Head of Equity Strategy
Summary: Silicon Valley wants investors to believe in the non-GAAP metric and essentially ignore stock-based compensation expenses. Not only is this misleading, it's everything that's wrong with tech, venture capital, and cheap money in general.
In my opinion, it’s embarrassing, and it tells you everything that’s wrong with cheap money, the venture capital industry and Silicon Valley. Later today I will extend this conversation to startup exits and why I believe we have a systemic problem in our economy. But first, back to Box’s Q3 result and why it offers important lessons for investors looking at technology companies (whether private or public).
Stock-based compensation is a cost
When Box’s CFO says that the company is likely to achieve profitability in Q4, it is based on something called non-GAAP metric – a metric that has been criticised multiple times by the US Securities and Exchange Comission. According to Box’s Q3 financial statements, the US GAAP net loss was $40.2 million in Q3, but it was only $8.4m measured in non-GAAP terms.
This is done by subtracting the $31.8m stock-based compensation expense. Box wants essentially investors to ignore this expense as if it’s irrelevant; at least, that’s how we must interpret the company's statement about profitability. Box does not care about stock-based compensation. We will come back to why that’s a big mistake.
Prior to 2005, the US Financial Accounting Standards Board did not treat stock options as an expense, leading obviously to perverse use of these compensation instruments for executive pay as they did not hit the bottom line; the trick was to link executive compensation to net income. After the 2005 change, US companies were quick to innovate and started linking compensation to EBITDA (earnings before interest, taxes, depreciation and amortisation) which excludes stock-based compensation. So again, the link is to an accounting measure not incorporating the cost of the compensation – it's really quite smart.
Warren Buffett was always clear in this debate, arguing that stock-based compensation is an expense and should be booked accordingly. The reason is simple: stock options are non-cash compensation so if they are not used, the company would likely have a higher cash salary expense. This quote from Box’s Q3 earnings release confirms this view: “...additionally, in the case of stock-based compensation expense, if Box did not pay a portion of compensation in the form of stock-based compensation expense, the cash salary expense included in cost of revenue and operating expenses would be higher, which would affect Box’s cash position.”
In other words, stock-based compensation is clearly an expense and often a quite sizeable one. In the case of Box, stock-based compensation is around 20% of revenue in Q3. This means that if Box shows a non-GAAP profit in Q4, the actual profit margin including stock-based compensation is still around -20%, and this after being in business for 13 years! As frequent readers of our analyses will know, we are always using EV/EBITDA metrics in our valuation analysis but we have erred in not adjusting the EBITDA numbers for stock-based compensation. This naturally makes the EV/EBITDA more attractive on a relative basis for companies using a lot of stock options; we will make efforts to correct this going forward.
When everything else fails, use the cash flow statement. This has always been the mantra in financial analysis of companies because the cash flow statement is more difficult to manipulate. In the case of stock-based compensation, however, it masks the true nature of the employee expenses, which are typically cash-based and thus subtract in the cash flow from operations.
In the last nine months (ending October 31), Box reported $24m in net cash from operating activities; this is the true number. But as the company stated concerning stock-based compensation, if it could not issue those options the cash-based salary expense would be much higher. If we assume that the fair-value of stock-based compensation, at the grant date, is higher than what would have been the true cash-based salary to attract the employees, then we should not adjust cash flows from operations by the full amount of the stock-based compensation. Assuming a deduction factor of 0.5, the $24m in cash flows from operations would still be -$20.6m for the past nine months – a big difference when analysing Box.
Ignore non-GAAP metrics
The message is clear: don’t use the non-GAAP metrics provided by many US companies. They may be published with good intentions but they portray the wrong picture for a potential or existing shareholder. Our warning is very important for investors investing in technology companies as these are the heaviest users of stock-based compensation.
Another recent example of misuse of non-GAAP metrics came when Groupon wanted to IPO using gross billing as its revenue figure, obviously to inflate the numbers. But the SEC declined this use and stated that net revenue (Groupon’s cut of the gross billing) is the right and proper metric to publish. According to S&P Global data from 2015, the total equity-related remuneration was $30.4 billion or 12.1% of pre-tax profits for the 67 technology companies in the S&P 500 at the time. This is a significant adjustment in valuation figures for an entire sector.
The fact that stock-based compensation is a sizeable expense for shareholders in technology companies is one thing. Even more worrying, though, is that it’s quite likely that companies, and hence shareholders, end up paying too much for labour as the stock-based compensation has a convex payoff profile in the case of success. It would probably be much better for VC firms in the pre-IPO phase to inject more cash into startups, paying with cash and not stock options. To provide some valuable colour on why using company shares as payment in acquisitions or to attract employees, we will end this discussion on stock-based compensation with a wonderful story from Warren Buffet:
"I made an even worse mistake when I said “yes” to Dexter, a shoe business I bought in 1993 for $433 million in Berkshire stock (25,203 shares of A). What I had assessed as durable competitive advantage vanished within a few years. But that’s just the beginning: By using Berkshire stock, I
compounded this error hugely. That move made the cost to Berkshire shareholders not $400 million, but rather $3.5 billion. In essence, I gave away 1.6% of a wonderful business – one now valued at $220 billion – to buy a worthless business." — Berkshire Hathaway, 2007 shareholder letter