Barron’s recently published a list of the 100 most sustainable public companies. The research was presumably rigorous, but the analysis was so misleading I had to say something. It’s a troubling trend that most people don’t even read beyond headlines and hooks anymore, but still endorse bad material with a like or share. Publishers make money by getting clicks and views, that’s fine. But I draw the line when terms like “research” and “study” are used to make opinions look like facts.

Barron’s list of “The 100 Most Sustainable U.S. Companies” is just one of many examples I’ve stumbled upon recently, and it helps me make a couple points:

1. We must be careful about indices in general, especially when their calculation is difficult to explain. This particular list takes stock of over 200 indicators to build out a single score that summarizes a company’s position on environmental, social, and governance (ESG) factors. These three factors are hardly related themselves. For example, it’s one thing to have social programs in place to alleviate homelessness (Cisco), and a completely different thing to have “robust corporate governance” (HP). Just how these two things can relate to each other enough to make up a cumulative “sustainability” score should concern the consumer of the information.
2. It’s a slippery slope that gets worse when rankings are correlated with actualized and expected business performance. The Barron’s article touts that the 100 most sustainable companies on its list lost only 3.2% in 2018, while the S&P fell by 4.4%. If you remove just two of the largest gaining companies, RingCentral and Deckers Outdoor, you get a loss of 4.59% for the 98 remaining companies, resulting in poorer performance than the S&P. That’s after removing JUST TWO companies. See how fast the value of this index falls apart?

Spurious correlations like these are misleading and erode trust. Tyler Vigen, who I believe coined the term “spurious correlations,” wrote a book about this stuff. Some ridiculous examples from the book highlight the problem: There is a 99.26% correlation between the divorce rate in Maine and per capita consumption of margarine; a 99.79% correlation between US spending on science, space, and technology and suicides by hanging, strangulation, and suffocation; and there is a 98.51% correlation between total revenue generated by arcades and computer science doctorates awarded in the US.

Correlation is not causation, context matters. We can call it optimism or wishful thinking, but there’s no objective way to validate poor analyses or misinformed opinions. So we need to keep our guard up and hold people accountable for the statements they make, especially when they’re made by individuals and organizations that have a high level of influence.

Opinions are not the issue, they are important expressions of our unique worldview. The issue is opinions that are intentionally packaged as authoritative facts. For instance, look at the 2001 book Good to Great by Jim Collins. Collins defined seven characteristics he observed in 11 great companies that were particularly successful at the time. Two of the companies experienced significant issues: Fannie Mae’s stock dropped from $80 to $1 from 2001 to 2008, and Circuit City went bankrupt. Maybe we’re okay with an 18% failure rate, after all, the other nine companies are still doing pretty well. But it’s still an 18% failure rate. I found this example in a 2017 paper I keep going back to titled “Biasing Research: Can Science be Trusted?” If you’re interested in learning how biases can ruin even seemingly sound research, you’ll enjoy this paper.

This “facts” issue is a real thing. In 2017, the University of Washington launched a course titled “Calling Bullshit in the Age of Big Data,” and Shiv Singh and Rohini Luthra have just finished writing their new book, Savvy: Navigating Fake Companies, Fake Leaders and Fake News in the Post-Trust Era, which publishes on February 28.

When we see a sensationalized headline, we get the first indication that the truth is about to be tested. When we read the story hook, the author’s intentions begin to unravel. If we stop there, who else will read further to discover the truth?