Having read PG's essays since 2002, I suspect his motive is that he wants to have good ideas. The kind that will be remembered after his death. He even said as much in one of his essays [1]. He's writing for legacy, not for profit.
YC itself was one of those ideas - it started out as an experiment [2], derived from a talk at Harvard Computer Society [3]. The easiest way to test if an idea is a good one is to go do it, so he did, and it turned out it was, and that led to profit.
I don't think it's fair to attribute Machiavellian scheming to PG. There are some other people in Silicon Valley that will take a position and then influence public opinion to make that position come true, but it's pretty rare, and I don't think it's the case here. Rather, it's the same reason all VCs talk their book: they start with a hypothesis about how the world is, then invest their book according to the hypothesis, and then tell everyone what the hypothesis is. The worldview comes first, the investments come second.
[1] http://paulgraham.com/speak.html
[2] http://www.paulgraham.com/bronze.html
[3] http://www.paulgraham.com/start.html