The Accruals Anomaly

When Accounting Earnings Diverge From Cash Flows, Patterns Follow

You have seen the gap between reported profit and operating cash flows in financial statements. Researchers discovered that gap carries systematic information about how firms are priced relative to their future performance — and built three decades of empirical literature around explaining why.

Explore the evidence

How the Accruals Anomaly Became One of the Most Studied Effects in Empirical Finance

From Sloan's foundational discovery in 1996 through decades of replication, international testing, and ongoing debate about persistence in modern markets.

1996

Sloan Documents the Foundational Anomaly in The Accounting Review

Richard Sloan publishes the landmark study showing that investors appear to overprice the accrual component of earnings relative to the cash component, producing predictable return differentials between high-accrual and low-accrual firms in subsequent periods. The balance sheet decomposition method introduced in this paper becomes the standard measurement approach used in nearly all subsequent replication work.

1996
2002

Behavioural Explanations Gain Traction in Academic Literature

Hirshleifer, Hou, Teoh, and Zhang publish evidence that the anomaly concentrates among firms with lower analyst coverage and institutional attention, supporting the hypothesis that the effect reflects limited investor capacity to process publicly available accounting information rather than compensation for an unidentified risk factor. This framing becomes one of the dominant interpretive frameworks in subsequent research.

2002
2004

Dechow and Colleagues Refine Accrual Decomposition Methods

Dechow, Richardson, and Sloan publish a detailed disaggregation of the accruals construct, demonstrating that different components of total accruals carry different degrees of return-predictive information. This granular treatment advances both the measurement methodology and the theoretical understanding of why the earnings-cash gap matters for pricing.

2004
2007

International Evidence Tests the Anomaly Across Global Equity Markets

A wave of cross-country replication studies examines whether the accruals anomaly generalises beyond US data to markets operating under different accounting standards and institutional environments. Results are broadly supportive but reveal meaningful variation in signal magnitude across geographies, prompting researchers to examine which institutional features moderate the anomaly's strength.

2007
2012

Anomaly Decay Debate Emerges in Post-Publication Period Research

Studies examining returns in the years following Sloan's 1996 publication find evidence of a narrowing return spread between low-accrual and high-accrual firms, raising questions about whether increased awareness and arbitrage activity have partially corrected the mispricing. Researchers disagree about how much of the apparent attenuation reflects genuine decay versus structural changes in listed firm populations.

2012
2018

Machine Learning and Alternative Measurement Methods Enter the Literature

Quantitative researchers begin applying non-linear statistical methods to accrual components, finding that combinations of accounting variables outperform the simple aggregate accrual measure in predicting cross-sectional return variation. These papers reopen methodological questions about optimal accrual measurement and introduce new debates about overfitting, interpretability, and economic significance.

2018
2024

Algorithmic Markets and IFRS Effects Prompt New Empirical Questions

Recent research examines how the proliferation of algorithmic trading and the global adoption of IFRS accounting standards affects the accruals signal. Preliminary evidence suggests faster price adjustment in highly automated markets and meaningful differences in accrual measurement comparability across accounting regimes, adding new dimensions to an already rich empirical debate.

2024

The Mechanics Behind the Accruals Anomaly — What the Research Actually Shows

Before you can engage with the empirical debate about whether the accruals anomaly reflects mispricing or a missing risk factor, you need a precise understanding of what accruals are, how they are measured, and why the gap between earnings and cash flows carries any information at all about future return variation. The following three sections build that foundation in sequence, starting from accounting mechanics and moving through measurement methodology to the return prediction evidence itself.
balance sheet showing accrual components and working capital

What Accruals Are and Where They Come From

Accruals are the non-cash component of accounting earnings — the portion of reported profit that is not yet reflected in actual cash receipts or payments. They arise from the matching principle in accrual accounting, which requires revenues and expenses to be recognised in the period they are earned or incurred rather than when cash changes hands. Common sources include changes in accounts receivable, inventory, accounts payable, prepaid expenses, and deferred revenues, as well as non-cash charges like depreciation and amortisation. When accruals are large and positive, reported earnings substantially exceed the cash the business actually generated. When accruals are negative, cash generation exceeds reported profit. Sloan's foundational insight was that this gap is not random noise — it carries information about the reliability and persistence of the earnings figure itself, and markets appeared to price that information less efficiently than theory would predict.

quantitative researcher computing accrual measures from financial data

How Researchers Measure and Test the Anomaly

The standard approach to measuring accruals uses the balance sheet method: operating accruals are computed as the change in non-cash current assets minus the change in current liabilities excluding short-term debt, less depreciation and amortisation, scaled by average total assets. Researchers then rank all firms in their sample by this accrual measure at each annual measurement date and form portfolios based on accrual decile or quintile. The anomaly test examines whether the portfolio of low-accrual firms subsequently outperforms the portfolio of high-accrual firms on a risk-adjusted basis. Critical methodological choices include the filing date lag applied to avoid look-ahead bias, the treatment of financial firms whose accruals have different structural properties, and the risk factors used to adjust raw return differences. These choices produce meaningful variation in reported results across replication studies, which is why understanding the methodology is inseparable from evaluating the evidence.

cross-sectional return data showing accrual anomaly patterns

What the Return Evidence Demonstrates

The core finding is consistent across the original study and many replications: firms ranked in the lowest accrual decile have historically produced higher subsequent returns than firms ranked in the highest accrual decile, with the spread persisting after controlling for market, size, and value risk factors in most sample periods and geographies. The magnitude of the spread varies across studies, time periods, and measurement approaches, and there is credible evidence that the effect has attenuated in post-publication data in some markets. The interpretation of the spread remains contested: the mispricing hypothesis holds that high-accrual firms are overpriced because investors do not adequately discount the lower earnings persistence implied by large accrual components; the risk-based hypothesis argues that high-accrual firms share fundamental characteristics that justify lower expected returns on a risk-adjusted basis. Both interpretations have empirical support, and neither has been decisively eliminated by the data available to date. Past performance documented in the academic literature does not guarantee future results in any market context.

Side-by-Side Analysis

How High-Accrual Firms Differ From Low-Accrual Firms Across Key Dimensions

Understanding what distinguishes firms at opposite ends of the accrual distribution helps you connect the abstract anomaly concept to the concrete accounting and return characteristics that researchers actually observe in financial statement data across different market environments and time periods.
Learn more

High Accruals

SAR 19.00 Per measure

Firms where reported earnings substantially exceed operating cash flows

Earnings cash conversion Weak conversion
Earnings persistence Not present
Working capital build Present
Historical return spread Lower returns
Analyst forecast accuracy Not present
Subsequent earnings revision Often downward
Balance sheet accrual load Present
Anomaly signal strength Strong short signal

Neutral Accruals

SAR 59.00 Per measure
Firms where earnings and cash flows are broadly aligned over time
Earnings cash conversion Moderate alignment
Earnings persistence Present
Working capital build Not present
Historical return spread Neutral returns
Analyst forecast accuracy Present
Subsequent earnings revision Mixed direction
Balance sheet accrual load Not present
Anomaly signal strength No clear signal

Low Accruals

SAR 99.00 Per measure
Firms where operating cash flows equal or exceed reported earnings
Earnings cash conversion Strong conversion
Earnings persistence Present
Working capital build Not present
Historical return spread Higher returns
Analyst forecast accuracy Present
Subsequent earnings revision Often stable
Balance sheet accrual load Not present
Anomaly signal strength Strong long signal

Accruals Anomaly Terminology

A reference guide to the key terms and concepts you will encounter across the academic literature on accounting accruals and cross-sectional return prediction.

Accounting

Accounting Accruals

The non-cash component of reported accounting earnings, representing the difference between revenues and expenses recognised under the accrual basis of accounting and the cash actually received or paid in the same period. Accruals arise from the matching principle and include changes in working capital items as well as non-cash charges such as depreciation.
Accounting

Operating Accruals

The subset of total accruals arising from a firm's core operating activities, typically measured as the change in non-cash working capital minus depreciation and amortisation, scaled by average total assets. This is the primary measure used in the foundational accruals anomaly literature and most subsequent replication studies.

Accounting

Total Accruals

A broader measure of the difference between net income and operating cash flows, including both working capital changes and non-operating items. Some researchers prefer total accruals to operating accruals because it captures a wider range of earnings management activity, though the two measures produce correlated but not identical return predictions.

Measurement

Balance Sheet Approach

The original method for computing operating accruals introduced by Sloan in 1996, which derives the accrual estimate directly from changes in balance sheet items rather than from the cash flow statement. The balance sheet approach and the cash flow statement approach produce different numerical estimates of accruals due to differences in scope and accounting treatment.
Measurement

Cash Flow Approach

An alternative method for computing accruals that derives the estimate from the difference between net income and operating cash flows reported directly in the statement of cash flows. This approach became more practical after the widespread adoption of mandatory cash flow statement reporting and produces different accrual estimates from the balance sheet method.

Accounting

Earnings Quality

A broad construct referring to the degree to which reported accounting earnings accurately reflect the underlying economic performance of a firm and are likely to persist into future periods. High earnings quality is typically associated with a large cash component and a small accrual component, while low earnings quality is associated with earnings that are heavily reliant on non-cash items.
Finance

Cross-Sectional Anomaly

A pattern in which a firm characteristic observable from publicly available data predicts systematic differences in subsequent returns across firms at the same point in time, after controlling for known risk factors. The accruals anomaly is a cross-sectional anomaly because the return prediction is based on ranking firms relative to each other on accrual intensity rather than on time-series variation.
Finance

Anomaly Persistence

The degree to which a documented cross-sectional return pattern continues to produce statistically and economically significant return differences in data collected after the original study's sample period ended. The persistence of the accruals anomaly in post-publication data is one of the most actively debated questions in the recent empirical literature.
Quantitative

Factor Models

Statistical frameworks used to decompose asset returns into components attributable to exposure to systematic risk factors such as market return, firm size, and book-to-market ratio. Factor models are used in the accruals literature to test whether the return spread between low-accrual and high-accrual firms reflects compensation for risk exposures or a genuine pricing anomaly unexplained by known factors.

Finance

Alpha

The component of a portfolio's return that is not explained by its exposure to systematic risk factors in a given factor model. A statistically significant positive alpha for a portfolio of low-accrual firms relative to high-accrual firms is the standard empirical criterion used to claim that the accruals anomaly represents genuine mispricing rather than risk compensation.
Methodology

Look-Ahead Bias

A methodological error in which financial data is used in a backtest at a date earlier than it would have been publicly available to market participants, causing the test to overstate the anomaly's historical predictive power. In accruals research, look-ahead bias arises when fiscal year-end accounting data is used without applying a lag to account for the time between fiscal year-end and actual filing date.

Accounting

Earnings Persistence

The degree to which a given level of reported earnings is likely to continue in future periods. Academic research consistently finds that the cash component of earnings is more persistent than the accrual component, meaning that a dollar of cash earnings predicts more future earnings than a dollar of accrual earnings — a finding central to the accruals anomaly's theoretical foundation.

Theory

Mispricing Hypothesis

The interpretation of the accruals anomaly that attributes the return differential between low-accrual and high-accrual firms to systematic investor error — specifically, the failure to recognise that the accrual component of earnings is less persistent than the cash component, leading to overpricing of high-accrual firms and underpricing of low-accrual firms relative to their fundamental values.

Theory

Risk-Based Hypothesis

The competing interpretation of the accruals anomaly that attributes the return differential to rational compensation for an unidentified or imperfectly measured systematic risk factor, rather than to investor error or mispricing. Proponents argue that high-accrual firms share characteristics associated with elevated fundamental risk that justify the lower subsequent returns they have historically produced.