...spouses, children, nephews, the guy who gave you a jump-start in the Walmart parking lot...
And it's not necessarily wrong.
It creates significant inequities when everyone does it at scale.
Leaving aside all questions of legality and punishment (the only reason you care about punishment is if something goes wrong), what is the right way to solve this? How, in an ideal world, do we fix the significant inequity that many people today happen to have homogeneous networks, and hiring from your network means that you perpetuate any homogeneity, while also saying it's totally okay and encouraged to found a startup with a close friend instead of with someone who passes an exam?
And how do we get from here to there?
Is some amount of inequality at scale _necessarily_ a bad thing, if it arises from a fundamental systemic property of the socio/business/economy context we've bought into? I'll make a rather absurd parallel; look at money itself. To have more money enables gaining more with less effort, (citing the common "boots" theory of economics, basic laws of investing returns, etc) Therefore, enabling anyone to have more money than someone else creates significantly compounding inequality. Should we ensure that can't happen? I acknowlege it's a bit of a strawman but I would hope it also conveys a point both of implicit systemic cycles and tradeoffs of redressing them. As you concede, there's a tangible benefit towards hiring known entities and cohesive teams, even though at scale it tends to favor the incumbent majority.
Rather than ask if the inequality is in itself bad, I'd try to look at the funnel and see if there are any truly improper barriers to the out-group, and focus on rectifying those. Like in CS I'm always wary of KPI based iteration since it can cause "overfitting"/ignoring transitive effects, and "raw diversity numbers" to me is a very opaque KPI.
(Whether the systems that lead to these patterns is good is a totally separate discussion and one so huge I don't think I have the background to speak on it well, but I'm trying to question pragmatic steps to be taken that don't involve a burn-it-down-and-rebuild or throwing the baby out with the bathwater)
Which is statistically close to impossible.
That seems unlikely.
Q.E.D.
I'm not suggesting that the change was a statistical fluke. I'm suggesting that maybe in this case some women at Yahoo! were better management candidates than some of the men. In a ratio that's large but not unprecedented. Look at the rations found in other companies - so often its 80% men but that's not seen as unusual or wrong.
Sounds similar to the wage gap argument here: if it exists, where are all the companies hiring cheaper workers capable of doing the same work, exploiting the supposed market inefficiency?
Caucasians are still a huge portion of the Silicon Valley work force. It's women, hispanics, and african americans who make up some tiny percentage...
Palantir was recently accused of discrimination against Asians because they hire them at a much lower rate.
https://www.mercurynews.com/2016/09/26/regulators-accuse-pal...
Even if they were not deliberate about it, they must have talked about how this might be perceived by employees.
As bad as this issue is, I think hiring friend/referral/former colleague, especially en mass is much bigger issue that's rarely talked about because it's not necessarily illegal.
However, I cannot overemphasize how demoralizing it is, especially when they aren't proven to be any better.