Thread by @benskuhn

More late-stage companies are switching from four-year vesting to one-year grants. Stripe, Lyft, and now Coinbase (blog.coinbase.com/how-coinbase-i…).

Seems like a good time to talk about why being an early startup employee is (used to be?) a much better deal than it might seem:

1. Picking winning startups is actually not *that* hard: benkuhn.net/vc-emh/ It's not *easy*, but some VCs, and TechCrunch award voters, do it regularly.

"If it's not hard then why aren't you rich?" Because plebs can't invest at those valuations, only name-brand VCs can.

("Why do name-brand VCs get a good deal?" Because many companies don't optimize very hard for valuation when fundraising. Instead they try to minimize distraction, risk of down rounds, etc. See post for a more fleshed-out argument.)

2. 4-year vesting stock grants are more valuable than a naïve calculation would suggest: benkuhn.net/optopt/

Briefly, this is because you can quit and move to a different company if it looks like the startup is doing badly.

The option of quitting is actually an option in the technical sense, and you can calculate its value using standard option-pricing math. If you do so, depending on your assumptions, the equity offer looks 30-100% more valuable than a naive expected-value calc would output.

Combine these, and you get a strategy of: pick the best-looking company, work 1-2y there, & if they're not a winner by then, repeat.

If you're as good as a TechCrunch audience member at picking startups, this gives you pretty good odds of making a good-VC-level return.

(Is this competitive with working at Big Tech? Hard to say—a lot depends on how good you are at picking startups, or at getting Big Tech to pay you a lot. It's been way better for me, but I got quite lucky!)

Anyway, this relies on 4y vesting to get enough mileage out of the winners to make it worthwhile. With annual grants, it's prob not worth it 😞

Understandable that Coinbase/Stripe/Lyft did this though, since nobody actually understood how to value the 4y grant (I sure didn't!)

This change is probably a Pareto improvement for Coinbase: they can offer a comp package with a higher "face value," while the actual cost to them is less because years 2-4 will be at a higher valuation. (Assuming number go up, of course.)

For that reason, I'd expect this to catch on, at least at places that are close to liquidity (+ hence have a more tangible expected-value for their equity).

I want to say "too bad for early employees," but I think most of us didn't realize what we were getting 😢