Palantir Buyback Plan Shows Need for New Silicon Valley Pay System
nytimes.com
nytimes.com
From NYTimes:
> He [Scott Kupor] also suggests a longer period for employees to exercise options after they leave, up to 10 years. That figure is endorsed by Y Combinator in an argument that any lesser period is unfair to employees.
Makes it seem like Scott Kupor is on the leading edge of caring about employees, in agreement with YC (which has actually been employee friendly in words and actions in regard to stock options).
However, read Scott's actual blog post and he refers to the 10 year exercise idea incredulously:
> The 10-year “solution” thus takes money/option value out of the pockets of the current (and growing) employee base to line the pockets of former employees who are no longer contributing to the business.
> Talk about disenfranchising your remaining employees and not being able to attract new ones.
Good reminder to always read your primary sources.
Stock options seem attractive as a form of compensation, since the hope is that you're working for a unicorn and one day you'll be fabulously rich, but the cash equivalent is better more often than not. It's just a form of risk transfer from those who have plenty to those who hope to have plenty by dint of talent and hard work. Having been burnt, I won't do it again.
There are a few problems I see here: 1) stock option grants are completely arbitrary and sometimes end up very wrong 2) it's hard to fix that in the future because you'll end up at a higher strike price 3) end up being expensive and tax inefficient to exercise
The closest I've seen to people who seem to get this and have sensible solutions are Andrew Mason at Detour (progressive equity) and Dustin Moskovitz at Asana (larger grants, but back loaded into years 4-6).
I have great respect for Adam D'Angelo at Quora for suggesting a solution to the problem, have known him in school he's certainly smarter than me on almost every axis of intelligence, but I think there are other potentially creative solutions that might be better (although I don't know tax compliance).
For example I think you could keep the status quo, but offer the option for employees to exchange their options for shares (white meat) at the time they can exercise. Example you have options for 100 shares at a strike price of $50, at the time you leave the shares are worth $100, instead of having to come up with $5000, you just get $50 shares free and clear. I think there's still a tax hit issue, but at least it's not doubled with paying for the shares.
In fact, I was pointing out that the poster I was replying to's complicated scheme was actually an "end-run" towards paying way too much in taxes.
They already do in the form of health and retirement benefits.
Retirement isn't really 'tied to an employer', in that you can still open an IRA without an employer[0], or use a non-tax-advantaged account for retirement savings (most people outside the military or government service use non-tax-advantaged accounts for at least a portion of their retirement, since the IRA and 401(k) contribution limits are too low for most people to survive on during retirement).
This might have been different 50 years ago, where employer-driven pensions were more common, but today, the only real way your employer impacts your retirement is the 401(k).
The purpose of both the IRA and the 401(k) is to provide people with an extra incentive to plan for retirement. Putting away $450/month towards your retirement[1] can be unpleasant, but if you're getting, say, $90 that back (in the form of lower tax withholdings/taxes due), it makes it a bit easier, because that's effectively only $360 out-of-pocket.
The incentives work similarly for the 401(k), except the tax savings work out for the employer as well, meaning that they are incentivized to give you some portion of your compensation in the form of 401(k) matching (ie, they have an extra incentive to nudge you towards saving more of your own money for retirement).
Personally, I do believe that, if you do not have access to a 401(k) through an employer, your IRA contribution limit should be raised by $17,000 (which is the 401(k) contribution limit for individual contributions). But without employer contributions, at most that's saving you less than $6,000 - and that's if you're already at the very top marginal tax brackets (even making $100K gross in NYC, the most heavily taxed jurisdiction in the country, won't be taxed at 35%).
[0] Well, you can't contribute more than your total annual income to an IRA, but if you're making less than $450/month and living in the US, retirement planning is not your most immediate problem.
[1] ie, enough to max out your IRA contribution limit
Either the former employees were paid full market rate salaries during their employment, in which case the company colossally fucked up by giving shares to them for literally no reason, or those former employees made a capital contribution just like any other investor.
Having said that, I think he's over-applied this mentality to regular employees with much smaller stakes. It is reasonable to expect a founder to stay 10 years with a company. Not most employees.
According to crunchbase they're basically owned by a private equity firm now (which is rarely a fun place to be) and are raising something like a billion dollars a year -- which basically appears to be around what their operating costs are (employee count of that year * $250k/yr).
They're either not bringing in any real revenue, or growing at the rate of revenue. Multiple raises per year (of weirdly different values) indicate frequent requests for more money.
Are they growing or are they dying? Either way they aren't doing it through revenue, and they're not going public so the financials stay very hidden.
Edit: forgot noncompete clause.
When an employee gets something "in kind" for the non-compete agreement, which arguably this is, then it may be upheld. For example, if a company pays you non-salary money in exchange for a non-compete agreement, then it may be upheld. Consider the case of former HP CEO Mark Hurd when he went to Oracle.
What's not enforceable in CA are non-competes in, for example, normal employment contracts that apply to everyone.
Edit: To be clear, Mark Hurd was allowed to work at Oracle, but he had to give the money back he got for signing the agreement.
They are definitely bringing in real revenue. Many firms have raised substantially more than they need to, because they have been well aware that funding could dry up.
They've raised 15 rounds that are publicly known. Every single raise since Sep 2013 has been under private equity. This sounds more like the company is being sold on the private market and each new owner is putting in a year's worth of cash infusion for operating costs.
Are those fund raising rounds? I dunno. But it's one of the weirdest fund raising profiles I've ever seen.
One reason for this not mentioned in the article is that in the US the tax burden is extreme - partially because when it was implemented it expected companies to go public.
If you hold options in a private company you get taxed on the exercise of those options based on the fair market spread which is the difference in price between your original strike price (the price of the options when they were granted to you) and the current fair market valuation. This is taxed as income.
This is problematic since once exercised you're holding shares of an illiquid asset (since the company is not public) and they're difficult to sell. This means even if you save up enough money to exercise your options you'll get hit with a potentially enormous tax bill due that year that you can't easily sell your newly exercised options to pay for. Additionally when you sell the actual shares after you've exercised them you get taxed again on the sale.
The one exception to this is if your options are ISOs (incentive stock options) then the delta between the strike price and the fair market value isn't taxed immediately, but it does count towards AMT (Alternative Minimum Tax) and it's fairly easy to hit the AMT while exercising options (meaning you could only exercise a tiny amount per year tax free).
All of these things make it extremely difficult to realize any value in a private company without enormous amounts of upfront cash and also losing roughly half to taxes. It also makes it extremely difficult to exercise options outside of a liquidity event. This can also make it hard to leave a company since the agreements are often 90 days to exercise after leaving or you lose your options (there's also usually a ten year expiration date).
If companies in SV intend to stay private and don't want their employees to view the options as impossible to liquidate we'll probably see an uptick in liquidity events like this one. The companies that value their employees will probably figure out a way to make this work.
The 409A valuations are real and there are rules surrounding how exercise happens. You can't just sell them outside of that and not pay taxes. You could theoretically sell the shares once exercised to some other private investor if you can find one, but you'd still have to follow the same exercise rules.
That's why shares that trade openly are called 'public market' shares.
Reading this thread is fun, it's like trying to see Silicon Valley try to reverse engineer something hiding in plain site.
95% of these problems would go away if these companies were forced (by investors or employees) to go public.
One clarification with what you said, is that with the AMT ISO exercise is that you're not actually taxed on both the exercise and the sell. What's actually going on is that you're prepaying your taxes when you sell the stock. When you sell the shares, you'll only have to pay the taxes (either regular income or capital gains) based on the difference of the fair market value of the stock when exercised and when sold. If it went up, and you sold in less than year from exercise or less than 2 years from ISO grant, then it's regular income, otherwise it's capital gains. So you could actually get a tax refund when you sell. (Same is true if the market price actually declined between exercise and sell.)
The argument is that when you exercise, you received something of value for less than market and so you made money, but in reality you actually haven't realized any gains, and actually are at cash loss. I understand the argument, but I don't agree with it, because you did not actually realize any gain.
FWIW, San Jose's congresswoman Zoe Lofgren has repeatedly tried to fix the AMT and ISO taxation, but hasn't had much success.[0]
Where I disagree with you is thinking that private companies are going to "figure out a way to make this work" in a way that's beneficial for workers. I'm sorry, but I've never seen high finance work out for workers. It's basically a play for the financially desperate. It's no better than selling you shares on sharespost or something. If it's illiquid market, you're never going to get full value, and you know damn well those that are buying are going to expect a few multiples in gain. They can just wait a bit longer. Workers on the other hand, are busy trying to scrape together a down payment on a $2,000,000 shack in the valley.
[0] https://lofgren.house.gov/news/documentsingle.aspx?DocumentI...
I agree the argument that you make money on exercise doesn't make sense - especially since the company could easily crash afterwards and you can still get stuck with a huge tax bill for value you never actually realized (except on paper).
Agreed you won't get full value, but I think the existence of liquidity events like this one is an example of private companies "figuring out a way to make this work". They're letting employees realize some value from their equity.
Nothing can be done about the 2 million dollar shack in the valley though (unfortunately).
One other option sometimes available that I didn't mention is filing an 83b election with the IRS and exercising options early before they've vested and before there's a fair market spread. You can avoid taxes this way, but you're putting money at risk very early and often you're not able to do this anyway (there are rules about early exercise).
Can you sell it?
If not, it's not really "worth" $100. That's the difference.
This is a bit misleading. The same income isn't taxed twice. You'll only be taxes on additional gains that accrued since the time of exercise.
AFAIK, this is a response to Facebook employees selling equity on private markets.
> [...] employees who sell their shares agree agree that they will not compete with Palantir for 12 months or solicit any Palantir employees during that time [...] [and] agree to a nondisclosure arrangement that forbids them from even talking about the repurchase and waive any claims they might have against the company.
The nyt put forwards a reasonable idea that this is a mechanism to increase their perceived value, and I'm inclined to agree.
EDIT: the buzzfeed article [1] paints an even worse picture:
> If they [employees] get any inquiries about Palantir from reporters, the contract says, they must immediately notify Palantir and then email the company a copy of the inquiry within three business days.
[1]: https://www.buzzfeed.com/williamalden/palantir-seeks-to-muzz...
Perhaps you are more aware of the terms on those shares than the rest of us, but generally you aren't quite "free" to sell the shares on the private market.
https://en.wikipedia.org/wiki/Non-compete_clause#Exceptions_...
Palantir's attorneys surely can't have missed this, but I don't see an alternative.
I would bet they don't apply that to existing employees, but they'll give it a shot on anyone hired since the change.
If this turns into a trend, it will mean a very interesting change to silicon valley culture. I read a blog post a couple of days ago that the average tenure for a software engineer at a medium to large company in SV was 1.5 to 2 years. With an effective way to push non-compete agreements on employees, that would be... significant. It would take a lot of companies doing it at once, though. Otherwise everyone would simply avoid the handful of firms that did it. After looking at this summary, I wouldn't work for Palantir. If I did, I'd have to consider the stock completely valueless.
That smells funny and oddly specific.
Where it may make sense is in so-called "private IPOs", i.e. those 9 or 10 digit dollar rounds. There's enough money there to hand out. In olden days, companies would have been public by then and employees would have had liquidity. Planatir raised $880M last year so, yeah, they can afford it.[2]
[1] http://venturebeat.com/2012/09/30/need-to-cash-out-a-bit-pre...
[2] https://techcrunch.com/2015/12/23/palantir-has-raised-880-mi...
Edit: I don't know for a fact that they are coworkers in a meaningful sense, of course a university is a large place, but I would expect a professor of law to have better reading comprehension than was on display here.
It does open up some creative financing though. Imagine a "common only" startup, one where every share bought by investors or earned as an option had the same rights and liquidation preference. Or my favorite[1] starting a company with a fixed 5 billion shares all at a par value of $1 each. Then only pay your employee in shares, only get investments by selling shares to investors.
[1] I call it a favorite since we joked at a party once that if you did this, and convinced a friend to buy a hundred shares for $100 you could call yourself a self made billionaire! And be completely truthful.
This is simply a falsehood. Exercises of options of do not dilute shareholders -- new issuances do. Former employees do not issue new shares.
On a side note, you have been defending Kupor in the other thread too with such nitpicks. Why?
Now they know that hundreds of employees are going to go blow the whistle so they have to pay them off with buybacks while they figure out an exit strategy.
My 2 cents.
The true solution is to give stock options that can be exercised early as long as the value of the stock is very low such that an employee's hiring bonus after tax could cover the cost. There is no tax owed by the employee since he purchased shares with no gain and then you vest outright stock. Once the stock value goes up it would be best to grant RSUs of convertible notes that convert into stock.
Citation?
It is known they will not likely go public as it would be detrimental to their business and in-Q-tel (cia vc arm) is a major shareholder. While Silicon Valley does need to rethink some of the ways they compensate employees; especially at late stage private ones, I would not consider Palantir indiicative of a typical SV unicorn
* left palantir in 09
* is a major shareholder
* left after he "had successfully replaced myself in both parts of the company we'd created and had fully vested".
So I am not going to speculate whether he was pushed out, but it sounds like he has much less latitude for steering the company. Especially when you consider the power of the other founders and the initial VC capital that was put in. So this (claims of palantir going public) are what I would expect a major shareholder without control of the company to say. It is effectively the only way to put pressure on them.
Even if this is wildly incorrect, and it quite possibly is, that comment was made several years after he had little more than an advisory role at palantir.
Go figure.
Yes, it's an offer. I don't see why everyone is so up-in-arms over an offer. People can reject offers just like they can make them. Other similar offers include employment offers. I'm going to comment more generally on this issue between founder and employee deals:
Here's what I think. HN is mostly of the employee class and so there's a politicized negative sentiment about companies not leaning the way of the employees.
Or maybe people think these offers and deals are bad for both the company founders and the employees. So here is my proposition if you believe that. Start your own company and enact whichever agreements you think are best. Do what Palantir does, or what YC advises, or whatever you make up. Give a 20 year exercise period if you want. Your call.
But you have to actually found a company.
I find it hard to believe in a moral "good and bad" on this issue. We're just talking about deals and contracts between people. And if people act with agency with regards to accepting and declining offers, and inventive individuals can come up with new systems and agreements that work better for everyone, then it will be fine. A moral bad would be something like Google and Apple and Facebook colluding behind closed doors to keep engineer pay below a certain threshold.
For the employees who will have to negotiate: you can't get a deal that's good for you if you aren't prepared to walk.
Either way I'm very skeptical of anything involving NDAs or non competes. I can understand a non poaching clause (but am also opposed to that on ideological grounds). If I could afford it I wouldn't sign anything that has NDAs or non disclosure.
I also doubt that many of the ex-employees need liquidity as they are likely to hold well paying jobs. I suppose some could need it but those are probably exactly the people least likely to compete with Palantir.
I could get behind the scheme you commented with, but I'm unsure of generally why we have the system we currently do, so I would want to understand the rationale for the status quo.
Further, you don't think businesses are averse to golden handcuffs, do you?
H.R. 4351 also includes an Incentive Stock Option (ISO)
provision that corrects a severe inequity in the AMT tax
system and finally resolves the ongoing AMT Incentive Stock
Option crisis. This essential provision provides relief for
hard working Americans who are treated unjustly by the tax
burden caused when exercising their Incentive Stock Options.
Under this important provision, employees will pay a fair and
proportional tax on money actually made and will be relieved
of the impossible burden of ongoing liability to pay taxes on
income never received. This provision will help ensure that our
companies continue attracting the best and brightest with
competitive compensation packages.The only thing that I would count on an offer in the future are RSUs in a public company.
If you ever get a palantir job offer, you'd be better off just forwarding it to Uber, or Google or FB, and waiting for them to give you a 50% increased counteroffer on the spot.
ASAIK, Goog/Uber/FB starting salaries are closer to 110k base with $40-50k RSUs vesting per year (refreshes each year), performance bonus 10-15% salary and sometimes a generous signing bonus.