Hacker News new | past | comments | ask | show | jobs | submit login
Lies, Damned Lies, and Stock-Based Compensation (tanay.substack.com)
170 points by pvsukale3 40 days ago | hide | past | favorite | 46 comments

So people have a tendency to read a headline/submission title like this and without reading the article they launch onto their soapbox about their pet issue like, for example, equity compensation at startups should be treated as being worth $0 if the company is not listed.

The article isn't about that. It's about companies misrepresenting their expenses by not accounting for stock-based compensation ("SBC") costs, which is completely fair. Google and Facebook (quoted in the article) do. Others (eg Workday, Splunk, Okta and Atlassian are quoted) seem to muddy the waters by stating they're unprofitable on a GAAP basis (which includes SBC since 2004) but profitable on a non-GAAP basis (where SBC isn't treated as an expense, I assume?).

So, caveat emptor for investors, basically.

are you cletus from stackoverflow lmao you used to help me so god damn much back in 2008 during my college years when i was writing c#. small web!

Yep, that's him.

Here's a thought experiment that helped me reason through SBC and GAAP.

Scenario A: Company hires an engineer with a base salary of 100k/year, and 100k/year worth of SBC.

Scenario B: Company convinces an investor to invest 100k/year... and also hires him as an engineer with a base salary of 200k/year.

From a business fundamentals and margins perspective, the two scenarios are identical. And yet, in scenario A, the company is spinning their non-GAAP annual expense as being 100k. Whereas in scenario B, the company wouldn't even try to spin their annual expense as being anything other than 200k.

The above thought experiment becomes particularly powerful if the hypothetical company has no other expenses and an annual revenue of $150k. Using the non-GAAP estimates from scenario A can mislead investors into thinking that the company has a very healthy gross margin, and is a lucrative investment. Whereas the actual GAAP numbers from both scenarios A and B, make it clear that the business is not profitable at all.

Good explanation. I'll nudge it even more with a Scenario C to further illustrate that ignoring non-cash comp leads to an distorted picture.

Scenario C: Company hires an engineer with zero base salary and $200k/year worth of stock that gets paid monthly with the correct number of shares to get $16.7k in value. The shares can immediately be sold back to the company for the full value.

The engineer clearly receives $200k of economic value. Whether the stock is sold or not has no effect on the value transfer.

Does this hypothetical company have 1) a 100% profit margin, or 2) is it losing money?

If you picked 1), let's say the engineer quits and for whatever reason the company needs to pay the replacement $200k/year in cash. To do so, the company sells $16.7k of shares to an investor each month. Is it still a 100% margin company?

If you can convience employees to trade 100,000 cash for stock then is it really an expense or a risk shifted to the employee?

It read to me like they were intentionally ignoring details like the risk adjusted value of the SBC, not because they don't matter, but because they obscure the point being made that employee payment schemes can be used to sweep unprofitability under the rug.

If they are telling employees that the stock is worth $X and then accounting for it as costing $Y that sounds remarkably close to fraud.

Isn't there also Scenario C, where the investor invests $100k, employee gets a base salary of $100k and SBC of $100k?

You've now got an extra $100k in the bank for other requirements. Effectively the SBC makes the employee an investor also, exchanging labour for equity (instead of just for cash). This seems to me to be different. Obviously there is some dilutive effect to the share allotments, but especially in earlier stage non public companies you probably always want to have more non cash vs cash expenses if you have the option?

That is identical to Scenario B - if the stock is actually worth 100k then they could have sold it for $100k and given the employee $200k as a base salary, while keeping the investor's money in the bank. Obviously there are some tax complexities in practice but in theory it balances out.

Thanks, I think this is convincing (where most of the arguments in TFA, especially the Buffett quote) were pure equivocation.

As an outside investor, if a company is not profitable accounting for stock compensation then the book value of your stake (or your share of the NPV of future profits, or whatever) is going down.

>>[...] especially the Buffett quote [...] were pure equivocation

I interpreted TFA to cast the Buffet quote in a positive light, and not as an equivocation.

>>As an outside investor, if a company is not profitable accounting for stock compensation then the book value of your stake (or your share of the NPV of future profits, or whatever) is going down.

Sounds like you, Buffet and the FASB are completely aligned on this point -- all the more reason to use/weight the GAAP numbers more heavily than the non-GAAPs and adjust our decision models appropriately.

Right -- I thought TFA's conclusion was correct, but the argument was unconvincing.

I did not read the article - but that seems quite obvious - so the interesting thing would be the other side 'why SBC was not in GAAP expenses originally?'.

I think because the GAAP metaphor is that revenues and expenses determine profits, which accumulate into equity, which is claimed by equity holders according to fully diluted shares. SBC was "supposed" to be accounted for by calculating dilution of the newly issued shares. The expense approach is basically a shortcut for analysts.

From a business fundamentals perspective, the two scenarios are completely different. In the second scenario, the investor turned employee has invested $100k cash. This is an actual liquid asset that does not exist in the first scenario.

Business fundamentals are concerned with cash flow (which the investment does not change because it’s not part of income/expenses).

You’re talking about the balance statement, which it would benefit but unrelated to profit.

In other words, yes they have more money, but no, it does not make them more profitable.

Business fundamentals are concerned with cash flow and the balance sheet. It's irrelevant to have positive cash flow if it can't service your debts (see, e.g., almost every company destroyed by a private equity firm, for example, Toys R Us).

In other words, yes they have more money, but no, it does not make them more profitable.

They're not profitable in either Scenario because they don't have any income in either. But in one scenario, they have $100k spendable cash, and in the other all they have is sweat equity that they claim is worth $200k compensation. With spendable cash, you have runway and the opportunity to acquire external products or services that they need for the business. With sweat equity, you have sweat and dreams of success.

I don't see why it's not the same (Company gives $200k, issues shares, gets back $100k).

2nd point: GP said "invest 100k/year", thus he do make additional investments.

Because a liquid asset is different from a liability. And the transactions have different tax, legal, and accounting outcomes.

At the simplest level, you can buy things and services with the $100k cash in Scenario 2 at the start of the year as soon as the investment is made. You don't have that cash in Scenario 1.

Those scenarios aren’t remotely identical.

Consider the difference if the employee quits or is fired before vesting their options.

Consider the difference if the company is sold for less than the strike price of the options.

A lot of FAANGs are moving to using RSUs rather than options and eliminating the vesting cliff. They really are identical in that case.

Then again, the companies that are doing so are generally the ones using GAAP earnings and including SBC. Companies still using options are often private, which means that Scenario B isn't even an option for the general public, and the only way for a non-accredited investor to own stock is to work for them.

We should be glad that public companies are forced to comply with FASB and issue GAAP financials that (since 2004) mandate that stock-based compensation be classified as a non-cash expense. By definition, non-GAAP figures are up to the company to specify and state, which indeed means that investors should be actively updating their own models when making investment decisions if looking at non-GAAP.

> “ On February 28, 2017, the Compensation Committee of the Board of Directors of the Company (the “Board”) granted Mr. Hu a time-based stock option for 900,000 shares of the Company’s Class A common stock vesting over four years, three performance-based stock options for an aggregate of 555,000 shares of the Company’s Class A common stock, each with a per share exercise price equal to the closing price of the Company’s Class A common stock on the date of grant, and a time-based restricted stock unit grant for 100,000 shares vesting over four years. Each equity grant is subject to the terms and conditions of the Company’s 2016 Stock Option and Incentive Plan (the “2016 Plan”) and the applicable form of award agreement thereunder.” [1]

I’m surprised Twilio isn’t listed. Just there COO alone was issued ~1.5M shares. At today’s market price ($208) his shares alone are worth $312M. Their annualized revenue is ~1.4B. So just their COO alone was issued SBC of 22% of the companies revenue.

And that doesn’t factor in the SBC of all of the other employees either.

[1] https://www.sec.gov/Archives/edgar/data/1447669/000110465917...

Good catch about Twilio missing from the SBC as %revenue chart. Without digging into Twilio's 8-K's I feel compelled to note that options would be expensed as each tranche vests. So as a very rough estimate the COO being issued ~1.5M shares over 4 years wouldn't be 22% of annualized company revenue but perhaps something on the order of 6%, notwithstanding the vague language about the performance-based options conditions and vesting schedules.

>>And that doesn’t factor in the SBC of all of the other employees either.

You're right -- that probably pushes Twilio's SBC %revenue higher to the Salesforce line.

You're misunderstanding options vs shares here.

He was only given 100,000 shares ($5 million/year).

The options only give him the right to purchase shares at today's price in the future. If the stock goes up substantially, he can profit on it, but the options could also be worthless if the stock price remains flat or goes negative. If the stock goes up 10%, the options would be worth ~$29 million (option to buy 1.4 million shares at a price of 208 then sell them at the future market price of 228)

It's not cash, so where does the money come from? Isn't the answer "investors"? Wouldn't their decreased share value from the dilution be the source of "value" used to pay these options to employees?

Exactly this. I find the article unconvincing as far as its conclusion is concerned. At the end of the day, shareholders pay for the stock-based compensation with dilution. There’s no expense to the “company”.

As a thought experiment: say someone works for $0 in salary and 100 shares of Worthless Corp. After a year of work, our employee attempts to sell their shares, only to find no buyers. Unsurprisingly, their shares of Worthless Corp are worth $0. Did Worthless Corp incur an expense somewhere?

As far as I can tell, Worthless Corp received a years worth of work at no expense.

It's a bit weird to try and conceive of the corporation without thinking of its actual owners. They're the ones who the accounting is ultimately for.

If Worthless Corp had 100 shares outstanding before this employee, the expense was half the company - whatever the valuation ends up being.

This is not the same as "nothing", and accounting should at least attempt to reflect that - such as by placing an estimated market value on the shares.

That's only true if it happens to be true for a specific scenario. If giving some new employee 100 shares (doubling shares outstanding) actually increases the value of the company by at least 2x by some measure, then the investors holding the original 100 shares should be happy.

Regardless, this is a silly, contrived example. No public company is minting anywhere close to 100% of their total share count every quarter in SBC. It'll be a fraction of a percent, probably? And investors shouldn't care, as long as the company is performing well at metrics that actually matter: acquiring paying customers, where the cost of that acquisition is less than the new customers spend. That, and things like efficiency improvements that cut costs, are the only things that actually matter, because those things are what drive stock prices up.

Half a percent per quarter means you've given away a quarter of the company in 14 years. That's not something an accounting rule should allow you to just hand-wave away.

> They're the ones who the accounting is ultimately for.

I absolutely agree with you. I just don't agree with the conclusion this article reached: that stock-based compensation must be accounted for as an expense that affects income because, in the case where shares are issued via dilution, it doesn't affect income.

Is there a better way to present this to owners than GAAP vs. non-GAAP? I think so. Have I given it much thought? Not really. I just don't agree with including dilution as an expense that makes companies appear as if they're losing money hand-over-fist.

I'm open to being convinced otherwise, but this article didn't do it for me.

> If Worthless Corp had 100 shares outstanding before this employee, the expense was half the company - whatever the valuation ends up being.

> This is not the same as "nothing"

Well, in the example, the valuation ended up being "nothing", and it seems fair to say that half of nothing is indeed the same as nothing. The owner in that scenario can retake 100% equity by paying nothing for the outstanding shares.

Can companies just dilute or do they need to buyback shares from the market for this stock based comp? The latter does have a real cost.

That's the crux. :) I don't have an answer for you, but that seems to be why the different sides are talking past each other in this discussion.

If a company can just dilute, there's no real tangible expense outside of the company transferring value from the shareholders via dilution. If a company has to buyback shares from the market, then I absolutely agree that stock-based compensation needs to be recorded on the income statement as a real expense.

That's my feeling as well. If I'm a publicly-traded company, and I give one of my employees a bunch of shares, I haven't spent any money to do so. I just don't see the function of the term "non-cash expense"; if it's not money, then how is it an expense? Sure, you can try to value things like "goodwill", but I don't think that's where we're going with this.

As an investor, sure, I'd want to know how much the company was diluting me every quarter by issuing new shares, but the only thing that materially affects me is the absolute value of the stock price when I want to sell (and the value of a dividend, if they have one, which most of the referenced tech companies don't).

The stock price doesn't need to reflect anything about the level of SBC. As long as people are bullish on the company's prospects at actually continuing to make real money from more and more real customers, then the price should go up, or at least not decline.

In the absence of dividends, stock prices are just a speculation game. If I'm not actually getting a share of future profits, it's just about whether the company will perform well enough (for whatever definition investors feel like that is) to increase speculative buyer demand, which increases the price.

Put another way, SBC doesn't affect money in the bank, or cash flows. A potential customer is not going to be turned off because a company gives out too much stock to its employees. It doesn't affect COGS, capex, or what the company can charge for its product.

I get why GAAP requires disclosing it, as it's a material fact regarding how the company runs its business. It's a risk: if the stock does not perform well, talent that is depending on that stock price for future income may bail, which could then hurt the company's revenue; the amount of SBC gives a window into how bad that problem could be. But docking it as a simple expense makes no sense to me.

The wild thing is, if ones accepts modern financial theory, from the eyes of a risk-neutral investor, stock-based compensation is actually much more costly because of imbedded option value and time value of money.

Say one needed to hedge the other side of a 4y employee stock grant. Let's assume the new cliff-less, monthly vest structure that is now market at some of FAANG. The counter-party would need to borrow a large amount of money to buy some fraction of the shares that the employee is likely to vest based on historical data. This also assumes they are just "delta hedging." There is definitely a "negatively convex" situation where if the stock price increases employees are less likely to leave and if it goes down, employees will find a new job that pays market.

Given the immense volatility in earlier stage companies, the counter-party may elect to hedge the gamma exposure as well, meaning they may need to buy calls in the open market against the RSU position.

In my opinion, Netflix has realized this, and given the implications decided to just pay cash. Dollar for dollar, to a well enough capitalized employee, the stock package is much better. This advantage increases with the volatility of the underlying asset. This completely ignores the career risk of working for a failed startup, but the culture in SV seems to minimize that.

Not sure I get this; who is this "counter party" you are referring to? My understanding was that the company doesn't actually buy shares on the open market when an employee vests RSUs or exercises options, but instead they either have a pool of shares waiting around (which they've never sold before), or they just mint new shares (and dilute existing investors).

For the pool structure, sure, there's an opportunity cost (the company could instead sell those shares on the public market). But in neither case does the company have to go out and spend money to buy up shares.

In modern financial theory valuation = cost of replicating the position. One can approach that cost from how the company or the employee would replicate it. My point here is that it is much more expensive than just paying the employee cash due to the optionality and the financing costs associated with such.

There is nothing that forces companies to fix the amount of stock ahead of time-- they do so for the retention benefit of triggering loss aversion. If that benefit wasn't worth the cost, they could drop it.

For mature companies it's pretty simple, stock-based compensation can be valued using the current share price and standard accounting principles.

For young companies, employees typically believe that their equity is worth more than the fair market value. So if anything, the non-gaap income should use a higher number for stock-based compensation expenses.

> For young companies, employees typically believe that their equity is worth more than the fair market value. So if anything, the non-gaap income should use a higher number for stock-based compensation expenses.

That's backwards, no? If employees are delusional and over-value the stock, they'll be willing to accept less of it and be just as happy, which reduces the amount of the "expense".

Right, and this reduced expense is the one that is in fact reflected in the GAAP numbers. But it might be useful for investors to know how much salaries would cost if they had to be payed in cash, which would be a higher number.

what’s the issue with this as long as the investors are also provided with the GAAP numbers? it’s not like the compensation is hidden

I never got this criticism either. If investors are too inept to look at both numbers and do their due diligence to figure out why there are differences (and whether or not those differences are in some way legitimate), then that's on them.

And I also just don't agree that SBC is an "expense" that materially impacts a business to the degree the article implies it does. I think there are a lot of other numbers that change between GAAP and non-GAAP that are much more relevant and interesting to look at. It seems like missing the forest for the trees to pick on SBC like this.

This is a shitty article. He doesn’t spend a single sentence defending why he thinks SBC should be included in non-GAAP reports. He just states it should be but never once says why. It’s a terrible article and a waste of time.

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