If diverse companies perform better, this would surely mean that there is some sort of difference between the sexes? How else could diversity have an impact on the bottom line of a company?

Rather, it shows that having people from diverse backgrounds and diverse experiences works more effectively than a social 'monoculture'.

Well, if women really make 87% of what men make for the same work (or whatever the latest number is), it could save the company 13% or more on car insur- I mean, labor costs.

If it is really true that women make 87% of men for the same work, any company that is driven by profit would exclusively hire women. The alternative would be that sexism is so rampant that it trumps profit driven motivation, am I missing something, or is that what you are suggesting?

I'm not really suggesting anything other than that, it's a hypothetical explanation that disproves the notion, "if diverse companies do better, it MUST be because of differences in the sexes". Admittedly not a high-effort comment.

In answer to the rest of your comment, it's anybody's guess and nobody agrees why. Maybe the stat is wrong, maybe the sexism is just that rampant, maybe the women-only candidate pool is inadequate (small), maybe there is a performance difference, etc.

Ah I see. Thanks for expanding.

>if there is proof that diverse companies perform better then his reasoning about how diversity initiatives are wasteful must be incorrect.

Absolutely not true. There are many other possible reasons for such a correlation. A few examples:

It's possible better-performing companies could attract more non-white applicants.

Or, that better-performing companies could be have extra resources, which they then spend on diversity initiatives, where poor-performing companies are desperate to survive.

Or the independent variable: Being in the Valley makes companies perform better, and also makes them want to hire more non-whites, but these are both independent outcomes simply caused by being in the Valley. In this scenario, it's even possible that diversity harms performance, but the effect is more than compensated for by being in the Valley.

Or, it could be random chance; outcomes are dominated by a few high-performing companies, which just happen to be genetically diverse.

Or any of many other possible connections and confounders. This is really basic stuff.

Thanks for expanding on that aspect! This is exactly the sort of things people should have brought up.

I feel like the suspicious thing is that people didn't react in this way.

Why was this not a glaring problem that his supporters rushed in to answer?

If I told you I had an elegant model for some chemical reactions and could predict, say, the thermal energy released in an experiment, you might look at the model and think it's great.

If it doesn't match what is measured in the real world, then it's possible the model is correct: could be an environmental factor or a problem with the apparatus etc. But I need to answer questions around the gap between prediction and reality, rather than to rely on my perceptions of the elegance of the model.

That's where I think this veers off into identity politics.