There's been a lot of discussion since, including https://news.ycombinator.com/item?id=2445447 and https://news.ycombinator.com/item?id=3489719.
Also, at the time, Dan Shapiro argued against it here: http://www.quora.com/What-do-you-think-about-Joel-Spolskys-a...
I also think the share distribution Wizards of the Coast (Pokemon, Magic the Gathering) accidentally used was interesting: founders had no shares, and worked their way up into the single digits, which supported small, individual investors, but it's probably not recommended if you're planning for traditional investment: http://www.peteradkison.com/blog-entry-2-wizards-of-the-coas... and http://www.peteradkison.com/blog-entry-3-wizards-of-the-coas...
A team has a capacity for arguments that depletes over time as arguments exhaust the team members. Arguments happening in rapid succession set up a vicious cycle, because there's a migraine aura of bad communications surrounding any big argument, and difficult decisions that happen in that aura spark needless new arguments. Lots of arguments also carry a potential for resentment, which creates a longer-term communication problem which sometimes insidiously builds as the company runs.
The "trial arguments" theory that Quora comment suggests seems to me a little like those parents who throw "chicken pox parties". It's probably fine and maybe even pragmatic, but it's a risk.
Why do I say this? Oh, no reason.. I just like typing things in boxes on the Internet!
- Joel confuses "easy" (50/50) with "fair" (working out the right number)
- It is better to argue yourselves to death early, when nobody else is affected, than later, when people are depending on you
- The expected value of an IOU is negligible because investors usually force you to waive them as a precondition of investing and they go to zero if the company fails
But perhaps I'm wrong. I'm expecting a round of innovation in equity allocation as companies heed sama's advice and try new things. I'm very curious to see how it works out!
What did work out well was Garfield's equity share, etc.
The IOU solution is not a good one:
1. Not taking salary when a startup starts is basically a very risky loan. An IOU simply doesn't take into account the risk involved.
2. This is not symmetrical to how investors are treated. In both cases there is an investment in the company which can be measured in terms of dollars. In the case of the employee he is only getting an IOU, but in the case of the investor, he is getting shares. I don't see any reason why these should be treated differently.
What you're trying to avoid is bringing company valuation into totally mundane cash flow problems like "who pays for plane tickets to first customer meeting".
It's a sign of very bad founding team cohesion when the founders look at each other as negotiating adversaries. Founders should prefer solutions that have a quick and intuitive sense of fairness over technical solutions that attempt to ensure fairness.
This is probably a bit more complicated in practice, but seems fair on the face.
Curious what other folks think.
Reasonable people can disagree on this point, but one thing that YC has said for years now that rings perfectly true to me: cofounder disputes can kill a company more abruptly than almost anything else. Rig your startup to minimize the possibility of resentment; you'll need all the unimpeded communications capacity you can get to resolve the problems that will arise intrinsically from your business.
If it were me, I'd force the founder to take the salary before I gave them more equity.
I finally decided I favor giving stock as bonuses based on individual merit, as a payback for any extra effort and dedication in the past and as a motivational tool in the future. Unlike Joel, I'm reluctant to see employee risk-taking as relevant or even measurable or fair, and I wonder if that is really the case at other startups. I mean, can one say their new hires are actually assuming uncompensated risk, beyond the reasonable risk anyone assumes switching jobs, as to be entitled to equity mainly for that reason.
Employee risk seems like an oxymoron to me.
For me personally, I want my employees to feel vested in the company. They are helping to build it, they are helping to mold it and shape it into something that will hopefully be very great. I want them to have equity because it gives them responsibility. (We don't have any employees at the moment, so it's easy for me to say this now).
Most importantly, I want the employee to feel like they are in an arrangement that they are comfortable with. If they are doing it as just a job, working exactly 40 hours per week, then a salary without equity makes sense.
If you were profitable from day one, I assume you run a services business?
If we were not profitable, we would not be able to hire them or pay salaries etc. so most of them would probably leave or sue. The resillient ones that stay behind would be given IOUs. I don't see much risk taking, unless you're really working for future pay that may never come, which may be the case for employee 1, 2, 3, but I doubt that's the case for layers and layers of new hires.
How common is this straightforward approach, where everyone is diluted in the same ratio of existing shares to new shares? I'd be interested in hearing about experience/knowledge other people may have had here.
I've never known startups to be steady long-term job providers. Seems like most live on the edge, always with not more than 3 months cash in the bank. Even when you get a big round of funding and hire more people, the investors want to use that money even faster than your last round.
Founder equity also compensates the founders for more than the risk that the company will fail and zero out their contributions; it also implicitly covers the upside risk of the founders, which upside was demonstrated by the fact that the founders created a company and presumably could have created others (or done something comparably lucrative) instead.
2. Skills risk, being at the top of your profession globally requires constant focus and professional support. The atmosphere at a very small company is hostile to this level of focus by necessity. It has a dulling effect.
The longer you have survived, the more mature you likely have become. This means you go from a demo, to a prototype, to a working product, to having a pilot customer, to have paying customers.
It's certainly true that some companies get tons of money without really being a mature company (especially in these days). The investors backing them are really pushing for a moonshot, so they invest tons of money and expect to spend the money quickly. Those are cases of less mature companies basically playing the lottery, and I'd agree it's pretty risky.
But most people don't have the connections, money, and drive to become founders. So those who do pay themselves well, relatively speaking. It's really just that simple.
EDIT: According to Spolsky in his hypothetical situation, Founder B was not a co-founder because he kept his job. Founder A, OTOH was essentially unemployed and took on all the risk, and therefore was a "legitimate" founder.
For the top layer: the real nature of a person comes out when they have a perceived opportunity to win big at someone's expense... hence good friends can make good cofounders. Failing that, find someone that plays well with others and considers the long-game of their actions. (You're gonna stick together for the next venture if this app doesn't work out, right?)
For the second layer: should be people you'd like to work with that may be founders in the future or people that have been recommended.
Third layer are more/less startup employees. So not regular corporate like employees that need to be thought for or are super niche, but T-shaped folks that can take initiative and worry a little more about details.
If a founder can't live with a slightly unequal share distribution, he is probably going to be the kind of guy who measures office sizes with a ruler. You're doomed anyway.
It's probably good to avoid a hugely skewed share distribution, but if you really have to pay people different amounts of cash, the loser in that deal ought to get shares.
Beyond tax implications and VCs, the IOU system strikes me as particularly terrible advice. In what realm is it reasonable to simply ignore hard interpersonal problems until they go away? If some group of people can't quickly come to an equitable arrangement for the division of equity, they shouldn't form a business. After all, founder breakups are a leading cause of failure.
Different loans have different tax implications. If the "IOU" you're taking is a deferred salary arrangement, and you are eventually paid a year's salary as a lump sum, that will obviously be taxed as income. If you're paid back a loan you made to fund operational expenses, and the loan carried no interest, the tax implications are likely to be minimal.
In any case, if your equity is worth anything, you're working with an accountant. Actually: if there's money changing hands in any direction, you're working with an accountant.
An IOU doesn't "ignore hard interpersonal problems". It's one of several resolutions to those problems. Your last sentence can be true without IOUs being unreasonable.
Also, the moment you have investors, the person's shares go below 50% and deadlock goes away.
(a) It only works when employees have control over revenue; the partner model disenfranchises important company roles that happen to be distant from revenues.
(b) It rewards the best salespeople and punishes people who prefer less business-facing and more technical-facing work.
(c) It works for investments/companies who are valued on continuing revenues from services, but breaks down totally when the company is valued based on forward revenues, which almost every software firm is.
(d) It creates an up-or-out model in which it is almost axiomatic that team members who fail to make partner will leave; in other words, it creates teams comprised of short-timers led by an aristocracy of long-term strivers. It also begs for churn and selects for ladder-climbers.
Even lawyers don't like the biglaw partner model. It does work, but know what you're getting into.
(I co-manage a consultancy that is larger than most YC companies).
Funded startups end with a pretty radically egalitarian share distribution by the standards of other industries. They'll generally IPO with regular working stiffs owning ~10% of the company. Do you know how much of the company non-management employees own at e.g. McDonalds? Wal-Mart? FedEx? The New York Times? NBC?
The only major industry which has as high a degree of employee ownership is -- and I'm aware of the irony -- finance.
The "enterprise value" of a law firm is very near zero because as the partners stop paying attention to everything the company falls apart.
Most startups make a product which can be sold largely independent of the number of hours worked by the employees. Certainly in a non-linear fashion. As a result a startup might have a substantial non-zero enterprise value.
He makes it pretty clear that that's an example and might not work for everyone (for example, if your hiring doesn't accelerate like that you might end up giving stripe 1 and stripe 2 hires the same number of shares with the suggested distribution, in which case you'd obviously want to move towards something more like your suggested 40-30-20-10).
