquidity-aware costssurvivorship free
  • Tested on a survivorship-free 1077-name US common-stock panel, 2005-2026. Realistic modelled costs.
  • Placebo / robustness test: real result vs random baskets or shuffled signals (real vs the 95th percentile of random)
Source: Heston & Sadka (2008), "Seasonality in the cross-section of stock returns", J. Financial Economics
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.