Taking a Red Pencil to Corporate Speak

An accounting professor separates fresh insights from hot air

by Mark Ray

photos by Emily Paine

efore becoming a professor at Bucknell’s Freeman College of Management, Kate Suslava worked as an auditor for accounting giant Ernst & Young (EY). It was a job that involved listening to plenty of companies’ quarterly earnings calls. Like her EY colleagues, Suslava paid careful attention to the numbers companies reported — things like revenue, expenses and the ever-popular earnings before interest, taxes, depreciation and amortization (EBITDA).

But she also paid attention to the words chief executive officers and chief financial officers (CFOs) spoke, especially euphemisms like “headwinds,” “rightsize” and “transition period.”

“I noticed that any time things were not going great, managers would switch to this type of language,” she says. “When things were going well, there was none of that.”

That observation fueled her doctoral dissertation, which involved teaching computers to parse the text of corporate disclosures for those euphemisms (a possibility that advances in natural language processing had just begun to make feasible). Her continued curiosity led to several recent, groundbreaking studies she authored or co-authored, most of which focused on how corporate executives speak during earnings calls. What she found was the clear effect those executives’ words have on analyst perception and stock performance.

In one study, published in Management Science, Suslava analyzed the transcripts of more than 78,000 earnings calls held between 2002 and 2016. She looked for euphemisms like those she once heard at EY and that appear in books like Oxford’s Dictionary of Euphemisms.

Not surprisingly, the terms from the glossary she compiled were plentiful, cropping up in 70% of the transcripts she studied. Usage increased during the global financial crisis of 2008 and when companies had disappointing news to report, like negative earnings surprises.

Of course, investors and analysts are more interested in numbers than words, so Suslava compared the use of euphemisms with companies’ stock prices. “Euphemisms are negatively associated with the stock returns on the day when the conference call happens,” she says. “If there is a conference call and an executive uses a lot of euphemisms, the stock will react negatively. So, investors take this as a sign of bad news.”

But that’s not the whole story.

Suslava found that the stock of a euphemism- happy company will continue to drop over the next three months. Although analysts know a euphemism when they see one, interpreting the bad news these words are disguising takes awhile. “Numbers are numbers; you can plug them into a model and quickly see what happens,” Suslava says. “But when it comes to words, it takes investors and analysts a longer time to decipher what they really meant.” (This also may explain why people are slower to react on days when lots of companies report earnings.)

In a second study, published in The Journal of Financial Data Science, Suslava and two colleagues took a similar approach with clichés — those trite, often wordy phrases like “quite frankly,” “forward momentum” and “at this point” that are often devoid of any real meaning. Unlike euphemisms, clichés aren’t necessarily associated with bad news, but they can still raise questions. “When somebody says ‘to be honest,’ it’s strange,” she says. “Was this person dishonest before that?”

As with the euphemisms study, Suslava and her colleagues found that the use of clichés had a measurable impact on a company’s stock price, probably because investors view clichés as a negative signal. They even constructed a stock portfolio that included some companies whose executives had used no clichés on their most recent earnings call and some whose executives had used at least four clichés. Over 142 months, the portfolio of no-cliché companies proved to be more profitable than the cliché crowd. An investor holding long positions in the first set of companies and short positions in the second set would have enjoyed an average monthly return of 2.05%. (A long position means buying and holding a stock; a short position means betting that the stock will lose value.)

“If you invest in companies that have more clichés, it’s more likely they will deliver more negative returns,” Suslava says, although the impact only lasts a few days after an earnings call. “If you have powerful software to count how many clichés and run this analysis right away, then you could make money off this strategy.”

Suslava’s research on euphemisms and clichés looked at words used by both male and female executives on earnings calls. Her latest research, which is pending publication, is different.

In the new study, she and two colleagues considered gender differences in the language used by CFOs. At publicly traded companies, about 10% of CFOs are women. “My gut feeling was that I wouldn’t find any difference because by the time a man or a woman gets to the CFO position, they have been polished so much by trainings and coaching and their experience that they probably speak the same,” she says.

Surprisingly for Suslava, that’s not at all what the researchers saw. When they crunched the numbers, they found significant differences across five verbal characteristics. On the whole, women were more concise and less optimistic in their presentations. They used less complex language, fewer euphemisms and clichés and more hard numbers. In short, they used what the researchers termed “careful verbal behavior.” “I didn’t expect any of that,” Suslava says.

The researchers also found that analysts appreciate the way female CFOs (and male CFOs who use careful verbal behavior) communicate. “The stock price tends to react positively to women’s verbal patterns,” Suslava says.

Although Suslava’s research is both intriguing and important, she’s quick to point out that it would be hard to get rich on the stock market by only applying what she’s learned — and she’s certainly not quitting her job to become a day trader. The fundamentals of investing still matter, of course, as do factors like the tenure of corporate board members, another topic she has studied. (The sweet spot, by the way, seems to be an average of around nine years per board member. At that point, board members have enough corporate knowledge to monitor and advise management effectively but haven’t yet become complacent.)

What’s more, corporations are well aware of the research people like Suslava are doing, and some are adjusting the language they use to compensate. “You need to always keep following what corporations are doing next because they also read our research and react to it,” she says.

None of the letters behind Suslava’s name — MBA, CPA, CFA, Ph.D. — have much to do with words. But her interest in the language of business makes sense when you know her background.

A native of Belarus, she holds bachelor’s and master’s degrees in linguistics from Minsk State Linguistic University, where she learned how to teach people new languages. After moving to the United States two decades ago, she took a job at a local community college teaching English as a second language. She eventually moved on to accounting so she could both pay the bills and have health insurance. “I quickly realized that accounting is also a language,” she says. “It’s just a language of business, but it’s kind of similar to learning a foreign language.”

After crunching numbers at EY for several years, she discovered what she was missing from her former life. “I learned a lot during my years in the corporate world, but I never forgot about my passion for teaching. I missed that a lot,” she says. So she enrolled in a doctoral program at Rutgers University, learned about textual analysis technology through her collaboration with an asset management firm, and dedicated her dissertation to the study of “corporate jargon.”