From the entrepreneur's side, or from society's, who cares?
(I agree Sequoia, Benchmark, Accel, and a few others are ballers, and I'd love to someday be an LP in a few, but that's not really what we're discussing.)
If a fund has a way to raise money steadily based on past reputation, market irrationality, a large collection of compromising photographs of leaders of oil producing countries, or whatever, and invests that money in a bunch of bad-investment startups but also my going-to-be-successful startup, it's a win for me and my users.
And those "bad investment startups" might actually be amazing companies, producing wonderful products which make the world better -- just at a really high valuation. There is surely some price at which a good company is a bad investment.
Maybe there's an argument for sustainability, but that's more over the long term. In 1999, or if you believe 2007 or now is an overpriced period for startup valuations, it still makes sense to raise VC if it will help your business, even if it doesn't make a risk-weighted great return for the VC. I don't believe this is the case now, but it's irrelevant to the argument of whether a startup should take VC.