Relatedly, the valley still does not have a good story for first employees. Given the economics of stock options, there is currently no good economical reason to go working for a startup when the option of working for an already IPOd & liquid Corp exists.
I wonder how much the proliferation of start up failures has to do with this very fact versus a poor product/model/founder/execution/etc.
Not saying that the employees are to blame but I'm sure there's plenty of good ideas out there that just don't have the right people trying to implement it.
Or that's such a small amount that it really doesn't contribute.
If I were to start a VC-backed startup I would just employ the paradigm most big publicly traded tech companies use: pay competitive market rates for good engineers as FTEs, and contract with body shops for all the other work. Given that FANG are some of the best performing companies in the world I'd say it's a safe assumption that they've done the math and found that it's more economical to pay a high rate for a limited number of high qualified engineers rather than pay medium rates for less qualified engineers. And then you can have contractors making less than what "medium" engineers make do the less important/skillful stuff.
Not to totally dismiss your point, but don't forget that one of the reasons, a very big reason, why FAANG companies pay that high is because they literally print money like crazy, so they can afford to behave that way. They have tons of cash to burn and they're not going to run out of it anytime soon.
If you are a startup that just raised 10M$, you just can't go around and pay a few good engineers $500k a year in cash compensation (which is what they would make at FAANG, since the RSU portion of the compensation is basically cash-equivalent, perhaps discounted at 80% if you want to be conservative), even if "a few" is a very small group.
You'll have to settle with offering them $180k base + stock options that on paper might bring their compensation to $500k, 5-10 years down the road, and any reasonable good engineer should value them at 10% of their pitched value, if not less.
It just really never makes sense to join a startup for financial reasons, and I say this as a person who was lucky enough to have a ~1M$ windfall from stock options of a former startup employer: amortized over the amount of time I spent at that company (4 years), I'd have been much better off financially had I been at FAANG the whole entire time (which is where I am now btw, completely done with the startup bs).
I think this is a relatively recent development. 8-10 years ago, working at a startup or working at FANG paid about the same, but the startup came with this lottery ticket, you learned more at a startup and it wasn't a big co with big co politics and process. Back then it definitely made a lot more sense to maybe give up %10-%20 less income in exchange.
Startups & VCs started getting a lot of money in the global search for yield as interest rates dropped to 0 and QE started printing money and giving it to banks who would invest it. The google/apple/etc wage collusion lawsuit increased the amount of competition for engineers. FANG was losing engineers who would rather do their own thing for a little bit less. The H1B lottery system limited the supply. Housing started getting even more expensive in the SFBA. And as you said, FANG prints money so they can afford to keep on going up in the bidding wars.
Compensation started going up in this competitive system as a result until we are at the point we are today, where startups are definitely not competitive comp wise to being at FANG.
Given my experience, I don't consider FAANG engineers any more qualified than other engineers on average. Actually, people coming out of agencies and consultancies that sell their development expertise are usually the most qualified candidates I interview, whereas FAANG companies are banking on hiring more engineers than anyone else and then building systems so even shitty engineers can productively contribute.
