// STUDY RESULT · 1998-01 TO 2026-09

Accruals Anomaly

DATA VALIDATED

Earnings driven by accruals persist less than earnings backed by cash, but prices act as if investors weight both the same, so high-accrual firms later disappoint.

How this run differs from the source: Accruals are net income minus operating cash flow, not Sloan's balance-sheet accruals; long only, 30 names.

	THE STRATEGY	$SPXTR BUY & HOLD
CAGR	17.3%	9.3%
Sharpe ratio	0.82	0.66
Max drawdown	-53.0%	-50.9%

// GROWTH OF $1, LOG SCALE

AMBER: THE STRATEGY · GREY: $SPXTR BUY & HOLD

Out-of-sample (from the split): CAGR 14.5% against 14.0% for the benchmark; Sharpe 0.89 in-sample, 0.72 out-of-sample.

After 25 bps of trading costs: 16.1% a year. EDGE OVER RANDOM

// DECADE BY DECADE

ERA	STRATEGY	BENCHMARK
1990s	32.7%	20.9%
2000s	17.2%	-0.8%
2010s	15.5%	14.0%
2020s	15.7%	15.2%

// YEAR BY YEAR, LOSING YEARS IN RED

1998*37.8%27.2%

199934.4%21.0%

200027.7%-9.1%

200116.8%-11.9%

2002-2.1%-22.1%

200376.7%28.7%

200421.6%10.9%

200524.6%4.9%

200620.0%15.8%

20078.2%5.5%

2008-38.7%-37.0%

200955.2%26.5%

201033.9%15.1%

2011-3.0%2.1%

201219.4%16.0%

201350.2%32.4%

2014-3.3%13.7%

2015-6.4%1.4%

201625.2%12.0%

201744.6%21.8%

2018-5.7%-4.4%

201919.7%31.5%

202015.1%18.4%

202129.4%28.7%

2022-20.1%-18.1%

202319.9%26.3%

202417.2%25.0%

202536.7%17.9%

2026*10.6%11.8%

Losing years: 2002 (-2.1%), 2008 (-38.7%), 2011 (-3.0%), 2014 (-3.3%), 2015 (-6.4%), 2018 (-5.7%), 2022 (-20.1%). Each cell: the strategy, then the benchmark below it. * a part year.

// THE IDEA, AS PUBLISHED

Sloan (1996), The Accounting Review · 1996

Accruals = change in non-cash current assets - change in current liabilities (ex short-term debt and taxes payable) - depreciation, over average assets
Rank annually into deciles, four months after fiscal year end
Long lowest-accrual decile, short highest

RUN IT YOURSELF ▸ IN THE LIBRARY

A real computation on the QuantGPT warehouse, walk-forward. A study across the whole market includes only part of the companies that later delisted from 1998 (survivors only before); a study on your own list of tickers is not affected. Gross returns, no costs except where stated. Past performance is not a promise. Findings, not advice. COMPUTED 2026-11-09 21:38 ON THE QUANTGPT WAREHOUSE. COMPANIES THAT LATER DELISTED ARE ONLY PARTLY IN THIS UNIVERSE (A FIX IS IN PROGRESS), SO THE RESULT LEANS TOWARD SURVIVORS. FUNDAMENTALS BY FILING DATE. NOT INVESTMENT ADVICE. PRIVACY