The development of a cash-based operating profitability measure that excludes accruals sheds light on the accrual anomaly. The authors find that this measure outperforms profitability metrics that include accruals, that a strategy including this measure can increase the strategy’s Sharpe ratio, and that the measure can explain expected returns as far as 10 years ahead.
What’s Inside?
Accruals are the noncash component of earnings. Prior research has shown that expected returns increase with profitability but decrease with accruals. The authors develop a cash-based operating profitability measure that excludes accruals. They find that this measure better explains the cross section of expected returns, that investors would be better off adding a cash-based operating profitability measure to their investment opportunity set rather than an accruals proxy and a profitability proxy, and that cash-based operating profitability explains expected returns as far as 10 years ahead.
How Is This Research Useful to Practitioners?
The authors’ results imply that investors can generate higher risk-adjusted returns when a proxy for cash-based operating profit is included in their portfolio. Specifically, the authors find that using a five-factor model that combines market, size, value, and momentum factors with a cash-based operating profitability factor yields a higher Sharpe ratio than does a six-factor model that combines the four factors with both an accrual and operating profitability factor.
In addition, if the cash-based operating profitability metric can predict returns as far as 10 years out, then a direct implication would be that investors could profit from the persistence of this factor. The authors suggest that this persistence could indicate an initial market underreaction to cash-flow information that is gradually corrected over a decade or that the cash-based operating profitability measure shares a common determinant with expected returns.
Moreover, the authors help explain the accrual anomaly, which is the finding of a negative relationship between accruals and the cross section of expected stock returns. When a cash-based profitability measure is included in their regression model, no relationship between accruals and the cross section of stock returns is found.
How Did the Authors Conduct This Research?
Beginning with CRSP and Compustat data and all firms (excluding financial firms) trading on the NYSE, AMEX, and NASDAQ from July 1963 to December 2014, the authors obtain a series of variables, including an operating profitability measure that is calculated as sales minus cost of goods sold minus selling, general, and administrative expenses. The authors then remove the accrual component from this operating profitability measure by subtracting changes in accounts receivable, inventory, prepaid expenses, deferred revenue, accounts payable, and accrued expenses. The result is their cash-based operating profitability measure.
To conduct the empirical tests, the authors use Fama–MacBeth regressions. (The Fama–MacBeth regression is a common statistical methodology that allows researchers to include a large number of control variables when examining interest.) They estimate the regressions using monthly data over the full sample period.
The authors create augmented models that add operating profitability, accruals, or cash-based operating profitability to the Fama–French three-factor model and examine whether the augmented models price accruals in the cross section of stock returns.
Abstractor’s Viewpoint
In finance, we view cash as king and tend to believe that accruals distort various financial metrics. Therefore, it is perplexing to find that the accrual anomaly somehow persists. This research offers an explanation of such findings—namely, analyses that yield such findings omit a critical variable.