The Economic Consequences of CEO Integrity
This article was originally published for Forbes on April 23, 2020, by Joseph Brazel, professor of accounting.
Lower chief executive officer integrity is associated with higher audit fees at large public corporations and deteriorating performance, according to a recent academic study published in The Accounting Review.
Researchers poured through over 30,000 shareholder letters, scrutinizing the words used by CEOs, according to the study, titled “CEO Behavioral Integrity, Auditor Responses, and Firm Outcomes,” from the University of Virginia, INSEAD, Duke University and Yale University.
The research team focused on language to measure CEO behavioral integrity or whether the executive “walks the talk” by the extent to which the CEO’s deeds and words failed to align and required explanation.
“Global surveys of CEOs suggest integrity is one of the most important leadership attributes, behind only creativity,” says William Mayew, a Duke University accounting professor and an author of the study. “Despite this importance, we know virtually nothing about how CEO integrity impacts firm outcomes in large samples. We wanted to provide some evidence to fill this void. We did so in the context of auditing because financial statement auditors are supposed to consider the integrity of management as part of their procedures.”
Key to the study was developing and validating a measure of CEO integrity without having access to surveys or interviews of corporate employees. “Using computational linguistics, we measured how many causation words were used in each shareholder letter. Causation words are words like ‘because’ and ‘therefore’ and are used commonly when providing explanations,” says Mayew.
Of course, CEOs have lots of other reasons beyond integrity to provide explanations. To address this concern, the study employed statistical techniques to account for other factors such as company performance, uncertainty and complexity. Importantly, the study demonstrated its integrity measure was CEO-specific by illustrating how it differed from the words used in the section of an annual report where management addresses the company’s performance as well as terms used by the CFO in earnings calls.
Next, the study validated their measure of CEO integrity. “First, we obtained a small sample of survey responses on CEO integrity and found our measure was highly correlated with it. Second, we obtained a large sample of employee rankings from Glassdoor.com,” Mayew reports. “We find that, when our measure suggests a CEO has high behavioral integrity, employees rate the CEO higher and have a more favorable view of the work environment.”
With the integrity measure developed and validated, the study then proceeded to determine the economic outcomes of CEO integrity. Their first stop was the examination of audit fees, with the thought being that auditors are charged with assessing management integrity and incorporating such perceptions into their testing to determine if the financial statements are materially misstated. In line with this train of thought, the study found that audit fees were higher when CEO integrity was lower, suggesting that auditors expand their financial statement testing and charge higher fees to mitigate integrity concerns.
The study then asks: Were the higher fees put to good use? The answer appears to be yes, as the researchers find no relationships between CEO integrity and financial statement restatements or a measure of fraud risk.
But what about areas where auditors might focus less? The study examines option backdating and observes that “low behavioral integrity is associated with a higher likelihood of option backdating. This is consistent with low integrity leading to bad behavior for outcomes that the auditor is not overseeing,” Mayew says. He notes that a big take-away from the study is that “CEO integrity could lead to some bad outcomes in areas that are not properly monitored by auditors or the board.”
On the bright side, the study does find that higher CEO integrity is associated with stronger future performance and notes that more research is needed to determine the extent which companies or boards value integrity and other attributes when making CEO hires. Mayew does stress that as a caveat, their measure of integrity is “at a distance” and their findings are of association and do not demonstrate causality. He also feels that more research is needed in this area to draw more definitive conclusions about the topic of CEO integrity.
According to Mayew, future research is needed to address topics such as “Are there different aspects of integrity beyond behavioral integrity that matter? What are the trade-offs between integrity and other executive attributes? In the end, optimizing firm value requires broader thinking about all of the attributes that make up the CEO.”