Google's 2015 restructuring into a holding company called Alphabet was a bit puzzling at the time.
Essentially, the company took its very well-established and well-regarded business, which accounts for 99 percent of its revenue and more than 100 percent of its profit, and put it under a holding company alongside a bunch of tiny experimental companies, like self-driving cars (Waymo), health-tech projects (Verily) and venture capital investments (GV, formerly Google Ventures, for early stage bets; and CapitalG for later-stage investments).
At the time, CEO Larry Page wrote the restructuring "allows us more management scale, as we can run things independently that aren't very related." He noted that each of the independent companies would have its own CEO, while Page would remain in charge of the overall holding company.
In practice, this also allowed Page to appoint a new Google CEO, Sundar Pichai, to manage everything associated with Google, including uncomfortable tasks like speaking on earnings calls and testifying before Congress, so Page could spend time on the areas that interest him most.
The split also reassured investors that, despite all the noise the company made about these long-term bets, it wasn't blowing a lot of money on them. In the fourth quarter, for example, these "Other Bets" lost $1.3 billion on revenue of $154 million, while Google proper earned $9.7 billion in operating income on $39.1 billion in sales. The side projects made for a sizable but affordable investment.
But there's a more subtle reason that started to become clear on Alphabet's fourth-quarter earnings call on Monday. The company is apparently treating some of these companies like true start-ups — all the way down to issuing equity in them that's not strictly tied to the value of publicly traded Alphabet shares.
On the call, Alphabet CFO Ruth Porat noted:
In 2018 we also saw continued progress on our goal to nurture new businesses outside of Google, our Other Bets, with the model of independence.
While I've said previously that there is no monolithic approach to how the Other Bets execute against opportunities, our shared principle is aligning employee interest with the long-term value creation by these companies.
Since inception this has been an important part of our approach of Google with stock grants comprising a meaningful components of overall compensation across our employee base. We're increasingly following that approach in the Other Bets with compensation programs that align employee and company interest.
You've seen the impact of some of those programs in the fourth quarter as we accrued compensation expenses to reflect increases in the valuation of equity in certain Other Bets. While these expenses will recur the timing of valuation events is unpredictable and can vary between bets, which can affect quarterly comparisons.
During a Q&A, Porat reiterated the point:
In certain Other Bets, employees are compensated through equity-based programs, and that's because we believe that this alignment of interest is valuable.
Alphabet's annual earnings filing, released Tuesday, clarified further:
Additionally, stock-based compensation includes other types of stock-based awards that may be settled in the stock of certain of our Other Bets or in cash. ... The fair value of such awards is based on the valuation of equity of the respective Other Bet.
In other words, Alphabet knows these long-term bets are risky and may never pay off. To succeed, the employees who work for them need to put in the kind of hours normally associated with start-ups. And to convince employees to do that, they have to be able to share in the upside if the value of these companies increases 100 times — otherwise, why not just go down the street and work for the hot start-up of the hour, where there's a real chance to become spectacularly wealthy?
The revelation that some of Alphabet's Other Bets have their own equity structure leads to some interesting questions:
- Who determines the valuation of each Other Bet, and what methods do they use?
- Can equity in these companies be traded among private investors, or are there restrictions that allow Alphabet to retain control?
- Do the shares in the Other Bets come with voting rights, or does Alphabet retain complete control?
- Could Alphabet try to spin these Other Bet companies off into separate publicly traded entities with their own shares, and how would existing Alphabet shareholders — who nominally own a piece of them today — be compensated?
- If you're an investor today and you think Google's best days are behind it but think there's long-term potential in self-driving cars, is there any method for you to invest your money in Waymo without buying Alphabet shares? It appears that Alphabet is selling equity in at least some of these companies to private investors — Silver Lake just led a $1 billion investment in Verily late last year. Is there a path for smaller investors to get involved?
We've put these questions to Alphabet's investor relations team, and we'll update this article if we hear back.
The whole arrangement highlights the quandary that any successful tech company faces. Everybody in the industry has read Clayton Christensen and is cognizant of the cycle of disruption — the mainframe was disrupted by the PC, which was in turn disrupted by the smartphone — and how companies that are on top of the world one day can shrink into oblivion in a few short years (see Nokia and BlackBerry).
The only way out of this trap is to be your own biggest disruptor. But it's extremely difficult — a company has to be willing to forgo the profitable businesses that made it successful before they're finished delivering cash and profits.
Today's tech giants are handling this in different ways. At Amazon, CEO Jeff Bezos has tried to instill cultural values like "always Day One," that keep employees thinking like they work for a start-up. Microsoft has over the last decade completely shifted the company's focus away from Windows on desktop PCs and toward cloud computing.
At Alphabet, the solution seems to be to take the company's massive resources and apply them in a venture capital-style model to a bunch of promising start-ups. The experiment is less than 4 years old, so it's impossible to say whether it will succeed or not. But the methods are gaining clarity.