Hacker News new | past | comments | ask | show | jobs | submit login
Open Guide to Equity Compensation (github.com/jlevy)
482 points by zalzal on Jan 11, 2016 | hide | past | favorite | 74 comments

This is generally a nice guide for anyone wanting to get an overview of equity compensation issues affecting startup employees.

There is one error I caught, where it says that "restricted" stock is called "restricted" owing to the fact that securities laws restrict one's right to resell such stock.

While it is true that the securities laws do use this terminology to describe the stock of closely-held companies (and in that sense the use of the term of is accurate), all common stock granted in a closely-held company is restricted in this sense, whether or not it is subject to vesting - that is, it must generally be held for a stipulated period as set forth under Rule 144 before it can be resold by the recipient (Rule 144 technically applies only to public company stock but applies by analogy to that of closely-held stock).

In the startup context, "restricted" stock refers to stock that is granted to a recipient but made subject to a repurchase option by which the company may repurchase it at cost on termination of a service relationship. That is, the stock is subject to a substantial risk of forfeiture until it vests. For tax purposes, such stock is not deemed to be owned, and is not subject to tax, until the risk of forfeiture goes away (i.e., it vests). This in turn creates a substantial tax risk to the holder of the stock because there is an immediate tax on the value of the spread (difference between what was paid for the stock and its fair market value at each vesting point), a risk that is eliminated if a timely 83(b) election is filed within 30 days of grant but not otherwise.

If stock is granted without any vesting requirements (i.e., an outright grant) it is referred to in common startup parlance as an "unrestricted" grant. Such stock is "restricted" stock in the securities law sense that it generally can't immediately be resold or transferred without complying with the Rule 144 tests. But this is a technical issue for the lawyers. For every practical purpose relevant to founders and employees, it can be treated, as the street parlance says, as an "unrestricted" grant because there are no vesting requirements.

Sorry if this is too technical for this thread. But this is an important technical point that is mis-stated in this guide.

A brief observation: when stock options first became widely useful in the startup world in the 1980s, ISOs conferred a huge benefit to employees because you could be assured that you could exercise them when they vested without any practical tax risk whatever. Over the years, however, AMT, though first enacted in the 1960s as a "millionaire's tax" to ensure that the wealthy could not easily manipulate tax deductions to avoid paying any tax whatever, evolved into a general catch-all tax that is now used to fill serious deficiencies in the U.S. tax code and that now ensnares many people making pretty average incomes. Once that happened, it effectively killed many of the once-very-special tax advantages of ISOs for employees and turned ISOs into a form of equity compensation that is only slightly more favorable for employees and is often a real disadvantage (for example, the notorious 90-day tail for exercising vested options on termination of employment derives directly from tax-code rules imposed as special restrictions on ISOs alone but today functions to entrap many employees into having to stay in undesirable employment situations far beyond what they intended on pain of losing their vested options altogether if they quit).

A final theoretical observation on best types of grants in order of preference: first, unrestricted grants (here, you own it all and can't theoretically ever lose it and you usually have zero tax risk while trying to hold for long-term capital gains); second, restricted stock at a cheap price with a timely 83(b) election (while it vests, and you can forfeit it, the tax picture is near-ideal in giving you a path to long-term capital gains tax treatment with no landmines along the way); third, ISOs with a low strike price and an early exercise privilege (with these, you exercise early, file an 83(b) election, and in effect get the equivalent of a restricted stock grant); fourth, ISOs or NQOs without a 90-day tail on termination (these give you maximum flexibility to trying to work around or at least postpone potentially detrimental tax events while being able to wait as long as 10 years before being at risk of losing vested options); fifth, and worst of all from a tax standpoint, RSUs (which really are a super-high-value startup's way of granting very nice bonuses to employees in situations where the very high price of its stock makes it tax-prohibitive to use any other more favorable equity compensation vehicle). Of course, this is theoretical only and you get what you get in the real world depending on whether you are a founder, an early-stage employee, or a later-stage employee and depending as well on what is negotiated with investors concerning any restrictions they may insist upon as conditions to their investments.

Just a few thoughts on what is, overall, a nice guide to equity compensation.

Thanks for the correction. I've created an issue with the entire comment so we're sure to correct it, and perhaps include more explanation based on this and the helpful background you've written. https://github.com/jlevy/og-equity-compensation/issues/24

I always love it when people use git & github for more than code in the way you just demonstrated.

I only just started organising some of my non-technical writings in a repository and can already see the benefits.

Great comment. We will update the guide Grellas. Thank you!

> RSUs [] are a super-high-value startup's way of granting very nice bonuses to employees in situations where the very high price of its stock makes it tax-prohibitive to use any other more favorable equity compensation vehicle

Why does the price of the stock influence the choice? By price of the stock, in this situation, do you mean that 10 shares of XYZ at a strike price of $1.00 per share is treated differently than 1 share at $10.00/ea? Or do you refer to the price of the stock in the sense that the dollar amount of the value being given is high? (I.e., the executive is being given a bonus with a dollar value of $10 million as opposed to $100k?)

This is very useful, and not too technical. I've always wondered what it meant, and yours is the best answer I've found so far.


You mention the preference order as:

1) unrestricted grants

2) restricted stock at a cheap price with a timely 83(b) election

3) ISO, with low strike price and early exercise

4) ISO or NQO without 90 day tail on termination

5) RSUs

How do ISOs with a 90-day tail on termination compare? If I understand correctly, those are the most common.

Actually, apologies for being sloppy on this one (in having dashed this off in just a few minutes) - under tax laws, you can't have an ISO without a 90-day tail. This should therefore read "4) NQO without 90 day tail on termination". I would probably put ISO with 90-day tail and no early exercise as #5. It really is a toss-up, though, because the ISO does retain significant tax advantages for anyone who can keep his option exercises out of AMT. I give the NQO (with no tail) the edge here only if AMT becomes a factor, which it does for many people, primarily because it gives you have the flexibility to keep vested options alive long after leaving your employment. Hope this helps.

> This guide applies to companies in the United States.

Can anyone from Germany (or EU) provide some information on how much of this is applicable to Germany?

Someone created an issue to add non-US content: https://github.com/jlevy/og-equity-compensation/issues/26

I can say that the Australian tax law in this area is very different and using a US designed approach will cause a lot of tax problems for employees.

This is Josh, of the authors of the Guide. First, thanks to everyone for the comments and feedback.

Also, please remember this is a GitHub project for a reason! This guide has a lot of shortcomings as it's very new, but our goal for this is to be a "living" guide, not yet another read-and-comment blog post. After commenting here, please consider filing an issue or PR with questions or suggested improvements. A community as talented as this one knows vastly more collectively than any individual authors. We can then discuss and work to get suggestions incorporated.

I've posted this in a previous thread about equity options, but I made this Google Docs spreadsheet that helps model the potential payout of various stock options: https://docs.google.com/spreadsheets/d/1L-hwCRXKDwmOPqXCwQo4...

You put in your equity percent and how the "costs" associated with it-- both exercise costs, and the "opportunity" costs such as smaller salary over X years, or unvested shares at current company--- and it spits out a grid of how much your options would be worth in various combinations of dilution and exit values.

I think it'd be a good addition to a guide like this but I'm not really sure how to best integrate something like this in a .md document or another page that will run inside github. Right now someone has to make a copy to their own Google account and mess with the inputs to use it, which is a bit cumbersome.

edit: went ahead and made a issue in GH for more detailed discussion rather than doing it here: https://github.com/jlevy/og-equity-compensation/issues/27

> Also, please remember this is a GitHub project for a reason!

Totally understand. I've upvoted your comment, hope everyone else does as well so that people can see this first before expending energy on a comment that should have simply been a PR (or at least a filed issue).

Thank you! Love HN discussions but also want to see corrections and comments work their way back to the original doc.

Thanks for putting this together!

This is a most helpful guide. Up until now I've never understood why RSUs are less favorable for taxation.

Josh, thanks for doing this.

Kudos for putting this together. The guide is long, but worthwhile if you're going to be in this situation.

The employer-employee relationship is generally relatively straightforward if you're being paid in regular income/on contract: It's easy to put a value on what you're being paid because that is part of the contract. When you're being granted equity, things get complicated because now you have an ownership interest or something that's economically similar to this.

In order to value this, you have to understand the capital structure and corporate structure of your company. This can get complex, as the guide shows.

Of course, even if you fully understand these principles, the value of your equity stake may not be certain because the value of your company is not certain; but it will help you answer the question: How much will my equity stake be worth if my company is sold for X dollars?

If you don't fully understand how your equity value ties to the company's value, you could end up like the employees of Good Technology:


As an addendum: An example of capital structure. (Based on debt capital rather than equity capital, but just an example)

You and Fred have a common friend, Bob. Bob wants to start a new business and needs to borrow money.

First, Bob borrows $10,000 from Fred and later on, Bob borrows $20,000 from you.

Some months go by, and the business isn't doing well; Bob (or rather, his company) eventually files for bankruptcy, listing $6,000 in assets. How should this remaining value be distributed among you and Fred, which both have a claim to it because of the debt owed to you?

One way to divide up the remaining amount would be in a pro rata method: Bob borrowed $30,000 total, 1/3rd of which was from Fred and 2/3rds of which was from you. So, you should get $4,000 and Fred should get $2,000. (This is known as pari passu I believe)

However, this isn't the only way things could be done. One could say that because Fred lent money first, he should get paid back in full before you do. Or perhaps one could argue the opposite.

You could argue back and forth about what the most "fair" way to divide up the assets is, but in reality what matters is what the rules outlined in capital structure specify. They should define who gets paid out first, which debt is subordinate to other debt, etc. (And even then, it's not always entirely clear, as evidenced by lengthy bankruptcy proceedings at large companies)

Although I'm from the Netherlands, I read these articles with great interest. Some wisdoms are transferable (e.g., equity being worth only as much as the company is valuated at right now, rather than what it might be in the future), but many also aren't, such as specific ways in which options, equity etc. are taxated. Are there any "guides for humans" out there that are specifically aimed at us Europeans? (Or, even more specific, us Dutchies?)

Or even us polite, frozen neighbours! It's in the news more lately because the newly elected government says it wants to change some of the tax rules around stock options.

Perhaps splitting the content into universal and US-only advice would allow other (more knowledgeable) individuals to build on the universal advice and give more location-dependent advice separately?

It's a bit messy to do this, but yes. See comment above.

We'd love to make it more general, yes. But it's a difficult task since taxation and law interact a lot with equity comp.

In the short term, it's just for the US, but this could change.

Probably the best approach would be for someone knowledgeable in European law to write an addendum in the same style for Europeans, listing info and differences specific to EU or specific countries. Then in the future, perhaps we could have linked subsections devoted to different regions or countries (and perhaps even split out the US). If anyone here would like to help with this, please file an issue to coordinate it!

The private company I work for issues stock "promises" on fancy certificates. Each certificate claims to be worth X shares for an estimated value of Y.

Clearly they are not stock options, nor RSUs nor anything fancy or official like that. They are just fancy certificates with unverifiable claims. Is there any legal or financial value to these things? Is there a name for these types of pseudo-stocks? Where in the linked article is this discussed?

LOL. I have a coffee mug that says "World's Best Dad" at home (it looks very fancy). Does that mean I am really the best?

Reminds me of Will Ferrell in Elf:

> You did it! Congratulations! World's best cup of coffee! Great job, everybody! It's great to meet you.

"You think you're better than me?" - Izzy Mandelbaum

"Is there a name for these types of pseudo-stocks?"


You need to talk to your own lawyer right away. That sounds extremely fishy to me.

Are they signed by both parties? It could be considered a contract, or at least proof of a verbal contract. It isn't talked about in the linked article because that is some weird made up thing that sounds kind of fishy.

The guide should cover more on this, yes.

Added an issue to track. Please comment there if you have more details. https://github.com/jlevy/og-equity-compensation/issues/31

As long as there's a contract behind it, it's just like any performance-based bonus.

I've heard it called shadow-equity or phantom-equity. Note that if it ever paid out, you would pay income tax (not capital gains).

My genuine stock certificates look like someone printed them out in MS Publisher, so don't assume the production values invalidate them. But as others have said, talk to your lawyer.

Not an expert, but have you asked your boss?

Monopoly money stocks

Stock options are super-complicated, there are many gotchas, and there can be serious tax implications. Like having to pay taxes on "profits" from exercising options before having the revenue from selling them, which may prevent options from being exercised. I think all these drawbacks often make stock options have near zero value in the eyes of employees. Which makes it not that fit for purpose, which is supposedly to motivate your employees and align their interests with those of the company.

Is there a better way? If you run a startup and you want to achieve the same ends of providing a form of compensation that's tied to the future success of the company, but doesn't involve immediate outlay of funds, how would you structure it? It should be as simple and straightforward as possible with a minimum of gotchas, and avoid tax implications as much as possible. I'm kind of surprised YCombinator hasn't innovated anything in this space, like they did with "the Safe"[1].

What about something along the lines of making a pool of X% of the gross revenue of the company each year, and dividing that among all the creative employees fairly, and adjusted by how many months of the year the employee was at the company. Then just paying that out every year as the end of year bonus.

[1] https://blog.ycombinator.com/announcing-the-safe-a-replaceme...

It's better for a growth business to use profits to grow, not as compensation. Plans like you describe are what typical profit-sharing bonuses look like in more mature companies (or ones that aren't planning to ever exit).

The easiest fix is to set the expiry after you leave the company to ~10 years rather than 90 days. In that case, you have your entire tenure + 10 years for the company to have some kind of exit.

If that were the case, you could simply exercise at the point when the stock was liquid and you would never have to pay tax before then. There would still be reasons to early-exercise (to get the clock ticking on capital gains), but that's something that incurs risk -- you never need to do it if you don't want to (and you don't lose the options if you don't)

Well, one could simply modify it to giving points to the employees and delaying the payouts until the company can afford it, then doing it by points so that those who've been at the company longer get more.

Anyway the details can be anything allowable by law, the question is can we do better than stock options. I feel the answer to that must be yes.

Many companies are never profitable before exit. Stock is already this point system where we call the payouts dividends. Growth companies don't pay dividends and they wouldn't want to pay point-payouts either.

If you want this, you can have it by joining a company with a profit-sharing plan. It's fairly common.

Typically movie stars negotiate a tiny % of gross revenue in their contract. They used to negotiate for % of profit, but studios got very good at fiddling with the numbers for the film until a successful movie had nearly no profit.

I don't see why that can't be done for software engineers. That is essentially dividends though, and it's obvious why growth companies don't want to do that. But it's also not clear that stock options have the intended motivating effect (it's not even clear that increases in salary or some kind of profit sharing would have the intended effect either[1])

I think it's clear that stock options are sub-optimal for the intended purpose, but it's not obvious to me how to fix that.

[1] https://www.youtube.com/watch?v=u6XAPnuFjJc

> If that were the case, you could simply exercise at the point when the stock was liquid and you would never have to pay tax before then.

Unless the AMT kicks in. Anyone with options needs to keep an eye on this. Consult a tax professional if unsure.

You do not have to pay AMT before exercise. The AMT income is on the difference between FMV and exercise cost.

There are risks if you don't immediately liquidate enough to cover your tax bill.

Altman has called for a change in legal barriers that would allow such an instrument to exist (currently not possible by law).

You can issue RSUs to your early employees if you're in the seed stage and incur minimal tax implications (on the order of several hundred dollars) and help your early team members through the 83b process.

In seed stage companies, what is the disadvantage of enabling your employees to early exercise and bonus-ing or loaning them the capital to do so?

The advantage seem clear: employees can get 25-35% more equity at effectively no cost difference to the company.

I don't know if there is a way to avoid this issue with some lawyering but loaning money for options was apparently somewhat common in the 90s tech bubble. [1]

Unfortunately what happened was that when the bubble burst, and many companies went under, the board and management resigned, leaving the company in the mercy of creditors. These creditors sometimes when after these employee loans and they had to pay them back while the stock they received was worthless.

[1]: https://www.fenwick.com/publications/pages/playing-with-fire...

Shameless plug: I'm an engineer at eShares, and we have been making a big push to bring transparency to employee compensation. Last month we published a guide for option grants, which includes information on taxes, vesting, expirations, etc. Another good source to help figure out what you are agreeing to.


What's the best way to find a tax professional that can help with one's specific situation? I see references to Teaspiller all over Quora, but it's apparently shut down :(

Any advice here?

Check on personal finance sites for references for Financial Planners or Financial Advisors. If you only want tax advice and not investing/budgeting/estate planning, that should be cheaper. (I looked on reddit.com/r/peronsalfinance, but didn't immediately see any listings of advisors.)

EDIT: removed redundant phrasing

Has anyone here had phantom stock options? I would love to see more details on how they tend to work, and if they have actually paid off for people.

From the guide:

"RSUs are less attractive than options from a tax point of view because you cannot make an 83(b) election with respect to an RSU."

I'm fairly sure this is incorrect. Every other guide that I've read (just google RSUs 83b) says that you can file an 83B with RSUs and I've personally filed 83Bs with RSUs multiple times.

There's a difference between "restricted stock" and RSUs. The first means you get all the stock up front but it has a buyback provision that goes away over a vesting schedule. The second means actual shares are released over a vesting schedule. Only the first is relevant to 83b.

See for example http://www.investopedia.com/articles/tax/09/restricted-stock...

I am fairly sure that you are incorrect. There are no consequences to filing an 83B with RSUs - you can send it to the IRS, and you will get no response back; they will not send you a letter that says "you did wrong".

The question to ask in this case is: later in the life of your employment with the company, when settlement happened and you received stock in exchange for your RSUs, did you owe tax immediately on that stock? If you did, then you filing an 83B was meaningless.

I always get responses back saying that the 83B was accepted.

Thanks for this.

This is the first time I've seen a github used for non programming purpose

It has happened before. Some examples are:

- German federal law and regulations: https://github.com/bundestag/gesetze

- DMCA takedown notices: https://github.com/github/dmca

- Map of US districts: https://github.com/benbalter/congressional-districts

- a novel: https://github.com/JJ/hoborg

- a well-known textbook on some niche (homotopy type theory) in mathematics: https://github.com/HoTT/book

- fonts: https://github.com/theleagueof

- Gregorian chants: https://github.com/CMAA/nova-organi-harmonia

Main collaboration functions on a 'repository' (or any set of data / documents that somehow belong together) such as forking and branching are not just only relevant to programming.

(edit: formatting)

Also a couple of my favorites:

- Remote job information: https://github.com/lukasz-madon/awesome-remote-job

- Liberland constitution: https://github.com/liberland/constitution

I think GitHub is being more popular among "non-developers" because version control is beneficial for all sorts of projects.

My favorite example of a "non-programming" project is Hadley Wickham's Advanced R book (http://adv-r.had.co.nz) which is hosted on GitHub and powered by Jekyll, knitr, and pandoc. I hope more authors decide to publish this way.

Github has been energetically chasing non-developers for the last couple of years:


We host our open source HIPAA policies on github https://github.com/catalyzeio/policies

Well then you're not looking very hard. It's use as a free, open markdown editor is pretty common.

Nice. Thank you.

Now is there anywhere a tool to make this readable after print?

A technique I've used in the past that usually works well:

Right-click the page and hit "Inspect Element". Scan through elements, looking for the ones that only refer to the header or footer stuff you don't want. Click them and hit "delete" on your keyboard.

Find the container that defines the width of the content you're viewing. Use the style editor to reset it to 100% (often, it's just disabling whatever width setting they're using, rather than explicitly setting it to 100%).

You should be able to CTRL+P from there and get yourself a PDF out of it. Then you can do whatever you want to the PDF. I think you could also just save the raw HTML, and the browser will preserve your edits.

This is a brilliant idea. Thank you.

If you are using Firefox, you can also use the Add-on "HackTheWeb" [1] that automates the suggestion by moron4hire.

Right-click on the page, and select "HackTheWeb". Whatever page element you hover your pointer over will be given a red border, and then you use hotkeys to perform actions.

To do the same thing as the example above, it would be: right-click on page, select HackTheWeb, point mouse at the content you want, hit "w" (widen) until you have the whole section you want, and then "i" (isolate) to remove everything else. From there you're ready to print to PDF.

For some reason I had to be viewing the actual README.md file, not looking at GitHub's auto-readme-display.

HackTheWeb also lets you view the element's javascript, show the xpath of element, and convert text to black on white (very useful for content whose text and background colors don't work well together).

I've used it for several years, and wouldn't want to give it up. It's based on the old Aardvark add-on that did the same kind of thing, but Aardvark wasn't updated as Firefox matured and eventually no longer worked.

[1] https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&c...

This is markdown file. I did not try but using pandoc you should get a nice printable pdf.

> pandoc -f markdown -t latex -o output.pdf -V geometry:margin=1in readme.md

The geometry stuff is too avoid too big margin

I spent $10k to exercise options last year. What are the chance that I at least break even?

If your Twitter profile is serious and you work for a startup that sends PB&Js to people, I'm going to go with 0.

Put $10k in a Vanguard index fund and leave it alone.

Source: https://twitter.com/AirPBnJ

impossible to answer without knowing any other information.

Are there general statistics? Like x% of companies don't pay back their investors.

Impossible to know without knowing anything else about the company. :-)

Here's the thing: it's incredibly difficult to say that. Most investors build themselves a nice legal framework to make sure that they are very likely to get some amount of money back.

Regardless of the company, investors most likely made sure that they get some sort of preference, whether it's in anti-dilution preference, liquidation preference, etc. If the company tanks, or even just doesn't live up to expectations, investors will most likely get some sort of compensation (maybe less than they invested, but it's unlikely that they'll lose 100% of their investment).

Then there's the question of even if things go reasonably well, is it worth it. First, you have to address the time value of money, which basically is a question of what else you could have done with the money in the meantime.

If you get $100k after 5 years at a company, that sounds like a good chunk of money, but is essentially $20k/year before taxes, and the tax implications that will likely be worse if you get $100k in a windfall vs. compensation.

Then consider that even if the company does "well", there's a decent chance that employees won't see nearly as much money as they expect. One question I've wanted to ask for awhile is what happened to employees at Loopt, Sam Altman's company. According to CrunchBase, they raised $39m+, and sold for $43.5m according to Wikipedia. If everything converted to common stock, employees with 0.5% would get $217k (before taxes).

However, depending on investors, there's a good chance that employees got nothing. Outside of dilution, there are a bunch of other things that can really ruin your "investment": the preferences mentioned above, cases where their initial investment is returned before the profits are split, and multipliers (investors are guaranteed a multiple of their investment in return).

So there's no real easy way to say what percent of companies pay back their investors. Colossal failures can still make money for investors (via the multipliers mentioned above), even though employees who exercised their options are screwed (they bought their shares AND paid taxes on them).

Applications are open for YC Summer 2023

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact