Seminal ‘Diversity = Profits’ Research Doesn’t Fare Well Under the Microscope

At last year’s World Economic Forum in Davos, Switzerland, the head of Goldman Sachs announced a new policy for the richest investment bank in the world: It would refuse to underwrite the stock offerings of any private company that did not have at least one woman on its board of directors — and that the minimum would move up to two in 2022.

“This decision is rooted first and foremost in our conviction that companies with diverse leadership perform better,” the Goldman CEO, David Solomon, declared.  {snip}

Solomon gave no source for this assertion, and Goldman did not reply to a request for comment. But much of the authority for claims like his rest on three studies done between 2015 and 2020 by the consulting company McKinsey, which were trumpeted as proof that large companies can boost their profits significantly by adding women and people of color to their boards of directors. “Companies in the top quartile for gender diversity on executive teams were 25 percent more likely to have above average profitability than companies in the fourth quartile,” the 2020 report says, while those in the top quartile for ethnic diversity are 36 percent more profitable than those in the bottom quartile.

“What our data show is that companies that have more diverse leadership teams are more successful,” the 2020 report concludes, recalling the two-word phrase in the title of an earlier report: “Diversity Wins!”

That line has been given credibility in major media outlets. “Diversity isn’t just a feel-good measure; it has bottom line benefits,” the Wall Street Journal reported in a 2018 piece on the McKinsey study. That conviction has inspired policies at large companies such as Goldman, and new state laws, including one in California that requires corporations headquartered in the state to have at least one director “from an underrepresented group” by the end of 2021, two by 2022 – and three such directors if the board consists of more than nine people.

Even as the McKinsey’s conclusions have become conventional wisdom in America’s power centers, there has been little outside scrutiny of its claim that companies enjoy bottom line benefits when they replace white men in leadership positions with women and people of color. But a new academic paper – the first detailed, independent analysis anybody has made of McKinsey’s findings and methods – concludes that while there is nothing wrong with diversity per se, McKinsey provides no evidence that it “wins.”

“Our results suggest that despite the imprimatur often given to McKinsey’s (2015, 2018, 2020) studies, caution is warranted to support the view that U.S. publicly traded firms can deliver improved financial performance if they increase the racial/ethnic diversity of their executives,” the authors report in a summary statement of their findings.

The study, carried out by Jeremiah Green, an associate professor of accounting at Texas A&M University, and John R.M. Hand, distinguished professor of accounting at the University of North Carolina, Chapel Hill, also suggests that McKinsey has misleadingly characterized its own findings to make the case for diversity.

“McKinsey’s studies are a little strange,” Hand said in a Zoom interview, “because they’re not structured to detect any causal evidence [that diversity generates profits]. They’re silent with regard to the fundamental causal question.”

The McKinsey study, written by four of the company’s consultants in its London, Chicago and Atlanta offices — gives each firm that it studied a “diversity score,” based largely on pictures and descriptions of executives on that company’s website and reports. It then measures each company’s economic performance. But as McKinsey acknowledges in the methodology section of its 2020 report, the financial data it collected on companies came from the period 2014 to 2018, while the data on diversity was compiled from Dec. 2018 to November 2019 — showing that the diversity data comes after or at the same time as the financial data, not before.

In other words, McKinsey’s own data shows, if anything, the likelihood that profit leads to more diversity, not the other way around.  {snip}

McKinsey acknowledges this possibility in its 2020 report.  “We are not asserting a causal link,” the methodology section states.  “It is theoretically possible that the better financial outperformance enables companies to achieve greater levels of diversity.”

Nevertheless, as Green and Hand point out, the company’s public interpretations of their results seem to set aside this crucial problem. They quote Vivian Hunt, McKinsey’s managing partner in the United Kingdom and one of the study’s four authors, saying: “What our data shows is that companies that have more diverse leadership teams are more successful.” Companies are implementing diversity, she continued, “because it’s a business imperative and driving real business results.”

{snip}

Green and Hand find other problems with the McKinsey report, including its assumption that the single factor of diversity could lead to the remarkable outcomes its study proclaims – a 36% improvement in the likelihood of having financial returns above the national industry medium for companies with the highest diversity indices. That’s an astonishing improvement to attribute to a single factor. Could having a few underrepresented minorities among a company’s executives really be solely responsible for such an immense difference?

“The output of high firm financial performance causally depends on a vast number of inputs, not just racial/ethnic diversity,” Hand said in an email. “This ‘vast number’ aspect is important,” he added, because, among other reasons, “it plausibly means that any one of the vast number of inputs is unlikely in and of itself to be a, or the, huge driver of financial performance.” Green and Hand also question the measure of diversity used by McKinsey to create what it called its “diversity score.” What they call the “key weakness” here is that the score given by McKinsey to any study “doesn’t benchmark against anything” – not against the actual representation of various ethnic groups in society, or what that representation was when the current generation of executives was graduating from college.

As a result, McKinsey’s score doesn’t say much about what diversity actually means in any given company. As Green and Hand point out, a company whose executives match the American population – ranging from 61% white down through to 1% Native American – would have the same diversity score as a company that had the reverse representation (61% Native American down through to 1% white).

{snip}

There are other reasons for skepticism about the McKinsey conclusions and the use put to them by those claiming the diversity-profit connection.  Among them is a comparison with ethnically homogeneous countries whose companies, despite their lack of diversity, have been successful and highly competitive on the international business scene.

{snip}