If you want to find out from executives if their companies have manipulated earnings or committed accounting fraud, the answer may depend on how you ask the question.
Understanding the prevalence of unethical or illegal financial reporting practices that mislead the public about company performance has always been a topic of considerable interest. The challenge is that, unless someone like a regulator successfully catches the misconduct and reveals it to the public, the only people who know that earnings were manipulated are the executives who actually did it – and got away with it. This creates a “tip of the iceberg” problem: How many manipulations are occurring under the surface that we can’t see?
Of course, any executive who gets away with manipulating earnings, especially in a fraudulent way, is probably going to lie if you ask them about it directly, even when responding to an anonymous survey. That’s why a team of researchers got creative in how they asked the question.
In a new study titled “Measuring the Prevalence of Earnings Manipulations: A Novel Approach” researchers anonymously surveyed and performed a “list” experiment with 971 executives from publicly traded U.S. companies. The executives participating in the study held a variety of positions, including CFO, Chief Accounting Officer, Controller, and CEO. The study is forthcoming in the Journal Accounting Research and is authored by Nicole Cade and Joshua Gunn, both from the University of Pittsburgh, and Alex Vandenberg from the University of Illinois Urbana-Champaign.
“The innovation of the list experiment is that it allows individual executives to answer truthfully about their own misconduct while still maintaining a high degree of plausible deniability. In fact, plausibility deniability is so high that it is impossible for anyone, including us, to know whether any individual executive admitted to misconduct. But as researchers, we can combine all of the responses to estimate the percentage of executives who admitted to manipulating their financial reports,” says Vandenberg.
When anonymously surveyed about whether their company had manipulated earnings, 27% of executives admitted that their company had manipulated earnings at least once in the past five years. Of these executives, 18% changed an operational activity to meet a near-term earnings target but sacrificed long-term shareholder value to do so, 9% intentionally obfuscated unfavorable information in their earnings reports, 7% exploited the discretion allowed by current accounting standards to report earnings that misrepresented the company’s actual performance, and 4% intentionally withheld material information from investors. Not surprisingly, when surveyed directly, not one executive admitted to committing accounting fraud.
Then a separate group of executives participated in the list experiment. For the less glaring forms of earnings manipulation, the executives in the experiment reported approximately the same rates of behavior as in the survey. However, for the two more egregious forms, committing accounting fraud and changing an operational activity to meet a near-term earnings target at the expense of long-term shareholder value, the executives in the experiment reported significantly higher rates for committing fraud (12%) and changing an operational activity (30%) than the executives in the anonymous survey.
Vandenberg concludes, “Our evidence suggests that earnings manipulations are not rare and that the number of financial statement frauds being committed is potentially much higher than the number being uncovered. These findings underscore the need for investors to maintain a skeptical eye towards numbers reported by company management because even audited financial statements might contain manipulations that have yet to be detected. Our findings also highlight the importance of parties like auditors and regulators who are tasked with detecting and deterring this type of behavior.”