Non-GAAP earnings numbers are better at predicting a company’s future earnings potential, but they are more subject to opportunistic manipulation and insider trading, a new study finds.

Sarah McVay

found that corporate governance and the Sarbanes-Oxley Act’s Regulation G help guard against such abuses, but investors still need to beware of non-GAAP numbers.

The group discount site Groupon was recently in the news for using a non-GAAP measure it called “adjusted consolidated operating income,” or adjusted CSOI, in the prospectus for its upcoming IPO until the SEC began asking questions and the company amended its offering documents (see Groupon Adjusts Controversial Accounting Measures).

The study found that unaudited non-GAAP figures are often better at forecasting a company’s future income than GAAP numbers, but often lend themselves to manipulation without oversight by independent board members.

In their paper, “Non-GAAP Earnings and Board Independence," Dr. Sarah McVay, an associate professor of accounting at the University of Utah’s David Eccles School of Business, and co-authors Richard Frankel, an Olin School of Business professor at Washington University in St. Louis, and Mark Soliman, an associate professor of accounting at the University of Washington’s Foster School of Business, concluded that independent boards can keep this opportunism in check.

Since it was enacted in 2002, Sarbanes-Oxley’s Regulation G in particular has done much to discourage such practices. However, McVay and her fellow researchers found that the regulation’s effectiveness relies on managers’ incentives, including ¬whether they have a direct stake in the information that is either released or withheld from a non-GAAP report.

“Regulation G requires a lot more exposure [of financial data],” said McVay. “It gets us a lot of the way to where we need to be for accuracy, requiring reconciliation between non-GAAP and GAAP numbers. There’s still some opportunism after Regulation G, but not as much.”

A further step would be to require non-GAAP figures to be audited to safeguard the accuracy of these reports.  

Making disclosures mandatory for a wider range of financial data, along with an emphasis on board independence and continued regulatory and investor scrutiny, would further eliminate the potential for non-GAAP inaccuracies, McVay noted.

The study found that companies with less independent boards are more likely to opportunistically exclude recurring items from non-GAAP earnings. Exclusions from non-GAAP earnings appeared to have a greater association with future GAAP earnings and operating earnings when boards had proportionally fewer independent directors. The results suggested that board independence was positively associated with the quality of non-GAAP earnings.

McVay and her fellow researchers plan to take a closer look at the consistency, or the lack of it, in non-GAAP reporting of so-called “transitory charges” in their next study. If a company excludes the charges that are deemed to be income-decreasing, they should also omit that data when it is income-increasing, she said. Preliminary analyses suggest this isn’t always the case, however.

McVay’s paper, published in the “Review of Accounting Studies,” can be viewed in its entirety at the SpringerLink Web site.