The Trouble With Diversity Data

Research has proven that diversity leads to better long-term outcomes — but the ways to measure diversity are flawed, according to Willis Towers Watson.

Illustration by II

Illustration by II

Asset managers with diverse workforces and executives are more than just good political eye candy for trustees: Diverse teams have been proven to be much more likely to deliver better long-term results. But the data used to identify them can be incomplete and misleading, according to a new report from Willis Towers Watson.

The consulting firm argues that an organization whose employees represent a broad mix of people will have cognitive diversity stemming from different work, educational, neighborhood, and other social backgrounds. “Cognitive diversity is hard to measure, but it is heavily influenced by quantitative metrics of diversity that we can collect,” WTW said in the report.

This data, however, can have limited value. For example, the study cautions investors about simply relying on minority and female ownership statistics to create a more diverse roster of asset managers. “We believe an asset manager having minority and female ownership doesn’t necessarily mean diversity is reflected across different functions within the organization,” the report’s authors wrote.

Instead, the consulting firm stressed that investors need to evaluate who is actually on a leadership team or portfolio management group.

“The people that comprise the investment teams and leadership are the primary source of alpha generation and responsible for the organizational culture that can sustain competitive advantages,” WTW said.

There’s no shortage of data to look at, even if some is limited. Investors have increasing access to better information as data on diversity and inclusion, as well as environmental, social and governance factors has exploded. A growing industry that includes index providers, credit ratings agencies, and other third-party data firms has sprung up to collect and analyze the data and provide investors with better transparency on the companies in which they invest and the asset managers that invest on behalf of them. These firms have used advanced techniques, such as artificial intelligence and machine learning, to gather data from regulatory filings, web sites like GlassDoor, and other governmental organizations. Most recently, financial information service FactSet said it would acquire Truvalue Labs, which uses AI to collect ESG data.

[II Deep Dive: Manager Diversity Is Increasingly Important for University Endowments. How They Measure It Is Still Unclear.]

Still, WTW warns about the dangers of using single pieces of information, such as minority and female ownership. This data could be relevant as these decision-makers are theoretically in it for the long haul and their paychecks are aligned with client results.

“However, when we examine the data reported to a third-party database provider which comprises over 6,000 firms, only about a quarter of them provided ownership details, making screening by diverse ownership extremely limiting,” the report stated. “Of the quarter of those who voluntarily reported ownership information, the minority/female equity stakeholders is still fairly low and not representative of society.”

WTW criticized the asset management industry for being a long way off from true diversity, saying it’s missing out on a lot of talent.

But investors and asset managers may be motivated by the consultant’s preliminary findings on returns after analyzing about 400 products in different asset classes over several years.

“Willis Towers Watson’s assessment of diversity and performance outcomes shows that investment teams with diversity, in particular ethnic diversity, tend to generate better excess returns,” the report’s authors wrote.