What are the national or ethnic origins of FTSE 100 companies? There has been a concerted effort from policy makers, influencers, and vested interests to force the diversification of company boards, spearheaded by the so-called Parker Review for race and FTSE Women Leaders for sex. All this requires some sort of method for making a classification of ethnicity/race and sex. The obvious and not to mention painstaking way of doing this is by googling everyone and seeing what everyone is then writing it down. This is time consuming and may be less effective for some positions where photos are not publicly available. Can computers help?
There are always plenty of clues in our names. For instance we can make a reasonable guess that Hardeep Singh is both male and of Indian origin as well as Sikh, while Richard Norrie is male and of Scottish origin. There is a slight snag in that while Hardeep is all of these things, he is actually from Slough and as British as they come, so we would struggle to tell the cultural differences between him and a namesake from the Punjab or Canada (or Glasgow), for instance. Nevertheless, names can provide some evidence that can be useful as well as readily acquired. All that is required is some form of automated classifier. Fortunately, there are some available.
But first, the data. A list of names of FTSE 100 CEOs between 2023 and 2024 was acquired from here. Names were then stripped of special characters (accents etc.). First of all I tried the rethnicity package for R. Unfortunately, this proved a bit of a red herring since it struggled to classify names, and would wrongly classify people as black. Here’s the results:
As you can see, it’s providing way to high a number of black CEOs – 31 per cent!
And here’s a few examples of people it has wrongly classified as black:
Dominic Blakemore – Compass Group
Adrian Cox – Beazley
Pascal Soriot – Astrazeneca
All very much not black.
Hmmm… Perhaps the problem lies in the classifier was trained on Florida voter registration data and so was not appropriate for a list of British-based CEOs. An alternative method was sourced from Namesor which appears much better.
Using this method, I found out 90.2 per cent of CEOs were of European origin (white) compared to 8.9 per cent Asian and less than 1 per cent African (black). Note that this method will struggle somewhat with black people whose names derive from white slave owners.
If we dig a little deeper, then we see the Europeans are mostly accounted for by those of Northern European origin with very few from Eastern Europe, perhaps reflecting the damage done by communism. Asians are split between Southern and Western while the African minority is Western African.
Finally, the graph below shows CEO names by country of origin. As we can see, just under half of British CEOs have names of British origin. Most interestingly, 11 per cent have names of Irish origin. This is a telling finding, since when the Irish arrived here during the great waves of Irish migration of the 19th century, they were extremely poor and often reviled.
So what should we make of this? Yes, classifying people based on their race/ethnicity or national origin can be done, but there are some products out there that are better than others.