Researchers from Indiana University’s Kelley School of Business and the University of Kentucky have revealed that CEOs might be giving away mental health challenges just by talking about their companies’ earnings to investors.
The study, published this month in the Journal of Accounting Research, utilized artificial intelligence to analyze chief executives’ speech recordings to identify traits of depression.
The universities debuted their measure of identifying the severity and prevalence of depression by training AI models to analyze the vocal patterns of CEOs. The analysis collected over 14,5000 earnings calls from S&P 500 companies from 2010 to 2021.
The study has shown that CEOs with depression face additional workplace challenges. Specifically, a CEO’s mental health challenges were associated with their company facing more significant risks, such as those that occur during litigation or volatile stock returns.
There was evidence that also suggested that CEOs with depression were shown to be more likely to have larger compensation packages and have a higher percentage of packages based on their performances. They were less likely to be older and women.
Nargess Golshan, an assistant professor of accounting at Indiana University and the study’s co-author, shared: “We want to really highlight mental health in leadership roles and how prevalent it is. Of course, it is important for the personal health of these executives, but also has far-reaching implications for the organization, the employees, the investors, and the broader economy.”
How AI Can Identify Depression
Researchers have utilized voice analysis as a tool for assessing chronic illnesses in the past, such as Alzheimer’s and Parkinson’s disease. The measure of depression through similar analysis is no different.
Rather than look at speech components previously used to assess some health conditions, such as pauses and filler words, which have been associated with depression, researchers are using AI to detect patterns too small for a human ear to notice.
“These machine learning models [are] more complicated than that,” Golshan shared. “They use numerical embeddings of pieces of the audio file that are not really perceptible by humans.”
Golshan collected data from a sample of non-CEOs who also took mental health assessments and cross-referenced the data with scores taken from reliable tools for determining depression, such as the Patient Health Questionnaire. She used this data to train her machine learning model, which could then identify small pieces of the CEO’s speeches that could indicate depression.
Among more than the 14,500 studies, over 9,500 CEOs were classified as having depression from the learning model.
Mental Health Stigmatized for CEOs
Mental health troubles and the effects they carry with them do not disappear outside the office. Three-quarters of the C-suite said they would consider quitting their jobs to seek a workplace that would better support their wellbeing. This was based on Deloitte’s 2023 Wellbeing at Work Survey, which polled employees across the United States, Canada, the UK, and Australia.
However, despite the prioritization of mental health among executives, discussing and seeking help for improving their mental health and wellbeing has been dampened by stigma about those struggles. Eight in 10 CEOs and 67% of employees believe that someone who has a mental illness is weak or burdensome, based on BusinesSolver’s 2024 State of Workplace Empathy survey, which polled 20,000 employees.
These types of mental health challenges can result in changes for a business. Blake Mycoskie, the founder of Toms, a slip-on shoe brand with a philanthropic bent, had sold 50% of the company to Bain Capital in 2014, citing loneliness and depression for the move. “I lost a lot of my clear meaning and purpose,” Mycoskie said.
