ANN ARBOR—Contrary to what companies want investors to believe, costs excluded from “pro forma” earnings are far from one-time or unimportant expenses. In fact, higher levels of excluded expenses lead to lower future cash flows and can mislead investors about a firm’s profitability, say University of Michigan Business School researchers.

Pro forma earnings are an increasingly popular measure of earnings that excludes certain expenses a company deems non-recurring, non-cash or otherwise not important for understanding the future value of the firm. Examples include restructuring costs, depreciation, amortization, losses on the sale of assets and a variety of other miscellaneous charges.

In a new study of more than 120,000 quarterly observations of company earnings announcements and cash flows from 1988 to 1999, U-M Business School researchers Russell J. Lundholm, Jeffrey T. Doyle and Mark T. Soliman found that every dollar of exclusions per share in a quarter results in roughly .83 fewer dollars of future cash flow.

“Prior research of pro forma earnings can be interpreted as supporting the position that investors are not misled by firms’ use of such earnings,” says Lundholm, professor of accounting at the U-M Business School. “We come to a very different conclusion. The market does not appear to appreciate the future cash flow implications of the excluded expenses. Rather, the market appears to be systematically fooled by the firms’ use of pro forma earnings.”

Lundholm and colleagues examined companies’ cash flows and stock returns for up to three years after an earnings announcement. They found a statistically and economically significant link between future cash flows and expenses excluded in a company’s definition of pro forma earnings, as well as a significant difference between the stock returns of firms with high vs. low amounts of excluded expenses.

Their study classifies exclusions, or the difference between GAAP (Generally Accepted Accounting Principles) earnings—which account for all expenses—and pro forma earnings, into two parts. These include “special items” that are relatively easy to identify (e.g., restructuring charges) and “other exclusions,” which are more difficult to ascertain (e.g., stock compensation expenses, losses on equity-method investments, legal settlement costs and operating losses from soon-to-be closed facilities).

While a dollar of excluded special items, such as a restructuring charge, in a quarter predicts about one more dollar of future cash flows over three years, a dollar of other exclusions predicts roughly three fewer dollars of cash flow over the same period, the researchers say.

“To the extent that the special items represent some type of restructuring costs, it appears that future cash flows benefit from the activity,” Lundholm says. “However, if the excluded expense is what we have labeled ‘other exclusions,’ then these expenses are much more varied and appear to have a strong negative relation to future cash flows.

“This could be because they are expenses that actually recur, or it could be that firms that resort to excluding these items are scrambling to meet some earnings target, and such firms have predictably lower cash flows in the future.”

In addition to measuring the impact of pro forma earnings on companies’ future cash flows, the U-M study also explored the response of the stock market to the reporting of such earnings.

Although initially the market reacts positively to a pro forma earnings surprise during the three-day earnings announcement period, the reward is diminished if the surprise is achieved by the use of exclusions in the definition of pro forma earnings, the researchers say.

But, more importantly, they add, in the three years after the earnings announcement, stock returns are up to 45 percent lower for firms with relatively large exclusions, compared with those firms with small exclusions.

“Overall, our results show that the current regulatory concern about the use of pro forma earnings may be warranted,” Lundholm says. “The expenses that are excluded in a firm’s definition of pro forma earnings predict generally lower future cash flows and are negatively related to future stock returns.

“While the stock price reaction to the earnings announcement is decreasing in the amount of excluded expense, the ‘penalty’ is not nearly sufficient. A trading strategy that goes long in low-exclusion firms and short in high-exclusion firms yields a large, positive abnormal return in the three years following the announcement.”

Business SchoolRussell J. LundholmGenerally Accepted Accounting Principles