Hacker News Reader

Comment by ditonal | Hacker News Reader

Startup ISOs are totally broken, and VCs and founders would rather write 30 page treatises on all their complexities (of course, glossing over the 99 ways to screw employees), than actually try to improve them.

A small percentage of people got rich off options a handful of times a long times ago, and since then countless people have been screwed.

Public RSUs for stock you can sell immediately on the open market are fantastic.

Common ISOs are toilet paper. At a _minimum_ you should get a 10 year exercise window, and if the CEO tries to say that would make it so early employees can hurt cap tables for future rounds, he's basically saying he doesn't consider your equity grant to be real equity as it deserves to be clawed back for the sin of not wanting to stick around for the 15 years it takes startups to IPO these days.

These exact same people will try to convince you that their ISOs are a valid subsitute to liquid RSUs, THEN say that they don't deserve to be "preferred" instead of "common" because the VCs put in actual money (hint: so did you if you turned down a public company to work at the startup. Biggest difference is only that you're way less diversified).

Am I ranting? Of course, but if VCs and founders are going to continually "educate" engineers on their equity offers, engineers need to stand up and inform each other of the pitfalls. I know countless people, myself included, who have been screwed by ISOs. You can actually lose money because the 30 day window forces you to pay strike price + taxes on gains, then you find out that the CEO sold the company at a bargain so liquidation preference kicked in and he just took a huge retention bonus instead.

The way out of this mess is not Github treatises on how to evaluate your equities. The way out is for engineers to continually tell founders/VCs that they will pick public companies instead of startups until ISOs get fixed. If ISOs screwed over VCs instead of engineers, they would have gone to DC and gotten this fixed 10 years ago. At an absolute minimum you shouldn't have to pay a dime in taxes until you've actually realized some money in your checkings account. VCs/founders don't care because they don't have to care because engineers are still too gullible and accept these bogus deals.

View on HN

> don't deserve to be "preferred" instead of "common" because the VCs put in actual money

I agree with most of what you said, but a nit: one perspective I've heard on the motivation for preferred stock is this:

Suppose I give you $10M to start a company in exchange for 10% of it. You then easily sell the company for $9M, keeping 90% * $9M = $8.1M for yourself and returning $900k to me. Preferred (non-participating, 1x) shares prevent this problem by making sure you can't just run away with the money: you have to actually use it to build the business.

People investing in the company in non-liquid ways (e.g., the founders or engineers, via opportunity costs) aren't in the same boat, because their opportunity cost can't be immediately liquidated.

This is one of the better explanations I've heard about preferred stock. I'd say the counter point for the engineers at least is that frequently the opportunity cost for them is vesting stock.

That lost revenue over the vesting period is IMHO the same as directly investing cash in the company that the company itself can liquidate. I would say that when shares are used for compensation with a vesting period they should be treated equivalently to investors that are directly putting cash in.

Also eng's typically get way way less grants than VCs putting up cash, so turning those shares to preferred shouldn't be a huge deal. But it is.

Right, I think the way to think about this is that common shares are for sharing in the value created at the company over and above the money invested in building the company. If your company spends all that excess value on perks and generating usage, then your common is more likely to get wiped out.

Except employees do invest money in the business by exercising their options. And that money, dollar for dollar, buys them an inferior product: lower priority in the liquidation stack; less liquidity because of abusive bylaws restricting secondary transactions; less tax benefits since your “gains” are taxed upfront.

The lower the pay, and the higher the relative cost of exercising, the worse it gets. So higher-paid executives are less penalized than entry-level employees who might spend their savings exercising options that will end up worthless.

"And that money, dollar for dollar, buys them an inferior product: l"

No, employees are usually offered options at market value when they are issued. They are essentially 'free' at the time of their issue. If they 'cost something' at exercising time, then they are 'worth at least' the cost of exercising, and usually more.

An 'option' definitely has value beyond a stock, because you don't have to actually take any risk if you don't want to. You can just 'wait and see' - which has very material economic value.

> No, employees are usually offered options at market value when they are issued.

It absolutely is an inferior product. Preferred Stock comes with a preference, in the absolute best case it's 1:1; owners of Preferred Stock will get 100% of their money back in a sale of the company before Common Stock owners see $0.1. There are sometimes worse deals (for the employees) where Preferred Stock gets back 2:1 or worse.

Common Stock (what employees get as ISOs or NSOs) is absolutely an inferior product.

I believe the grandparents comment was about options not common stock.

You are not accounting for the optionality of the ISO.

The ISO is an option to buy a stock, not actual stock. An option on a stock has material value due to the de-risked nature of it.

Someone with 'preferred' shares gets their money back first, sure, but someone with an 'option' never even has to put their money in the first place. An 'option' gives you all of the upside with none of the downside.

Would you rather put $1M in a company for 10K 'preferred shares', or would you rather be given an option to buy 10K shares at 1 cent each. Without knowing anything else at all, you'd probably choose the later.

> Someone with 'preferred' shares gets their money back first, sure, but someone with an 'option' never even has to put their money in the first place.

That’s not true. They are paid for at very high opportunity costs.

Yes but the context here is liquid investment, which is what preferred stock are meant to be a defense for (at least according to what I've heard).

(The opportunity cost is also paid slowly over years; I think VCs get better terms in part for front-loading all the money.)

> An 'option' gives you all of the upside with none of the downside.

To see any of that upside you have to exercise the option. If the company has the common 90-day post-termination exercise window, that means if you want to change jobs you need to come up with cash to exercise what’s supposed to be part of your compensation, and if the FMV is much higher than the strike price, hand a bunch of cash to the IRS too.

Now you’ve put in your time as an employee and reinvested the salary the company gave you right back into the company, and you’re likely at the back of the line with common shares.

Yes, windows on exercising are also part of the value, but you are fundamentally misinterpreting the concept of investing real money and optionality.

Your misunderstanding leads you to believe that somehow an option is worth a lot less than a 1x participating stock, and it's not.

Again you side-stepped the comparable:

As an employee, even with a '90 day exercise window' which would, you chose for compensation (or what would most people chose?)

1) the ability to invest $1M for a bunch of preferred shares; or

2) a bunch of ISO's at a low exercise price?

Everyone will take the ISOs due to their riskless nature, now why would they take option 2 if it's so 'inferior'?

A 1x participating preferred is a perfectly rational type of financial instrument for VCs, and some type of ISO is actually well aligned for staff. Though things might evolve here and there, there is zero chance that the 1x preferred aspect is going away.

As an employee, instead of exercising, say $30k worth of ISO, I would much rather buy $30k of preferred at fair market value. Even better if I got a discount on fair market value. I would also like the company to pay the 30k upfront as a loan or sign-on bonus, a common practice for executives already.

> I would much rather buy $30k of preferred at fair market value.

Sure, but VCs aren't getting shares at FMV either.

My bad, I meant "at the same price as VCs". You are correct that's not what fair market value means.

> 1) the ability to invest $1M for a bunch of preferred shares; or > > 2) a bunch of ISO's at a low exercise price?

Why is this the choice? how about:

3) a bunch of ISOs of preferred shares at the current FMV

Why is a VC putting in $1m riskier than an employee taking a full time job with the company? Why can't an employee who vests shares get the same preference when they reinvest their own money into the company?

Also: why do startups cover the new hire’s early exercise cost upfront when the new hire is an executive, but not for other employees ?

Because 'Cash in hand' is worth more than the risky nature of what you are able to deliver with your 'time in', esp. after receiving a salary.

That you will never get 'preferred' ISO's is a fundamental fact of the market, which accounts for such risks.

Implying it's 'unfair' is more or less like saying 'your salary is only 1/2 what it should be'. There's definitely an underlying reason your salary is roughly where it is, it's better to understand why it is what it is.

Options can be issued on any kind of stock. I think it makes sense to issue them on preferred stock. However I don't know what is the common way of handling this. In fact, the companies I know don't even have separate classes of stock.

In these discussions it is important to understand that each company might have totally different arrangements when it comes to equity.

> It absolutely is an inferior product.

I don't think they were disputing that? But it's not dollar-for-dollar equivalent -- VCs are paying more (a lot more, I think usually) liquid cash per preferred share than employees are paying liquid cash per option of common stock.

So use tranches and don't invest in founders who are going to run away with your money? If you're afraid of losing $10M because you don't trust the founders to invest it in their business there are safer investments you can make.

But what if they just keep raising money at crazy valuations because it's easy to get and hoard?

Exactly. Or just skip the tranches and only give founders as much as they can reasonably spend proving that they deserve more money.

As an investor, you want the CEO working on the company, not spending all their time and focus raising incremental tiny rounds and being one raise away from complete failure at every step.

As an employee, you want the exact same, only much more critically and personally.

Huge rounds lead to share dilution. The best thing to want is organic growth that preempts the need to raise rounds.

That actually makes a lot of sense, thank you.

>Suppose I give you $10M to start a company in exchange for 10% of it.

The scenario you present is a great excuse for preferred stock but the same result could be achieved by other methods.

For example the minority shareholder in this case could be given a veto over the sale of the company for anything less than 1.x times the most recent valuation. Or they could be given the right of first refusal on the sale of any shares.

But yeah I think it's just generally pretty common in the gamut of human relationships for people to try and put themselves in positions where they can't get screwed by putting themselves in positions where they can do the screwing. It's also pretty common for predators and control-freaks to try to elevate themselves to advantageous positions by justifying it in terms of having the right to defend themselves.

I think a good partnership is one where all sides work together to not only align their interests, but also to structure things in a way where everyone feels safe. It's much easier for me to totally throw myself at a cause, especially one that may involve making sacrifices, if I don't have to worry about the potential for getting screwed for my efforts. Having shares that I know will be just as valuable as all the rest can go a long way towards achieving that.

Exactly this. As far as I'm concerned the purpose of ISOs at companies more than a year old is to trick junior engineers into accepting a lower salary than they would receive elsewhere. Unless you have at least half a percent of a company you really, truly believe in I'd just ignore this article and put the value somewhere between $0 and a roll of lottery tickets.

Worse, as OP states, they can cause you to lose money. They can be worth less than zero dollars. Pay piles of taxes on the exercised options and then watch the company get gutted and preferred investors take everything.

Sadly, examples from the dotcom burst and housing crisis abound. It's a double edged sword. I have always preferred to go for early exercising.

"Many of these workers now owe far more in taxes than their stock is worth. Former Cisco engineer Jeffrey Chou, 32, owes $2.5 million in taxes on company stock he purchased last year that has since withered in value. Chou figures that if he were to sell everything he owns, including the three-bedroom Foster City, Calif., townhouse that he shares with his wife and 8-month-old daughter, the family still could not pay the bill."

* https://www.chicagotribune.com/sns-tech-taxes-story.html

That sounds like a single-edged sword to me.

Hell, plenty examples exists from when the general economy was just fine. Startups are startups.

Jeeze. The IRS must work with you on that, right?

Talk about getting screwed!

> $2.5 million in taxes on company stock he purchased last year that has since withered in value.

but if the stock dropped in value, doesn't the loss in capital offset the taxes? Or has he _already_ paid the taxes, and cannot claw back any losses (since he has to offset those losses on future capital gains from other sources)?

The stock is counted as income when it is exercised. The basis is the strike price then, once you sell later it’s considered either capital gain or loss. My understanding is that you can only apply 3000 in capital loses to income tax liability.

You can apply unlimited capital loss against capital gains and then an additional $3K of capital loss against income each year.

That is an NSO, not ISO.

That's not how ISOs work. You are taxed on the bargain element (discount to fair value provided by the option) at exercise under AMT, and taxed when you sell under the normal tax rules.

Let's say you worked at Uber and you exercised ISOs at the IPO that had appreciated such that the bargain element at exercise was $1,000,000. You would owe 28% Federal + 7% California = $350,000 under AMT (ignoring a lot of other things for simplicity) in April 2020. When you sell your shares, you will be able to reclaim some of this tax through credits for AMT paid in previous years.

By the time the employee lockup expired in November 2019, the shares are trading 43% lower than the price you exercised at and are only worth ~$577K. You still have a $350K tax bill under AMT due in April 2020.

Because you exercised in 2019, you have an out - by selling the shares in a the same calendar year, you can lower the basis under AMT, but you give up the ability to pay long term capital gains rates on the sale (0-25% fed + 0-13.3% CA) and instead pay ordinary income tax rates (22-43% fed + 0-13.3% CA). Let's just say it would be 40% - you could reduce your tax bill to $231K by selling all of your shares.

You have until the employee trading window closes in mid-December to decide whether to hold onto your shares for the eventual preferential tax treatment, hoping they recover before the tax bill ruins you, or to double your tax rate to save money on taxes this year.

Now, the benefit of hindsight:

If you had sold all of your shares, you would have missed out on $330K in gains in the following 60 days. If you held all of your shares, you are still facing a hefty AMT bill and years of careful tax planning to reclaim the AMT credits you generated. Oh, and even if the shares 10x in value, you are still going to have >$100K in AMT credits that will probably take 10+ years to recover.

This is absolutely crazy. I remember this exact thing happening to people in the dotcom era, they got taxed on gains that they were completely unable to realize.

If I understand it correctly, the problem here is the lockup period? If you get options that you are guaranteed that you can sell immediately after exercising them, you should be good, right? Then the proper course of action is to exercise all you can, sell enough to cover your tax liability, and then do as you please with the rest, keep, sell, whatever.

But if you can't sell them on the exercise date, you have to start thinking about the bullshit you just described above, right?

> If I understand it correctly, the problem here is the lockup period?

For the exact example of Uber, yes. Getting rid of the lockup period is one reason companies have begun using direct listings instead of bank-backed IPOs. There’s no group of investors who want their cut before letting common shareholders sell their shares.

Yeah, or just don’t exercise them hoping to hold for a year to get capital gains taxation instead of income taxation.

How does that work? Aren't options taxed as income on the difference between strike price and share price at the time of exercise? Capital gains require you to have shares first, right?

>but if the stock dropped in value, doesn't the loss in capital offset the taxes? Or has he _already_ paid the taxes, and cannot claw back any losses (since he has to offset those losses on future capital gains from other sources)?

It's the second. But someone getting 7 million in stock options should still be making bank somewhere and can probably offset it in the future. He's not really in that big of a pickle. Ultimately he will work out a payment plan with the IRS and pay it off over a few years.

He's also just an idiot. He should have sold enough stock to cover his taxes when he exercised the options. The people that get really screwed are those exercising options into illiquid companies. They may be paper millionaires as relatively junior engineers and have big tax bills. Then the company goes tits up and they lose everything without a reasonable expectation of future income high enough to offset their paper loss.

>> Or has he _already_ paid the taxes, and cannot claw back any losses (since he has to offset those losses on future capital gains from other sources)?

> It's the second.

No, it isn't? It can't be, because he's saying he can't pay the taxes. If he'd already paid them, they would already be paid.

I’m not familiar with his specific case, but it sounds likely that he’s already incurred the taxes but not yet paid them.

I'm not saying you're wrong at all -- most typical startup ISO offerings suck and are not good situations for employees.

That said, as a founder who custom designed our stock plan to be as employee-friendly as we could, there is a better way when actually done right and not just "following the standards".

Stock plan improvements like super long post-termination exercise, offering RSAs instead of just stock options, and most importantly in my opinion, adequately educating your employees on the risks and return possibilities of their equity!

It is irresponsible and manipulative. IMO, for a company to say, "we're giving you X shares worth Y dollars now and it might go up way more than that." While many employees' eyes will glaze over a bit if you start citing tax code, helping them get the basics and fully internalize the risk/reward tradeoff is both doable and necessary!

Equity can be a really valuable part of early employee equity and it's just as much of a peeve of mine to hear people say it's worth $0/lottery as to hear companies oversell the startup dream.

> It is irresponsible and manipulative. IMO, for a company to say, "we're giving you X shares worth Y dollars now and it might go up way more than that."

That’s why it’s illegal to tell an employee (or any potential investor) that their X shares are worth $Y. You must give a 409a valuation or some other evidence to supplant the valuation, and you must clearly state what percentage the share represents (or disclose total number of shares). Misleading and false financial claims are fraud.

VCs and startup founders are shooting themselves in the foot. It makes very little financial sense to work at a startup vs FAANG these days.

Ok cool, Il tired of seeing these startups popping up. Great thing that they are now thing of a past as they wont be able to hire anyone. /s

A few startups are realizing it is a rigged, unfair game and are trying to balance things a bit. They are offering stock purchase and exercise windows valid for several years, instead of just months or even days. Unfortunately it is still not a widespread practice. I have yet to see a company equalizing stock priorities of employees vs VCs.

Yeah exactly. Early in my career, I loved working at startups, and aside from compensation would prefer to do so again, but the opportunity cost now is way too high. The one advantage I see at a startup is that if you pick one with a decent engineering culture, you can learn a lot more than most FAANG roles (although you'll probably learn more at the best FAANG roles than at good startup roles? Unclear.)

I think startups are a bit of a crapshoot even in this regard. You'll learn a lot at a good one, but could potentially learn a lot of bad habits at a bad one.

> I think startups are a bit of a crapshoot even in this regard. You'll learn a lot at a good one, but could potentially learn a lot of bad habits at a bad one.

I've seen far more shit habits from habitual startup employees than those who come from Big N. The people who come from big public companies are almost always better at best practices and writing code than startup employees. I don't believe working at startups really gives you good sense for architecture, scalability, readability, or a variety of aspects of programming. It's almost all about getting that short term dollar to get to the next stage of funding. So, raw first time implementation speed gets prioritized over all other aspects. I don't find many startups have very strong technical voices either. CTOs frequently being product people in disguise, etc. That's my experience...

That is all true but you’ll get one thing working at a startup that you don’t get at a big company. A sense of legitimate urgency. I’ve worked at both and at a big company things are glacially slow and engineers love to build big “enterprise” solutions to tiny problems that don’t need it.

At a functioning startup if you overbuild you die. Everybody working at one knows this so people are much more pragmatic.

At a large company, your over engineered crap is lost in the noise.

> That is all true but you’ll get one thing working at a startup that you don’t get at a big company. A sense of legitimate urgency. I’ve worked at both and at a big company things are glacially slow and engineers love to build big “enterprise” solutions to tiny problems that don’t need it.

The "sense of legitimate urgency" is a bunch of bullshit. It's just as made up as it is at any other company. I've worked at startups from seed stage to unicorn and a lot of the real urgent things aren't things that engineering needs to do - it's other shit and usually has to do with partnering with someone. And that usually isn't an engineering problem. Engineering is rarely the blocker for things being rolled out. It's usually that product can't determine what the right product is - business can't get the partners they need - etc.

At a small company - your overengineered crap becomes the institutional bullshit that holds the rest of engineering back for years because no one can replace it because there's no time for refactoring. At least at a big company you can throw away those solutions that don't work for people. At a small company - it becomes the shit that holds you back from doing things efficiently for years.

Because startup employees are typically doing tasks where they don't have much prior experience, your chances of learning from your peers is lower in an average startup than in a large company.

Indeed. Real upside is the only thing that lets them at all compete for employees. Investors facilitating deals that screw early employees are poisoning the well for all their future investments.

> Investors facilitating deals that screw early employees are poisoning the well for all their future investments.

That horse has bolted the barn long ago.

Also these days a motivated eng will get exposed to many projects / processes at FAANG at usually a higher quality than at a random startup. You used to need to go the startup route for generalist experience but there's so much you can pick up at the big co's nowadays.

Classic tragedy of the commons scenario. ISOs were originally a great idea for how an early group of employees could take a big salary cut in exchange for a shot at getting wealthy. Founders and employees both had common stock and would sink or swim together.

Then over the decades information asymmetry started to chip away at this with investors realizing they really only needed to keep the founder happy, and they could steadily chip away at the outcomes for the rank and file given those young engineers would keep coming in based on the mythology of getting rich from the previous generation.

It took awhile, but now that trust has been fully eroded, and simultaneously FAANG have recognized the value of top engineering talent, paying salaries to ICs that would have been unthinkable in the 20th century.

It's going to be very difficult for investors to win back trust because now the information asymmetry works the other way, with young talent just assuming ISOs are worth nothing. Even if founders make an effort to do right by early employees, it's very difficult to demonstrate this to someone who is suspicious but not seasoned enough to know what questions to ask.

This information asymmetry is also solvable at the company level and IMO, good founders have a obligation to fight to educate engineers who are thinking of joining a startup.

I'm not saying you're wrong at all -- most typical startup ISO offerings suck and are not good situations for employees. That said, as a founder who custom designed our stock plan to be as employee-friendly as we could, there is a better way when actually done right and not just "following the standards".

While many employees' eyes will glaze over a bit if you start citing tax code, helping them get the basics and fully internalize the risk/reward tradeoff is both doable and necessary!

Early stage, yes. Top late stage (unicorns) tend to win (depending on your risk aversion).

"VCs and startup founders are shooting themselves in the foot. It makes very little financial sense to work at a startup vs FAANG these days."

Consider that it makes no sense to do almost any job if you can work as a dev at a FAANG.

Not everyone can work at a FAANG, moreover, this is not just a VC/Founder vs staff problem, it's an existential startup problem vis-a-vis FAANG and the rest of the world. You see the issue in housing affordability etc..

It seems that the fairest way to compensate employees at a startup is to pay them in cash, but also allow them to invest at each funding round with the same terms and conditions and liquidation preferences that the VC's are getting.

Reminds me of employee stock purchase plans at big (non tech, older industries) companies where you can buy your employer's stock with your pre-tax income. There might be something to this idea.

Starbucks employees get to apply up to 10% of their (pre-tax, I believe) income to a fund that quarterly buys stock at a discounted rate (iirc, 90% of the stock price as of start or end of the quarter, whichever is lower).

It's a nice benefit, and it'd be cool to see something similar in tech.

ESPP is available to many (all?) public companies.

Certainly not all. They last public company did not have a plan; and of course, some plans are better than others.

Apple has a program like this. It's a 15% discount by default. Last quarter it was like 50% I think because the stock plan started when AAPL was at $200 or something and ended at $320

Tech firms have ESPPs. I first learned of them when I interned at Texas Instruments in 2005.

That's relatively common; I haven't had it at my past couple of tech/public employers, but have it currently.

ESPP is never using pre-tax income. You are taxed on the money you use to purchase the shares, same as everything else.

A huge point of compensating employees with equity is that startups don't have cash, at least not very much of it.

If you can't pay in cash, you're a charity, not a business. At best.

Wow, this is the first time I've ever seen a VC described as "charity."

That's an absurd interpretation. First, I don't know any startups that aren't paying their employees a cash salary. But the tradeoff always used to be that startups paid somewhat below market (and I'm not talking hugely under-market, usually in the 10-20% range), with a much larger potential upside if the startup was successful.

The problem with that tradeoff nowadays is that in many cases you're just barely breaking even or coming out ahead with what a FAANG can pay even if the startup is wildly successful. The only way that tradeoff will start being worth it again is if founders and VCs give more real equity to employees so they can benefit if the start up is wildly successful.

Yes, but they can still be quite stupid about it.

Once I was told, well, we want to give you a raise of X% of your base salary, but money is tight. So here's X-5%, and 5% in stock options. Not stock, options.

So, not only do we have to go public, but we also have to double (triple really, because you know I didn't get that 5% as cash the second year) for me to get the raise you say I deserve. Plus, in order to exercise those options to get compensated, I have to pay you that 5% in cash. Super.

Honestly, and along the lines of another responder, a big raise is never as sweet as a modest raise and a little autonomy. If money is tight, a little raise and a little more autonomy can smooth things over pretty well.

But that causes employees to lose a powerful tool they have: ability to invest future earnings in the company and recall their future investment (by quitting) if things don't work out.

"Recall future investment" makes no sense. Isn't that just a fancy way of saying "not invest"?

Well not quite. If a startup offers me $800k in equity over 4 years, I'm fully invested.

If it doubles in a year, I'm now earning $400k in equity a year for next 3 years rather than $200k. (I recall my investment)

If it dies after a year, I only lost $200k, not the entire $700k.

If the company also actually grows and is profitable, there is also little reason not to give out cash, because the equity is more valuable than cash. Usually equity is given so that employees would work more like owners, taking care of the long-term value of the company. However I don't see this working for many people because many people just don't see the value of owning companies - they don't invest in stocks etc.

Some of the reasons VCs get a better class of shares than employees are structural and unlikely to go away no matter how ISOs are structured; for instance, VCs get liquidation preferences for reasons that are sort of intrinsic to the concept of equity investment.

The exercise window thing is a valid point and is a reason to devalue employee options.

Then perhaps the start up should just pay above market rate in salary and not offer any ISOs since those ISOs are so dang valuable but couldn't possibly be given the same liquidation preference?

I mean, I agree. I think employees generally under-value equity, for understandable reasons, and that it's probably more efficient to compensate them in cash. That does mean that when the company sells, the upside goes entirely to management and investors, but I guess you can't have it both ways.

> when the company sells, the upside goes entirely to management and investors

But for early startups that's basically an unlikely outcome.

I agree. Working for early startups is risky. I'm just saying, we're unlikely to mitigate the risk by eliminating preferred shares for investors.

What about the converse, giving preference constraints to early/founding employees?

Early founders & execs handle this by a combination of a) large initial stake to survive dilution and b) the ability to cut a deal eventually to counteract impact of liquidity preference etc. ... but you can only extend that to so many people.

I don't think it's insane to structure things in such a way that you are basically saying if anyone makes money off this, these people do in proportion to their relatively small stake.

I suppose the real problems here are that 1) early round investors don't get this protection (but I think they can just accept that) and 2) at some point you get a different class of employee equity which could be hard to judge.

I don't understand the proposal. I don't see what problem preferred stock for employees solves. I do see what problem preferred stock for investors solves. I do think employees should care about the preferences investors get, and judge their opportunities accordingly.

Idea is risk management for early stage employees,as a mechanism to make it hard for the company to later write a deal where they are entirely (or nearly) shut out of an event that has at least some decent return for founders and late investors.

I was just musing, haven't thought about it deeply.

I suspect we agree on the real solution, which is work out your business plan so that you can pay real market rates (and give them a chance to participate in early rounds)

I suspect you're thinking about this in generalities, like "everyone is taking some risk so everyone should have comparable risk protections". But preferred shares for investors address a specific kind of risk, involving management's incentives, one not shared with employees.

The rest of the protections VC gets for their shares are simply a result of market power. They have it; employees won't organize and don't.

More specifically perhaps that very early employees should organize, and push for this. In particular, to counteract management incentives that may otherwise throw them under the bus. Agree it’s not identical to VC case, but similar enough that as a mechanism it could work.

> I think employees generally under-value equity,

Do you mean by this that they attache a low value to the early stage equity they are given, or that they should push for a bigger part of the cap table early?

The former I think is really hard to support. After all, it's a common trope (for good reason) that people accept salaries well below market value at a startup on the basis of being convinced to over value the equity they are being offered.

They attach a low value and do tend to think in terms of their percentage of the cap table, which is irrational. Ultimately, I agree: developers should be paid cash.

We agree on the latter, certainly. Overthinking the cap table evolution isn't very useful, but ignoring it entirely is irrational.

Weighing your relative percentage ownership of the company against anchoring points and folkloric benchmarks is irrational. You need to evaluate the story that gives your shares value, have some clear idea of the numbers the company has to hit to get there, figure out what you'll make according to that story if things go well, risk-weight that, and see if the offer is competitive. Most employees don't do this, even when they're empowered to do so by management. Shares are undervalued by employees; many more of them should just be paid cash.

Cap table evolution is part of that story. I'm not talking about folkloric benchmarks, I'm talking about understanding who holds what now, what the overall capital needs are going to look like, what that is liable to look like. This is very different at different points in the evolution and and employees should understand that.

I don't think we are far apart on this actually, I agree you have to do this sort of evaluation, and that it is fundamentally imperfect. I'm not sure ton how many employees do this or don't, I've know many who do. Many of them fail on the risk-weighting part, and more of them have overvalued their state, ime, than under.

If you are really early stage, I also think along with a clear and vetted idea of the numbers the company is going to need to hit is a plan on what to do when that doesn't occur. The more time goes by, the more information you have, the more accurate your EV estimates should be. If they are dropping sharply there is nothing wrong with going back to management and saying this isn't making sense anymore to me, we need to reconsider my comp.

the equity being offered needs to be discounted to make it comparable to the cash salary. If the employee over-values the equity, it is to their own detriment.

Right, but many people actively encourage them to do this, so I think it happens a lot.

83(b) election is one way to make ISOs more on par with RSUs. 83(b) isn’t cheap... But! These days a lot of companies (especially FAANG, but start-ups too) are offering $20k-$100k signing bonuses; those bonuses essentially give the appearance of competitive total comp while in actuality depress base salary growth. It might be productive for employees to push towards having signing bonuses turned into an 83(b) election bonus that covers strike and (some amount of) taxes. Then fewer employees get roped into dumb tax games that only entertain investors, and we’d be moving the notion of competitive compensation towards reducing stock risk for those who can least afford it. (If salaries grow more rapidly as a side-effect, that’d be nice too).

Correct - this allows you to get QSBS treatment on your stock - a huge tax win.

The better startups will even loan employees "money" to exercise.

I've never taken a job in 20 years because of money and I don't advise anyone to do that. I choose a company for culture, the people I work with, the kinds of projects, and quality of life. Also a desire to ship things and not spend my days in meetings or dealing with bureaucracy, a desire not to work with jerks, and a desire not to work on lines of business that I find morally repugnant. The latter criteria have made most of my FAANG offers quite unappealing. But I love going to work at a startup with my buddies, building cool tools that help people, and spending plenty of time with my family.

They commenter isn't saying you should pick a job just because of money. But you should also avoid a job that is taking advantage of you, by giving you absolute crap compensation, by "selling" you ISO's as something valuable, when in most cases it isn't because of the constraints.

We all want to work on things that are cool, but we also have bills to pay, and retirement to prepare for.

A "culture" of founders making a ton of money for my work is not a culture I would consider good.

Every startup I’ve joined, I’ve always wished for the founders to make bank on my work. Because that’s the only way I’d do well.

My point is that founders frequently make bank while your stock options go to zero.

If you’ve never done it, maybe you shouldn’t be giving advice on it?

I just took a job at a smaller public tech company that’s paying me almost what I’d make at a FAANG (more than double what startups were offering), and I’m experiencing none of what you’re talking about. Great work life balance, WFH whenever you need to, people are in the office 40 hours or less, people take 5 weeks vacation per year, we don’t have a crazy amount of meetings or bureaucracy, and our business is selling useful software tools to customers who pay us money for them.

Pretty nuts to do that while getting paid nearly half a million per year.

I just treat high compensation as a necessary, but not sufficient, condition for employment. It works.

>... you find out that the CEO sold the company at a bargain so liquidation preference kicked in and he just took a huge retention bonus instead.

There's something potentially misaligned between the control of the founders and the final grant value on the other end. There are so many ways to distort this in the final exit deal and these are details that you probably won't learn about unless you happened to exercise stock before the exit happens. In general, the positive outcomes seem "capped" in that a founder is inclined to lock in their gains with an exit if the deal is reasonable for them.

Maybe the option shouldn't be as a share of the company but a share of everything the CEO/founder makes from there on out. :)

For really early stage, I generally assume that a founder is trying to at least 10x their net worth so any hand wavy exit estimate generally gets capped with that in mind.

I agree that the way a lot of companies structure their equity compensation is terrible. How would you structure equity compensation at a mid-stage startup where a share grant would come with a real, substantial tax hit, without any possibility of liquidity in the near future?

I'm asking because we're just starting to think about how to do this ourselves, and I agree that most equity plans are giving engineers a raw deal. The answer might just be large share grants, at least until the share value makes that unattractive. I'd love to hear other peoples' thoughts on this.

IIRC, much of the reason ISOs work the way they do is because of the way the IRS treats them, and companies that offer much longer exercise windows are having to work around these limitations to do so.

How about double trigger RSUs?

The downside is that employees pay ordinary income on the full IPO value, but there are no upfront taxes and no exercise dilemma for people that leave after vesting.

Honesty I can’t imagine how a private company can offer single trigger RSUs. I would never accept an offer at a place where I knew I would have to write them a year after joining to cover taxes for illiquid not-quite-equity.

Enough that's a nice bonus but still a minimal part of total compensation. IRS treatment of ISO/NSOs is part of the problem but the real issue is lack of liquidity. If you do make the equity significant at least give your employees the option to periodically liquidate to reduce the risk of them paying tax on money they'll never see.

Tax issues aside there are many startups whose equity is incredibly valuable on paper but the stock is trading for a third of the FMV on the secondary market, if there are any buyers at all.

Yeah, secondary market liquidity on anything but the top startups seems pretty lackluster.

It is good exercise to contrast what accredited investors think of startup equity compared to average startup engineer.

Having a 10 year exercise window on options helps quite a bit. In most cases there will either be a liquidity event within 10 years or (for companies that are successful but stay private longer than 10 years) the company will have multiple huge fundraising rounds and can allow early employees to cash out some shares in a secondary round.

To be fair the CEO should either eventually push for a liquidity event or should be open to allowing employees to sell shares on the secondary market.

Yeah, I think long dated options plus regular internal tender offers help a lot. I need to look more into the implications of long dating options.

Allowing sales on the secondary markets can help, but the interest/liquidity on those is super heavily concentrated in the household names, so if you're working at a $X00M startup, it tends not to work so well, even if the company is nominally friendly toward it.

TLDR: Offer RSAs and education!

From my other comment:

"I think the biggest gap that this guide and every one I've seen like it that they undersell the positive value for employees of the Restricted Stock Award (RSA). RSAs can be significantly better than ISOs in many situations.

There is no actual reason why companies can only offer RSAs to founders and super early employees. The taxes get trickier once the strike price gets high enough, but employees should have the choice to maximize their equity value!

From the very beginning of founding my company and still 5 years later, we offer our employees the choice to take their stock as RSAs, options, or a combination of the two. We also teach a mandatory "Stock 101" course every quarter for new employees that is a 1.5hr version of this guide.

Too many employees' (engineers in particular in my experience), don't actually understand how their equity compensation works and I think it's the responsibility of the company to educate them -- both for their current role and possible future job offers."

This stuff is a huge personal passion of mine and we've put a lot of effort (and lawyer dollars) into this at my company. Feel free to reach out anytime.

Cool, thanks! I might take you up on that. In any case, I think that's the direction we'll go, at least in the beginning.

Your post captures my feelings and experience perfectly. I will never trade cash compensation for ISOs again. It sucks because I'd rather be working for small companies, but I'll be God Damned if I'm going to get some asshole rich off of my hard work.

This! In addition to all of the above, it is very hard for the candidate to judge what "0.X % of the company" is worth. Like if one gave up, say 50,000 in salary, is the equity worth that much or some good multiple of it.

The "preferred shares" and various other gimmicks played by VCs mean they might let the "current value" appear inflated than it really is. I have seen startups "start" with a valuation of "50 million dollars" out of thin air. No sales, no product, but a combination of confusing paperwork makes it appear as if someone recently "invested at that valuation". It will be too late by the time you find that the investor was nothing other than the founder, who found a creative way to value his/her time or seed money.

Another major mistake made by new candidates is to miss the fact that there is a vesting period - the upfront large amount means nothing if you quit after a year. On top of it one may not even get any new ISOs in future years. Companies always show "current-year salary + total ISO issued" as "total compensation" - that is absurd. Even if you know what you are doing, you will fall for it after 10 repetitions.

The standards that have us thinking about this in terms of % ownership of the company is part of the problem IMO. Future rounds and other factors can quickly make this calculus impossible or with 1000% error margin.

IMO, the correct way to approach this is keep everything in terms of $$ values. There is available research online that can allow you to aggregate data for comparable companies on things like:

"if the preferred shares price was worth $X per share at the series A, what was the that same share worth at IPO/acquisition?" It's still not easy to answer as the data is sparse and widely varying, but thinking in these terms normalizes the conversation around dollars and gets rid of the "what dilution might happen" part of the equation.

For example, my company is in the open source data infra space. So while its hard to get very many data points, trying to get info about the preferred and common share prices at each round and all the up to IPO of companies like MongoDB, Hortonworks, Elastic, Cloudera, etc -- even though they all raised totally different number of rounds and IPOed at different valuations -- in aggregate you can start to build an insightful picture.

ISOs can't be over 90 days by law. You can do non-quals for 10 years but you'll lose half of it on taxes on exit. Pick your poison.

Are you sure about this? I thought the limit across the board was 10 years.

You can make options expire after 90 days but they lose their IRS protection and have to be treated as income and not capital gains. ISO 90 day limit is an IRS limit.

https://employeestockoptions.com/difference-iso-nso/

You'd lose most of the NQSO's taxes on exercise of ISOs too if the proceeds are at all meaningful.

I've been in that scenario, and honestly I can't imagine why any sane person wouldn't have them convert to NQSOs rather than the true poison of needing to exercise within 90 days. An extremely common scenario:

1. You work your ass off at a startup for years, but even after all those years the company is still not public and the stock isn't liquid.

2. You quit, so you have 90 days to decide if you want to cough up the cash to pay for those options, with no real way to sell.

3. Oh, don't forget the huge AMT bill you'll likely get that year.

4. Pray to God that company is eventually sold for more than all the liquidation preferences of the investors.

Would you rather have that or take the 24-37% potential tax hit (when you can actually sell your shares) vs 15% capital gains.

Also note you can write it so that the decision is totally up to the employee: exercise within 90 days and they're ISOs, after that either lose them or convert to NQSOs.

> Public RSUs for stock you can sell immediately on the open market are fantastic.

I mostly agree with this statement, but they are not entirely without risk. With the almost universal 6 month lockup these days, through an IPO employees could be left with a large tax burden, including having to make quarterly estimated payments, on a bunch of income that might be worth less than it was taxed at, or even worthless, if the company folded.

Sure, it's unlikely, but look at some of the 2019 IPO flops. If you are taxed on a something like a million dollars of income, but only actually pocket half that, then at the end of the day you effectively have a quarter of what you started with.

If your RSUs dropped that much in value before you could sell them, isn't that some sort of loss you can note on your taxes?

Depends. AMT losses (from ISO exercise spreads and subsequent collapse) are separate and can't offset normal capital gains. That's what caused tons of pain for folks during the first bubble burst.

None of that has anything to do with "Public RSUs for stock you can sell immediately on the open market"

> Startup ISOs are totally broken.

> Public RSUs for stock you can sell immediately on the open market are fantastic.

Both correct statements, but apples and oranges.

But for an engineer they are comparable, because many are deciding between the two. So do you want the bucket of beautiful, fresh Honey Crisps, or 3 month old moldy juice oranges?

If you are deciding between the two, you are deciding the wrong thing. You should go to an early startup (ISOs) if that's the environment you want to work in. You should go to a public company or late stage startup (RSUs) if instead you prefer that environment.

The choice between ISOs and RSUs shouldn't be a factor that you are deciding on; it comes with the territory for the job you are seeking, based on other factors.

Fully agree with you. How would you solve it then - besides engineers being more explicit with founders? (not trying to be smart with you; honest question).

Sam Altman did a post with some simple things a couple years ago. Three easy things:

1. Extend the exercise period to 10 years as GP says. Many companies have done this. Literally, all engineers out there, just walk away if they insist on 90 days. 2. Make the option pool bigger. Altman has a very good point, that companies try to pretend that the size of the option pool is something set in stone. The board just made that number up, they can change it, and if the pool is too small to support decent equity compensation, again, walk away. 3. Founders will need to give up more equity. A founder does deserve to make a lot more as they took the initial risk on the idea, but it's bullshit that a founder deserves to make, say, 50x a very early critical engineering hire.

At the end of the day I do believe the market will force a shift. The FAANGs are paying so much, and while it's taken some time employees, especially junior ones, are starting to get better at evaluating their comp offers, specifically because of comment threads like this one.

Hi Sam!

Offering Restricted Common Stock (Same as founders) as a solution is more pragmatic than granting preferred stock (Same as investors)

Are options considered an equity grant? I hadn’t thought so, but maybe I need to update my own vocabulary.

In common language when discussing comp, yeah definitely, I think so. "Legally" idk.

Yes. Options given to you by your company as a non-cash compensation is an equity grant because it gives you the right to equity ( shares ) in a company. If you "exercise" the options, you get X amount of shares in the company.

If only there was some sort of collective bargaining possible to address the unique nuances of our industry

Employees should have capped liquidation prefs!

Reply on news.ycombinator.com