American investors are used to the idea of seeing publically traded companies reporting results on both generally-accepted accounting principles (GAAP) and on non-GAAP. The latter typically flatter the corporate performance, often excluding costs that managements claim shouldn't be considered normal operational costs.
Early stage or growing technology companies are among the most frequent users — and abusers — of non-GAAP EPS, since they heavily employ stock compensation, as well as because of their relatively high costs from asset impairments, research and development and foreign exchange costs, among other things. Typically, when a company puts forward non-GAAP earnings Barron's looks at GAAP numbers too to get a clearer picture of the fundamentals.
Yet the new tax law, passed in December, has sparked a fresh debate about what numbers to use. Indeed, investors should probably pay closer attention to non-GAAP numbers for the companies in the Dow Jones Industrial Average (.DJI), according to a report from FactSet.
There are mixed opinions in the market about the use of non-GAAP earnings per share, acknowledges John Butters, FactSet's senior earnings analyst. That said, in the fourth quarter of 2017, thanks mainly to the tax changes, the numbers generated from non-GAAP EPS for Dow companies were less volatile and more likely a better reflection of profit growth from day-to-day operations than GAAP, he says.
In the fourth quarter, 28 of the 30 companies in the Dow, or 93%, reported non-GAAP EPS in addition to GAAP EPS with a mean difference between the methods of 110%. In 2016, just 68% reported both measures with a difference of 13% between them.
Consequently, in this particular quarter, thanks to the exclusion of a one-time tax charge or gain, the non-GAAP EPS figures for Dow companies were less volatile and more likely a better reflection of earnings growth from day-to-day operations than GAAP EPS, Butters adds.
Some Dow component fourth-quarter results illustrate the unusual nature of the quarter. For example, McDonald's (MCD) fourth quarter GAAP EPS showed a 40% decline, but non-GAAP EPS a 19% increase. Similarly, Intel's (INTC) GAAP EPS was down 121% but non-GAAP up 37%.
Overall, 19 DJIA companies reported a net charge because of the tax law, while nine reported a net gain and these were typically the largest single item accounting for the unusually large differences between GAAP EPS and non-GAAP EPS for these companies for the quarter, he notes.
Moreover, Dow companies that didn't typically provide non-GAAP EPS numbers in the past did so in the quarter.
It's too simplistic to say ignore non-GAAP numbers, but Barron's has consistently written that investors should keep an eye on the difference between a company's GAAP and non-GAAP results. Supporters of non-GAAP argue that it provides the market with a more accurate picture of earnings from the day-to-day operations of companies, as items that companies deem to be one-time events or non-operating in nature are typically excluded.
But critics argue there is no industry-standard definition of non-GAAP EPS, and companies can take advantage to exclude items that (more often than not) have a negative impact on GAAP earnings.
It's still wise to always mind the GAAP.