In an efficient market, the relative number of angel investors, VCs, and startups would always hover around an equilibrium. In reality, several factors prevent that from being the case. To name a few:
1) Glorification of the startup world, leading many to jump in even though it's difficult, stressful, risky, and not suitable for all personality types
2) Unrealistic expectations on the part of less informed investors, particularly angel investors, who may overestimate likely returns
3) Motivations on the part of angel investors other than pure profit (e.g. interest in better cancer treatments and contributing to companies aiming to develop those)
4) Greed and fear on the part of investors, including VCs and their LPs, leading to overreactions. For example, medical device returns were fairly weak from 2000-08, so many investors shifted entirely to healthcare IT, yet there's no certainty that the latter vertical will produce better future returns. If anything, the opposite is likely to be true, as only the strong startups will survive in an industry where funding is scarce, whereas an industry with lots of investor interest will see many weak startups get funded. This is why contrarian strategies often do well.
In general, I think there are too many startups and too many angel investors putting seed money into companies right now. This is the primary cause of the supposed "Series A crunch." I've see a lot of companies stay in the seed stage and raise convertible note rounds for 5-6 years, constantly one big break away from being "Series A ready." Needless to say, these companies are unlikely to produce great returns for the angels, at least in an annualized sense.
These early investors would be better served by investing only in very strong teams with a much better solution to a real, substantial user need and a proven ability to make progress. I've seen many startups that fail on one or more of these criteria raise loads of seed funding but never get anywhere. The risks facing these companies should have been obvious from the get-go.
Regarding later stage funding, I think the VC world is a little better at regulating the amount of funding it doles out than the angel world is, notwithstanding typical market volatility. If VC returns are higher than other asset classes, LP money floods in, and if they're lower, the reverse happens. Overall the amount of later stage funding is probably about what it should be, with temporary inefficiencies based on greed/fear around specific verticals (see medtech vs. health IT above) and macro economic conditions.