The most underreported self-dealing by the VC class isn't the 2-and-20, but the revolving door between VC firms and overpaid executive positions at VC-funded companies. Being a partner at a VC means you get a lot of executive positions to hand out to your B-school friends who turned out to be underachievers. This means that a bunch of people will owe you favors, 10 years down the line. This type of personal draw-out (funding suboptimal businesses to win future favors from founders and other VCs, rather than the best) is also impossible to prove or even measure.
VCs draw a decent salary for themselves, but the personal career benefits of being a king-maker (which are impossible to track) are even bigger. The savvy ones essentially assure that they'll become two- or three-digit millionaires (billions is unpredictable and hard to assure; you typically have to go all-in on your own company) in the next bubble cycle. You can't not get rich if you're a VC, and for 95-99% of these guys, I could do their jobs better and I'm a 31-year-old relative nobody.
For better or worse, they're the main (and perhaps the only reliable one) valve between top talent and the ocean of passive capital seeking better returns than what index funds offer, and they're not going to give that up.
It's a Cheap Votes Problem. Why is it illegal to sell one's vote? Because, even if the statistical power of one vote is low, the statistical power of a bloc of votes can be quite high (voting blocs grow quadratically in contribution to variance). If you allow vote-buying, the people who aggregate cheap votes gain a ridiculous amount of power. This was a major problem in the Gilded Age. Venture capitalists are cheap vote aggregators when it comes to passive capital and high-talent business formation.