The Earnings Quality Indicator Used by ETFs

Look under the hood of exchange-traded funds (ETFs) incorporating a quality factor and you will often find accruals being used. An accrual is an accounting entry recorded at the time a transaction occurs, regardless of whether not payment has been made. Accrual accounting allows businesses to report their sales and earnings performance without being beholden to the timing of transaction-related cash flows.

The downside of accrual accounting is the risk of those transactions not being completed in the future. Customers may not pay their bills or may push to renegotiate contracts. Inventories may be sold at lower-than-anticipated prices or even written down. There are certainly other examples, but one of the risks with accruals is too much optimism on the part of management. This excessive optimism can lead to lower future stock returns.

This was the conclusion reached by Richard Sloan in “Do Stock Prices Fully Reflect Information in Accruals and Cash Flows about Future Earnings?” (The Accounting Review, July 1996). In this oft-cited analysis, Sloan found that stocks with the lowest levels of accruals experienced abnormal returns of 4.9%. Stocks with the highest levels of accruals experienced abnormal returns of -5.5% over the following 12 months after being included in a portfolio. (An abnormal return is a return that is different than would be expected.) Subsequent studies confirmed the existence of an accrual anomaly. ETFs using an earnings quality component consider accruals.

There are different ways for calculating accruals, but the math is straightforward. The only thing you need is financial statements for the two most recently completed fiscal years and either a spreadsheet or a calculator, pencil and paper. Stock Investor Pro users can also create custom fields in the program to automate the calculations.

Using “Δ” to imply the year-over-year change, Sloan listed accruals in his 1996 study as equaling:

(ΔCurrent Assets – ΔCash/Cash Equivalents) – (ΔCurrent Liabilities – ΔShort-Term Debt Included in Current Liabilities– ΔTaxes Payable) – Depreciation and Amortization Expense

The accruals calculated by the above equation should be scaled (meaning divided by) the average total assets for the most recently completed fiscal year and the year prior. Doing so allows an apples-to-apples comparison between companies of different sizes to be made. Depending on the formula, accruals may be scaled to net operating assets at the start of the fiscal year instead.

Low levels of accruals are good and high levels of accruals are bad. The problem with calculating the data for just one company is relative context. There is no measurement of all companies to compare it against. Rather, the math must be repeated for all companies, with results ranked from lowest (good) to highest (bad). Index providers who do this winsorize the data. Winsorization reduces the impact on statistical analysis that outliers with extraordinary high or low values have. This might not be a daunting task for those who have mathematical backgrounds and are comfortable with spreadsheets, but it’s certainly not something most others desire to get involved with. (Index providers and quantitatively-oriented investment firms rank stocks based on specified criteria to ensure that a sufficient number of stocks can always be found. Using absolute instead of ranked values has the advantage of finding more stocks when conditions are favorable and less when conditions are unfavorable.)

Those of you seeking a shortcut can similarly look for negative values. An alternative is to avoid stocks whose net income exceeds free cash flow (cash from operating activities less capital expenditures). Another option is to determine how the asset turnover ratio (sales/assets) has changed on a year-over-year basis. An increase in this ratio shows that a company is more efficiently turning its assets into sales; a decrease shows lower efficiency. Scott Richardson, in a 2001 paper co-authored by Sloan and two others, suggested using net operating assets (NOA) instead of reported total assets in the denominator of the assert turnover ratio. NOA is:

(Total Assets – Cash and Short-term Investments) – (Total Assets – Short-term Debt – Long-term Debt – Common and Preferred Equity – Minority Interest)