Investors have proven to be sophisticated enough to dismiss implausible explanations from companies of their quarterly earnings results, according to a new study.

For example, a utility company might attribute their lower earnings to “warm weather and higher propane products costs,” while an insurance company might explain a good quarter by touting its “continued efforts on cost containment and operational efficiencies.” Self-serving attributions such as these, which typically blame outside factors for negative developments and claim that their own internal initiatives led to positive results, are a traditional part of corporate earnings reports and press releases. But that doesn’t mean investors generally believe them.

According to a new study in The Accounting Review, a journal of the American Accounting Association, market response to self-serving attributions depends in large part on two key tests of plausibility—how badly the company’s industry peers are doing and what the study calls “commonality,” the extent to which market or industry forces drive a company’s earnings.

The difference proved to be dramatic when those two key tests were applied to the 94 companies in the study, which was conducted by Michael D. Kimbrough of the University of Maryland and Isabel Yanyan Wang of Michigan State University. Firms with average positive earnings surprises who made the highest-plausibility attributions had three-day above-market returns of 4.77 percent on average, whereas those that offered the lowest-plausibility reasons actually averaged a slight decline of 0.79 percent. Meanwhile, among firms with average negative surprises, those with the lowest-plausibility attributions sustained average declines of 5.11 percent, while those with the highest-probability excuses had declines of only 1.42 percent.

“Firms which provide defensive attributions to explain earnings disappointments experience less severe market penalties when 1) more of the their industry peers also release bad news, and 2) their earnings share higher commonality with industry- and market-level earnings,” said the paper. “On the other hand, firms that provide enhancing attributions to explain good earnings news reap greater market rewards when 1) more of their industry peers release bad news, and 2) their earnings shares lower commonality with industry- and market-level earnings.”

“Collectively, our results suggest that investors neither completely ignore seemingly self-serving attributions nor accept them at face value, but use industry- and firm-specific information to assess their plausibility,” the professors added. “Further analyses reveal that investors’ use of industry peer performance and earnings commonality information appears justified because investors’ perceptions are consistent with the association between the plausibility measures and the ex post actual persistence of earnings surprises.”

In sum, “investors are somewhat sophisticated when interpreting these narrative disclosures,” Kimbrough and Wang wrote.

“Our findings ought to be of value to both investors and corporate leaders,” said Wang in a statement. “Hopefully it will disabuse those executives who are counting on the naiveté of investors to let them get away with empty words or phony excuses in their public communications. For investors, it provides standards they will need to meet to keep up with the investment community at large.”

“The tools needed to apply those standards are certainly available to institutional investors, even though determining commonality is probably beyond the reach of individual stock-pickers,” said Kimbrough. “Still, even they should have the means to stack up the claims of a given company against its industry peers, which can go a long way in assessing the plausibility of the firm's performance narrative.”

The study's findings are based on an analysis of press releases and earnings reports of 94 randomly chosen firms, a roughly equal mix of small, medium and large, over a seven-year period. Sufficient data was obtained for a total of 1,790 firm quarters, 1,023 of which featured self-serving attributions and 767 of which did not. The self-serving classification was assigned to quarters when companies attributed their success in meeting or beating consensus forecasts to internal factors, such as management strategies or introduction of new products, or blamed a negative earnings surprise on external factors, such as bad weather or rising costs or regulatory actions. Firm-years in the self-serving category featured at least one such attribution and an average of three to four in a given earnings press release.

The authors found a significant relationship between the plausibility of self-serving attributions, as determined by industry performance and commonality, and the market-adjusted cumulative return of firms' stocks in the three days centered on earnings announcements. In reaching that conclusion, they controlled for an array of factors likely to affect the market’s response to earnings announcements, including the size of companies, the volatility of their stock, and their book-to-market ratio.

What kind of companies are likely to issue suspect attributions? Preliminary evidence suggests, in the words of the study, “Firms which provide less plausible attributions are larger and have higher likelihood of insider trading around earnings announcements, higher analyst following, higher institutional ownership, higher return volatility, and lower book-to-market ratio. These findings imply that managers with insider trading incentives and those facing greater capital market scrutiny are more likely to offer seemingly self-serving attributions even if they lack plausibility, consistent with the ‘opportunistic behavior’ view of capital markets.”

This view, according to the paper, finds “that capital-market scrutiny combined with the linking of manager compensation with stock prices creates pressure for managers to prop up prices by biasing financial reporting. To the extent capital-market pressure is greater for firms with higher analyst following and/or institutional ownership, the ‘opportunistic behavior’ argument suggests that greater analyst following and/or institutional ownership may increase managers’ tendency to provide implausible attributions to either mitigate market reactions to negative earnings surprises or to increase market rewards to positive surprises.”

Still, given the hazards of implausible attributions, as revealed by the new study, why would managers make them? It’s a matter of what they believe, Wang and Kimbrough wrote. “If managers believe there is a chance that investors might be persuaded by their implausible seemingly self-serving attributions, they are more likely to offer them even if ex post it turns out that investors can see through them.”

The study, “Are Seemingly Self-Serving Attributions in Earnings Press Releases Plausible? Empirical Evidence,” appears in the March/April issue of The Accounting Review, published six times a year by the American Accounting Association.