I find this essay frustrating for a number of reasons, but the biggest one is that it feels like a rehash of his "Inequality and Risk" essay from 2005 - except this one is less direct and buries the lede until the end ("Of course the Gini coefficient is increasing").
Because here's the thing that's wrong with most of the thinking in this essay and in the last one: measuring the success of your economic system based on outliers tells us very little about whether that's a good economic system. For the majority of people in a given country, whether the wealth of the top .001% is inherited or built through some sort of ability to scale doesn't matter, it's still a bad system. If ten thousand people try to do something and only the one that succeeds gains anything from it, we are looking at a non-viable economic system for 99.99% of people.
And I see this all the time now, in all places. It's like when people say "Look, this guy makes huge amounts of money from Twitch/Youtube/Patreon (with the older standards being music/hollywood and pg's focus being startups). These are winner-take-all systems. That's why there's people without inherited wealth at the top of the charts - they won, and somewhere out there are tens of thousands of people that lost. This doesn't mean those people are destitute or homeless, mind you; people that "lose" in the software ecosystem go work for the winners, and things sort-of work out (its worth noting that things don't work out as well in other spaces). But consider - if I have a half a dozen people that start companies in the same space, and 5 of them drop for various reasons - bad marketing, bad design, bad customer support, whatever - does the guy who wins really deserve to be not just 10x, or 100x richer, but 1000000x richer? Is this a just system? Is it even an effective one? Are we really incapable of imagining an economic structure where people don't lose motivation to do great work without the possible reward of billions of dollars?
There's further complaints to be made - looking back only to the industrial revolution to consider 1982 the anomaly, when since the dawn of civilization inheritance has been the standard form of wealth transfer; the social impact of high inequality and how the massively wealthy have a tendency to warp society around them (this was at least examined to some degree in the other essay); and so on.
The future I see from pg's inequality essays is one of increased separation: one in which rather than having an accountant or travel agent in every town, these services are provided by a single company that serves to funnel that money from hundreds of thousands of local areas across the world to a single business, probably somewhere in SV - and a few hundred thousand middle class jobs disappear into the void. And as our technology increases in capability, it happens again, and again, until we're largely left with a few groups - the founders and funders that own everything; the few remaining high-skill jobs that make and manage the systems for huge winner-take-all tech companies; and the rest, which spend their lives doing work directed and optimized by computers until a point is reached where that work can be automated cost-effectively. I'm not sure that's a world I look forward to.