In August of 2022 we debuted a feature that allows users to watch two short videos, each of a different CEO, and choose which one they "preferred." There were twenty-two CEOs, selected from among US Fortune 500 companies.
Our our goal was to measure CEO likeability as a proxy for their ability to attract and retain talented employees. Given that the latter is perhaps the most important skill a CEO can have, we wanted to test if likeability and share price performance were postively correlated: In other words, could we predict share performance from the ratings a CEO received in our videos?
To best test our theory we included CEOs from similar pairs of companies (Stryker and Boston Scientific, Lowes and Home Depot, Coke and Pepsi, etc..) so that to the extent possible we could account for factors external to CEO charisma.
Once we had sufficient ranking data, we checked how it correlated to past share price performance.
On February 13, 2023, with 350 relative CEO charisma ranks, we conducted our analysis.
We compared the past share price performance of 11 CEOs that scored above average in "preferability" to the 11 that scored below, compared the top 5 to bottom 5, and the top 3 to bottom 3.
Ideally the results would show that the shares of the preferred CEOs outperformed those of the less preferred CEOs, with margin of preference correlated with margin of stock outperformance.
The data generally did show that. In the five years prior to Feb 13, 2023 the 11 most-liked CEO's shares outperformed the bottom 11 by 101 pct points, top 5 beat bottom 5 by 118 pts, and top 3 beat bottom 3 by 98 points. (See table at the bottom)
Our second set of comparisons was between pairs of CEOs in the same sector. For instance the FedEx CEO v the UPS CEO, Lowes v Home Depot, etc. The average five-year outperformance of the better-liked CEOs across the ten different pairs (two of the twenty-two CEOs did not have a natural comparison) was 118%.
This was inflated by a pair many, including us, might question if it was a true like v like: Ford v Tesla in the auto sector. Taking out that pair, the preferred CEOs shares outperformed by 52 points, on average, with eight of the nine pairs being in the black.
A flaw in our study would be if user preference of CEO wasn't based on how each came across in their video, but on name recognition and past share performance. The definitive example being if Elon Musk received positive ranks because people knew he was Tesla's CEO and knew those shares skyrocketed the past five years.
However, the rankings don't appear to be impacted by name recognition or past share performance. Our most preferred CEOs were not household names: The top five were Carol Tome, Marvin Ellison, Jensen Huang, Kevin Lobo, and Tim Cook. Musk came ninth. On the other side of the scale, the infamous Muck Zuckerberg finished tied for 14th, far from the bottom.
While past share performance did correlate to CEO preference, an equally important test is if the outperformance continues going forward. "Going forward" started on Feb 13, 2023, and as of this writing (July 11, 2023) the outperformance continues. The top 3 liked v bottom 3 are up 39 pts, top 5 v bottom 5 up 24 pts and top 11 v bottom 11 is up 6 pts.
In the sector comps, the higher ranked ten have outperformed the lower ranked ten by 10 pts on average, with a high of 74 pts and low of -22 pts.
These "out of sample" tests have only five months of data (as of July 11, 2023), check back here for updates.
If you have any thoughts on this test, please email us at contact@fatfingerdata.com