3 ms·
>options are always more valuable than the underlying asset with the degree of premium based on the volatility of the asset. No, you have this backwards. Optio
by gibybo 12y ago
>options are always more valuable than the underlying asset with the degree of premium based on the volatility of the asset.
No, you have this backwards. Options are always LESS valuable than the underlying asset. The degree to which they are less valuable is inversely correlated to their volatility. The more volatile they are, the closer in value they are to their underlying asset.
I.e. a January 2015 call for Apple is currently trading at $4.36 per share. That means despite apple shares being worth $103 each, you can buy options for $4.36 each. If you want a longer term you can pay a bit more. A January 2017 call for Apple is $16.65, still more than 5 times cheaper than the actual share.
What employees get are more like options on the options. This secondary option multiplies the value of the option as you have described, but it also presents really awkward conflicts of interest where employees often end up with their shares reduced or being pushed out of the company altogether (see Skype or Zynga). Even still, it's not clear whether this multiplicative value of the option ever makes it worth more than the underlying stock. It could just make it similar.
- shalmanese 12y agoI was curious if companies ever did fire employees to claw back vesting so I asked on Quora a couple of years ago: http://www.quora.com/Equity-Compensation/Do-immensely-successful-startups-often-fire-their-early-employees-before-their-options-fully-vest-If-not-why-not?share=1 http://www.quora.com/Equity-Compensation/Do-immensely-succes... and apparently the answer is, outside of Skype and Zynga, the practice is rare.
- gibybo 12y agoIt's not just firing them or taking back their options. The incentives change such that the company doesn't really have to put any effort in retaining them anymore. Bonuses/raises/soft perks/etc that the company might normally provide would be less likely. They don't necessarily have to worry about keeping you happy with what you're working on which is not something you normally think about when considering the value of equity. It's hard to predict exactly what will happen, but the broken incentives are clear. A better model would have them aligned much better.