And even if you factor in taxes, he got a heck of a payday.
But what it does do, is give the holder the sense of wealth which is directly tied to the value of the stock. Make the stock go up and you certainly "feel" a lot wealthier even if you can't actually take action based on that feeling[1]. But you're very motivated to get that feeling and make the stock worth more. (at least that is the theory).
At the C level and even at the EVP level, you're asked to make some pretty big decisions, decisions that could turn out to make you unemployable if they look really stupid in hindsight. Sometimes the only way to get someone to take that job is to give them some assurances that even if they become unemployable they won't become homeless. (Yes, your argument is still valid, 250K shares would probably do that _at the current valuation_, the question then is risk mitigation)
I'm not saying that I agree with the decision but I am familiar with some of the reasoning that the executive compensation committee goes through when coming up with these numbers. [1] One huge loophole used to be (and may still exist) is that you can sell stock to pay off a mortgage.
The reason the 83b is impossible then is that he doesn't know how many actual shares this will be. But that is ok since the company sells some of the granted shares to withhold for taxes.
This scheme of 'introducing a moderator variable' gives the company exceptional leverage over the employee as they can 'take back' all of their stock grant by setting the moderator to zero.
I don't know what his contract looked like but I'm guessing it has a minimum value exit (that was used to get him to change jobs) so it would have been hard for them to not pay him any of the equity.
The guy's cash compensation, $500k. That is a about 2x what Google pays their top engineers (about $250k) The rest of his package was stock. [1]
The compensation theory goes that if you do well the stock will do well, if you do poorly the stock will do poorly. So most of your compensation is a chunk of stock, in this case about 2.5M shares as "RSUs" (a restricted stock grant with performance tuners tweaked to company performance)
de Castro's stock did well because of the Alibaba thing not because of what he did. So that left him with a chunk of stock. Had the company done poorly that would have been worthless. It could still become worthless. The filing indicates he got 1.5M shares in restricted stock, he no doubt will have to sell a chunk of that to pay the taxes on those shares (it will be treated as ordinary income by the IRS) and California.
When these stories are reported they pick the biggest cash number they can, but the actual value may be significantly less. Large stock grants are a tool to keep executives interests aligned with the company interests, they are given a lot of stock and huge restrictions are placed on their ability to sell that stock.
Had he stayed at the company, the restrictions on selling would have prevented him from realizing that stock value immediately.
[1] http://www.sec.gov/Archives/edgar/data/1011006/0001193125141...