Stock options: a balanced approach

This might be the first blog post of yours that I don’t agree with, and I think I've read all your posts. I agree with the individual parts, but I don’t believe the conclusion is universal. Since you told some stories, let me return the favor and tell a story about an internship I took, oh, maybe 13 years ago now.

I worked with a number of people who were probably in their 50s who really weren’t happy to be working. That wasn’t true of everyone that age, but it was true of a significant fraction of people. In candid conversations, people told me that they’d be retired now, except that they spent time when they were younger doing something that didn’t pay off (examples included things like earning a PhD and then getting stuck in a “normal” job, and going the startup route without a big success). These folks did not minimize their regret by taking on more risk.

Looking back, the main thing that’s striking to me is that no one in my peer group is worried about what their 50s will look like at all (with the exception of folks who have health problems, which is a different kind of worry). I don’t think that’s because my generation is different and isn’t going to worry about that kind of thing once we reach our 50s. IMO, it’s more likely that it’s because it’s hard for people my age to imagine a risk like that.

BTW, I’m not saying that all people should necessarily do the safer thing. I’m just saying that some people will regret not doing the safer thing.

good spick thanks

@Dan: I'm very flattered by your assessment of my writing, I guess I'm surprised too 'cause I must have said quite a lot of crazy stuff over the years.

And in this instance – I must have phrased it wrong because I do agree with you, it is wiser to take the safe route, provided that its benefits are as large as you depict. Which is why I said "...if you're at a startup and there's no compelling reason to leave" and I listed a really well-paying job elsewhere as a compelling reason to leave. I'm actually dumbfounded by your numbers; I can't believe anyone would take a 3x-5x pay cut to work for a startup, it's utterly deranged.

Anyway, what I've seen a lot of people do is stay at a startup and disregard equity and that doesn't make much sense. And for them staying actually made sense in itself because they don't have the $250-350K/year option. And once you do stay it's unwise to just disregard equity because it tends to be worthless; you're already there so you might as well insure yourself against an unusual success of that place, so to speak.

So basically I'm agreeing with you and all I wanted to do is add the parts of the picture that are relevant for people facing different options.

Wages for a programmer is pretty much the same as what you state, in Australia.

Whatever anyone thinks of the thesis, this post makes one significant error and one significant omission:

* "The right answer is that now it will most definitely go up, albeit perhaps by a smaller percentage than in previous rounds" This is inaccurate. Private company valuations regularly go down, even after a late-stage funding round. See Fidelity's recent mark-to-market valuation changes for a dozen examples.

* "If the stock goes from $1 to $10 (a "large" 900% increase), you get $9 when you exercise the option. If it goes from $50 to $70 (a "small" 40% increase), you get $20." That statement is factually accurate, but it neglects that the size of the option grant will be far, far smaller (in at least # of shares, and probably in face value).

* They do go down (and they might go up later.) All I'm saying is that unlike a VC which usually loses most of the investments for the occasional big payout, later stage investors tend to make something on most of their investments. What is true is that I don't have the stats... But I'm pretty sure that early and late stage investors both tend to make money and that late stage investors have a much lower variance. I think it's not a controversial statement and to my surprise a lot of people do not get it at all; they feel that after a late round the price is no longer likely to rise, while previously it was likely to rise. I think it's both widespread and pretty wrong thinking. What is true is it's unlikely to rise by a huge percentage, and hence the next point.

* Regarding the grant size – someone can go ask for a grant and multiply the size by the per-share gain. In my experience a lot of people (PhDs included) do not know how to do the arithmetic and automatically assume that "stock options are worthless now". As to the grant size – it can vary very strangely with time. Where I work, a new employee would get a rather similar amount of options in 2002 and 2012 (a lot less than they would in 2000, a lot more than they would in 2014), even though the valuation went up by 7x over that decade, I think. The grant size shrunk maybe by 1.5-2x. I guess one reason was that management's estimate of what the IPO valuation is going to be, and how likely an IPO is to happen, stayed roughly constant, though I never read their minds. So all I'm saying here, one should know how the arithmetic works and then check their own situation. I do know a few people who got most of their stock options very late. Also I didn't "neglect" what you say I neglected, I said it a few sentences later.

YES YES YES:

Options are insurance against unrewarded success.

Since everyone rightly assumes they will be worthless, collect as much of the funny-money as possible!

and

Sell from your portfolio when you need the money. You can define 'need' any way you like.

Finally, someone agreeing with my two main points! That's encouraging...

I have to agree with Dan above, I've read all your columns (although I responded very few times), and it's the first time I have to disagree (at least on the financial side): if you know how to find your way in a big company, and you are good, then — on average — being 20 years in Google / FB / Apple / IBM / MS / HP would get you more or less the same amount of money as being early in a successful startup (you are not working in a successful startup, but in one of the most successful Israeli startups ever, so you are biased).
As far as I know higher tech-track salaries are same all over the world, so an HP DT, or a google/IBM MTS would make about the same in Israel or in the valley, and the numbers or more or less as in the piece you quote (although, it usually doesn't take 5 years getting there, even if you are good, more below).
I think the main thing he misses is that a lot of technical people would simply get lost in a big company (and thus never make it to that pay grade): pushing technical innovation in big companies (which is what gets you promoted on the technical track – duh) is not only about the technical details, it's abut understanding the company.

Regarding my bias – that very successful startup is also, shall we say, fairly conservative in giving out equity, as you can see from the published compensation of the top VPs. I'm guessing it somewhat negates my bias, based on what I hear about equity grants elsewhere.

Regarding it being realistic to make the numbers Dan cites working for, say, Google (not to mention HP) in Israel – either you are misinformed or I am; I'm sorta hearing that $150K/year is pretty, pretty good. Or, if what you're saying is that the numbers are lower in Israel but so are startup payouts – perhaps, but Dan mentioned a 3x-5x difference between startup and big company compensation (startup equity excluded) in the US, which definitely does not happen in Israel, not even close.

Overall what I'm saying is, one should look at the specifics. I once interviewed at a startup which got $100M from an investment bank very early on, and I kinda thought the company was shit; the investment bank ended up losing those $100M. I had very specific reasons to dislike that company. All I'm saying is, you don't want to have a rule that you apply regardless of your situation. If you want to have a rule, "equity is worthless" is not a good general rule; "equity is worth much less than they tell you" is I think a pretty good general rule.

An example: Jamie Zawinski sleeping under his desk at Netscape, whatever he says, was very obviously done for the money (he quit programming after cashing those options and he said he was proud with getting it done much more than how the code came out, etc. etc.) Again, whatever he says now, it was a smart move, Netscape being what it was back then. Of course not every startup warrants sleeping under a desk for 6 months. Sometimes it's a worthless company in all likelihood, at other times there's not much to be gained by this sort of sprint because it's a slower market, at still other times they don't compensate you for it so why bother, etc. etc. I just say, look at your particulars, instead of reading someone else's reasons with a correct (for them) conclusion, forgetting the reasons, and misapplying the conclusion to your different circumstances, which is what happens with pro as well as anti startup memes (at least this is how a lot of people I know reason about these things – in all too general terms.)

Oh, and it's always about understanding the company... things aren't what they seem, even if there are 10-20 people – you always have to figure out how things actually work, which is usually not exactly the same as what they tell about how they work.

I worked at HP for a few years (both in Israel and in the valley), and I know a few people who moved from HP to google. I'm pretty sure about the following facts:
- In HP, the higher you are in the technical career path, the lower the geography matters (around 10% on HP DTb/w Israel and the valley)
- In the US, the higher you are on the technical path, the closer the numbers are between companies (DT in HP US makes about the same as what-ever-they-call-mts-today at Google US)
- These numbers are north of 200K$ yearly (sometimes way north).
You can verify these three claims looking at glassdoor – you don't need me for this (I trust glassdoor because I know people in these positions).
Regarding the "knowing the system" – my experience is that it's much (much) harder to understand a big system (which is of the main reasons I'm not in one anymore).
So yes, it is always about the specifics, but the specific do not include only the company, or you technical abilities, it is also how you would fit the company, and if you want to get paid a lot, the last one is trickier in a big company.

Just to clarify, the 3x-5x is in places in the US with a weak tech job market, e.g., when I talked to companies in Portland some of them were offering $80k-$90k for experienced engineers and $60k for entry-level positions. It's only slightly uncommon for new grads to get offers with $200k total comp today, and certainly not uncommon to give out an offer that's 3x $60k for a new grad and still lose bidding.

For startups in the bay, most were "only" 2x worse.

I definitely agree with the point about how people underestimate late stage equity. If you believe HN comments, it's way too late to make meaningful money at places like Uber or AirBnB, but a lot of people who joined similar-ish companies at an equivalent stage in the past have done pretty well for themselves.

And I guess, if you can negotiate for 4x the equity, there's a good chance you're better off pushing for more equity than more salary, since most places won't give you an equivalent bump in salary, assuming the original x wasn't trivial.

Dude, it's all very interesting. (as usual i have no idea on the business side of things thought).

I have one minor question: did your employer ever make a fuss out of your writings here, or are you just too valuable for them for this kind of shit?

@David: Google salaries in Israel on Glassdoor look even lower than the numbers I cited at first glance. But I didn't look deeper nor did I look at other big companies. It's an interesting theme... I'll look more into it.

@Dan: 2x is still a lot; I don't think an Israeli startup will dare to compensate nearly as badly compared to a big company operating in Israel. I think overall we kinda agree in that we both think 2x, not to mention 3x-5x, is a crazy premium to pay for working at a startup, but if you don't pay that high a premium and you're there after all, you probably don't want to neglect equity.

@Michael: erm... I don't think what I write hurts my employer's interests (in fact, we hired one person who found us through this blog and another one recruited through an insider's personal connections was apparently positively impressed by this blog and also ended up taking the offer.) Most of what I write is not exclusive to my current employer and is influenced by experiences of other people elsewhere. Here in the comments I mentioned that my current employer was not giving out that much equity for much of its history and that's a specific comment, but I doubt that the shareholders would be upset by not being diluted very much by options-based compensation, and as to employees, they know why they stay (or leave), while prospective employees will know why they took or rejected an offer. And in any case, I'm careful not to mention the company by name, 'cause you never know who might google the name and then misquote you.

So no, nobody objected to my writings here while yes, some of my managers over the years have stumbled upon the blog, and I think they didn't object regardless of how valuable I am but simply because it's not particularly objectionable.

Interesting post and valuable comments, and I don't dispute most of it. I do take serious issue, however, with regret-avoidance as a strategy.

I don't think that regret-avoidance is truly rational for humans. I mean, you can assume that avoiding regrets is the most important thing, and then rationally strive to achieve that goal. But that's what I'd call a rational in a shallow sense. A deeper rationality involves being very careful about the goal you set for yourself, and I'm surprised that a human might honestly and carefully deliberate over his possible goals and still opt for regret-avoidance.

Money means having latitude in choosing your home, the health care you get, the education you get. And then there's your family and their housing, health care and education. Having money also means being able to retire early and do whatever you want, or working very little and devoting your time to art or play or politics. This is obvious, I know. And yet, when you prioritize regret-avoidance over wealth maximization, these are precisely the things you choose to give up. And for what? Avoiding regret? Life's full of regrets. Better get used to them, no?

Were this an alternate reality where someone were to ask my opinion about what is a good goal to maximize in purely financial decisions, I'd probably suggest something like:

max E(log(max(c,net_worth)))

where E is expectation and c is a person-specific constant.

The "max(c," is there because the cost of losing your socks is bad, but not infinite. The log is there because we always want more money, but money becomes less useful the more you have of it. When you're rich enough, you ought to be risk-averse in the big decisions and risk-taking in small decisions (because lots of small uncorrelated bets with positive expectation each are safe and profitable in the aggregate). Nobody would use this formula in real life they're not smart enough to ask my opinion, of course.

My point about wealth maximization vs regret avoidance was, if you see a sure opportunity to make an amount that allows you to do something that you really want, you shouldn't give it up for a 70% chance to make twice the amount and a 30% chance to make zero, even though 0.7*2*x > x. Using log(x) solves this one, but then if you tweak the constants, it no longer will, and the thing is, x tends to be a better known quantity than the 70% and the 30% which are more typically out-of-the-ass numbers. We tend to know more about what we want than about how the world behaves. The other part is, if x is too small to matter for you, and 2x is not, then IMO you should wait while log says cash out right now. Maybe you work in a way that makes you personally agree with log all the time...

So I kinda think of it along the lines of, maximization might work better on average as the various probability estimate errors cancel each other, and so 1000 people will do better with a maximization strategy, on average, and so will one person with 1000 positions in different securities. (And in these cases I guess you wouldn't want the log in there.) For a single position, however, cashing out after meeting a personal goal does not sound like a bad idea to me, and I think that on average, one will feel better about one's choice using this system than using any formula that does not take into account one's personal goals (unless, of course, one's goal is precisely to maximize the log of their expected wealth as determined by their personal choice of probability estimates!) I also think, however, that I'm not the greatest trader in the world and that in general it's hard to be smart about it; so the only piece of my advice that I "really" believe in is the one which I think most "professional maximizers" actually will agree with me on – namely, not to sell everything at the first opportunity if you can help it; on average, it slightly decreases the mean and greatly increases the variance and I think few would recommend it.

P.S. I have mixed feelings on "rationality"; in part because I think that our life is locally sensible, but globally nonsensical so rationality amounts to little more than optimizing a path to some ultimately deranged goal, and in part (or is it the same part?) because I think that our desires are too inconsistent between themselves at any point in time and between different versions of themselves over time for rationality to be definable. But that does not help one trade securities and in general moves us away from the topic; it's just kind of a background for my belief that having the fewest regrets is as worthy a goal/way of thinking for a human being as any other...

Myself, I sold all of my vested options almost immediately post-IPO, since I figured that tech stocks are fairy gold, meaning they're the most volatile asset class I know of (witness 2000-2002). I did not sell in order to buy anything in particular, I just wanted to get out of this huge concentrated risk. I seek long-term financial security and freedom to spend my time as I wish, and this step took me closer to that goal. Maximizing E(log) would have lead to the same outcome, I think. It's more of a way to hand-wavingly seek rationality than an actual rule I live by.

I'm pretty happy with having sold my equity (also because its price has gone down since then). My wife asked me at the time what would we have done had the stock price gone way up after we had sold. I said we would have felt bad at having sold, but we would have had a nice little hoard to cheer us up in our misery. I can handle missed opportunities in such a condition. Missing opportunities is what I do all the time.

I perfectly agree that we're doomed to remain inconsistent, but I still prefer to be more rational than I am. I'm not hell-bent on rationality though. Bounded rationality is just a great tool to have if you get it at a good price.

Post a comment

Author:

Site (optional):

Comment (Markdown):

Preview Post