Recent posts

  • The CAPEX explosion

    The bits

    • Some of the FAANGs (Facebook, Apple, Amazon, Netflix and Google) released earnings reports last week.
    • Growth in capital expenditure (CAPEX) continues to surge, so I decided to investigate.
    • The FAANGs are optimistic about their future prospects. Their CAPEX is about the same as major oil companies.
    • Facebook in particular is spending like there’s no tomorrow, with CAPEX up 822% in five years.
    • The FAANGs are loved by investors, so have no difficulty in attracting capital. But they need to spend it.
    • Facebook’s splurging on infrastructure for video, and is exploring virtual reality and AI.
    • I’m sceptical of the fervour surrounding AI and virtual reality but I understand why Facebook is investing.
    • Personally I think you’d be brave to buy the FAANGs this late in the cycle.

    I like to keep my eye on happenings that seem out of place, or what we economists call leading indicators. So I watched eagerly when the FAANG stocks (Facebook, Apple, Amazon, Netflix and Google) started to report their third-quarter earnings reports.

    A bit of background. Capital expenditure (CAPEX) has exploded in the last few years, especially in Silicon Valley. The chief culprits are the FAANGs, which around 2015-16 really started to ratchet up their spending.

    The figure below shows the CAPEX of four of the five FAANGs (2018 is my full-year estimate). I excluded Netflix as although its spending trend is similar - up 248% in the last five years - total CAPEX is just 420 million, well below the other four.

    The CAPEX explosion

    While Google is the biggest spender at an estimated $24 billion, the real outlier is Facebook. Its CAPEX has risen from $1.36 billion in 2013 to an estimated $12.54 billion in 2018, for an increase of about 822% in just five years. It now invests almost as much as Apple and about the same as Amazon, even though it’s a “social media company” (it’s not; see my earlier work to find out why).

    Misplaced optimism?

    Clearly these companies are optimistic about their future growth prospects; Facebook, Amazon and Apple are spending about the same as Exxon Mobil - a major oil company - and Google is spending more. They’re also spending big on research and development, with Amazon blowing $22.6 billion in the year ended 30 June, up 40% from a year earlier (Google spent $16.2 billion). But given the seemingly constant stream of privacy violations and intense competition in a constantly evolving sector, is that optimism misplaced?

    Amazon and Apple are your more traditional, consumer-focused companies. Apple makes phones and other gadgets and sells them directly to you. Amazon is effectively a massive retailer, even if its profits are almost entirely derived from its cloud services. They both need to invest in factories, retail outlets, logistics, and so on.

    But Google and Facebook are advertising companies. They really only need office space for their employees (Google spent $2.4 billion on a piece of commercial real estate in New York), along with data centres and computing equipment to power their services. It’s hard to imagine needing to spend nearly $15 billion when your products are essentially a couple of web and mobile applications.

    When quizzed on that topic late last year, Facebook CFO David Wehner cited the following reasons for the CAPEX splurge:

    • Sizable security investments in people and technology to strengthen its systems and prevent abuse.
    • Investing aggressively in video content to support the Watch Tab.
    • Continued investment in long-term initiatives around augmented and virtual reality, AI and connectivity.
    • Substantial investments in its infrastructure to support growth and improve its products.

    Mark Zuckerberg noted that the company was planning to have 20,000 people working on safety and security in 2018, up from 10,000 in 2017. But even if you paid them all $100,000 a year, that only accounts for 1 of the additional 6 billion in new CAPEX this year. So the rest must be going into infrastructure for video and connectivity, along with possible future ventures such as virtual reality and AI.

    Time will tell

    Only time will tell whether the new ventures will pay off and reward these companies for their investments. I’m sceptical of AI. I think there is little doubt it’ll improve productivity in numerous sectors (e.g. driverless vehicles), but also that firms are spending far too much on anyone or anything that says it’s an AI expert. It has limitations (which I’ll get into in a future note) and so much of it will turn out to be vapourware, with the capital ultimately wasted. I’m also not sold on virtual reality, but I suppose if you’re an advertising company and you don’t at least explore the technology, it could be your death knell.

    But whatever you think of the technologies, it’s a risky strategy to invest so much this late in the business cycle (we’ve now had 111 months of economic expansion, or nearly 10 years of constant growth). But it’s also hard not to. 2018 is shaping up to be the best year since the last recession and financial conditions are the easiest they’ve been since the early 1990s. You’d be crazy not to spend.

    Indeed, investors love the FAANGs in part because they spend so much; their lofty valuations are driven by future earnings prospects, which are predicated on today’s large investments paying off sometime down the road.

    However as the saying goes, the best laid plans of mice and men often go awry and it won’t take much for those valuations to come crashing back down to Earth. The Federal Reserve is raising interest rates, the fiscal stimulus from Trump’s tax cuts will wear off, and I suspect China is faring worse than many people realise.

    Personally I think you’d be brave to buy the FAANGs this late in the cycle, even if you thought the explosion in CAPEX had been perfectly allocated (note that I’m a relatively risk averse investor).

  • Is Gab the real deal?

    I dish out a lot of hate to Facebook and Twitter on these pages. They’re advertising companies that offer users a free social media experience in exchange for all of their (and their friends’) data.

    I feel uncomfortable about the amount of information these organisations are entrusted with, so I personally keep the amount I feed them to a minimum. Where possible, I prefer to maintain control of my own data.

    But for those who cannot do without a social network, there are alternatives. Gab.com (formerly Gab.ai) promises to be different. Gab says it is:

    A social network that champions free speech, individual liberty and the free flow of information online. All are welcome.

    But is that all Gab is; a promise to be different? I decided to have a look at its privacy policy and terms of service to find out how it differs. Here’s what I discovered:

    Terms of service

    • We will use your User Content to run the App; we won’t take your User Content and sell it to others.
    • We may modify the Terms at any time, in our sole discretion.
    • You can remove your User Content by specifically deleting it. However, in certain instances, some of your User Content (such as comments or messages you make) may not be completely removed and copies of your User Content may continue to exist on the Services.

    I don’t see anything too scandalous with the above. The only issue I have is that Gab can change its terms whenever it so chooses. That’s standard in these kind of agreements, but what would happen if Gab went bust? Would it be able to change its terms to allow the selling of user content? I suspect the answer is yes. Moving on:

    Privacy policy

    • We use automated data collection tools to collect certain information. Some third party services providers that we engage may also track and report information about how and when you interact with our App and information about your mobile device (such as device hardware, operating system, and location).
    • We may share aggregated information and non-identifying information with third parties… Information that we collect from our users, including your email address and, if you opt to add it, a photograph or picture of you (PII), is considered to be a business asset. Thus, if we are acquired by a third party as a result of a transaction such as a merger, acquisition or asset sale or if our assets are acquired by a third party in the event we go out of business or enter bankruptcy, some or all of our assets, including your PII, may be disclosed or transferred to a third party acquirer in connection with the transaction.

    The above is a bit more concerning. Gab has essentially given itself a ‘get-out’ clause, whereby it’ll be able to distribute user data to the highest bidder in the event of a change of ownership. That day may come sooner than many think: as I’m writing, Gab is still offline following the recent shootings in Pittsburgh.

    “Gab has been no-platformed by essential internet infrastructure providers at every level. We are the most censored, smeared, and no-platformed startup in history, which means we are a threat to the media and to the Silicon Valley Oligarchy. 

    As we transition to a new hosting provider Gab will be inaccessible for a period of time. We are working around the clock to get Gab.com back online. Thank you and remember to speak freely.”

    My initial view of Gab when I first signed up last year was that it didn’t offer anything that Facebook and Twitter wouldn’t quickly solve if demanded by enough of its users. It’s essentially the same product - a centralised, trust-based model - with a right-wing hue.

    Don’t get me wrong, I’m all for competition and innovation in the social media sector so it’s great that it exists. It’s just that I have serious doubts as to whether (in its current form) it’ll be able survive, let alone supplant the existing Facebook/Twitter hegemony.

  • The data industrial complex

    At the 40th International Conference of Data Protection and Privacy Commissioners, Apple’s CEO Tim Cook stole the headlines when he - without naming names - referred to something he dubbed the “data industrial complex”, where:

    “Our own information, from the everyday to the deeply personal, is being weaponized against us with military efficiency. Every day, billions of dollars change hands and countless decisions are made on the basis of our likes and dislikes, our friends and families, our relationships and conversations, our wishes and fears, our hopes and dreams. These scraps of data, each one harmless enough on its own, are carefully assembled, synthesized, traded, and sold.”

    He has two companies in mind, both of which happen to be competitors of his (one more so than the other). Those two companies are Google and Facebook, which survive by bundling and selling the information that millions of people, knowingly or not, upload to their servers on daily basis.

    This. Is. Surveillance.

    The above heading is from Tim Cook’s speech. He’s right, but I think he takes it too far. The only way these companies are able to track you is by using the information you give them. Facebook didn’t install a tracking device on your phone; you did (e.g. its Messenger app). Google only knows about your interests because you plugged them into google.com, or you used its mapping service to find a restaurant.

    Don’t get me wrong, the practice leaves a bitter taste in my mouth. I personally try to avoid the ‘data industrial complex’ companies wherever possible, for example by using DuckDuckGo instead of Google, Signal instead of messenger, and deleting the malware that is Facebook entirely. There are plenty of privacy-focused alternatives out there, it’s just people - demonstrated by their actions - don’t seem to care.

    I should also add that Tim Cook has a clear vested interest here. He may very well be an angel, but Apple doesn’t rely on advertising revenue to survive (although it has pocketed some of Google’s by selling its users’ data), so if the government regulated that business model away then it would likely benefit Apple. Thus I wasn’t surprised when I heard his proposed solution:

    “We should celebrate the transformative work of the European institutions tasked with the successful implementation of the GDPR [General Data Protection Regulation]. We also celebrate the new steps taken, not only here in Europe, but around the world. In Singapore, Japan, Brazil, New Zealand, and many more nations, regulators are asking tough questions and crafting effective reforms.

    It is time for the rest of the world, including my home country, to follow your lead. We at Apple are in full support of a comprehensive federal privacy law in the United States.”

    A federal privacy law might help on some margins, but I’m skeptical as to the ability of politicians to get it right. The winners will almost certainly be the FAANGS (Facebook, Apple, Amazon, Netflix and Google), as they will throw an enormous amount of resources at whatever legislation is proposed to protect (and further) their interests. Facebook has literally asked to be regulated in the past.

    The losers will be the usual suspects: consumers, who will probably have to deal with yet another annoying “we use cookies” pop-up box on every website they visit; and smaller companies, or competitors that do not even exist today, which will find it too costly to comply with the new rules.

    Will a new law actually protect people’s privacy? Not really. Legislators don’t know enough about it so will rely on industry feedback, which have the effect of watering it down. Terms of service agreements will probably get a bit longer and require a few more clicks. It may also force companies to get creative with how they use your information, but it won’t stop them collecting it.

    The only way to truly protect your data is through encryption, something the Australian government and many others are actively working against. Essentially, the very same people Tim Cook wants to draft up legislation to protect your privacy are hard at work undermining it.

    Go figure.

  • Social media

    Social media has copped a bit of flack in recent times. Facebook founder Mark Zuckerberg was dragged before Congress to testify about privacy, data mining, regulations and the whole Cambridge Analytica controversy. More recently, Facebook revealed that about 30 million of its users had their accounts breached via the theft of access tokens (which allow you to stay signed in).

    Twitter has been accused of “shadow banning” controversial (mostly right-wing) individuals and groups. If you get shadow banned you stop showing up in searches, user feeds and such. You essentially become a ghost, able to tweet but no one can hear you.

    Google gave up completely and closed its social media platform, Google+. The final nail in the coffin for the ailing network came after discovering it had left private profile data exposed.

    It all begs the question: have we reached peak social media? The New York Times, for the first time that I am aware, published an article helping people disconnect. How to Delete Facebook and Instagram From Your Life Forever provides a step-by-step guide for people who want out, from backing up your data to the deletion process itself.

    I purged my Facebook account over 5 years ago. I actually deleted it but apparently “delete” really meant “deactivate”, as everything was still there when I signed in again a few years later. So I went through the process of removing just about everything I had previously uploaded to the network (the process is easier than it sounds with the use of third-party tools). I did the same with Twitter just a few weeks ago.

    Now I don’t for one second believe I’m some kind of thought leader, moving well before the masses with my decision to ditch social media. I just wasn’t getting enough value out of them given how much personal information I had to sacrifice. And the data do indeed show the opposite: more people than ever are using social media.

    Facebook users Twitter users

    My decision to abandon Facebook was made around the middle of the first chart, right when growth in the number of users really accelerated. Conversely, I only quit Twitter a few weeks ago, yet its growth plateaued back in 2014! I’m clearly not ahead of the curve.

    But that doesn’t mean I can’t be skeptical about the future of social media as it currently exists. Social media is a great way for people to organise events, stay in touch and connect with one another. But I’m not sure how centralised platforms such as Facebook, the business model of which depends upon harvesting user data for profit, can survive in the long run.

    Hear me out. Facebook and Twitter are not social media companies; they’re advertising companies. 98 percent of Facebook’s revenue and 86 percent of Twitter’s comes from advertising. The two companies provide platforms for their users to network primarily for two reasons:

    1. Direct advertising. Much like radio and television, Facebook and Twitter display advertising on their platforms. That includes allowing organisations to promote posts/tweets, accounts and even trends. They use algorithms to make sure promoted posts/tweets make it into the right users’ feeds.
    2. Data mining. Everything you share on social media, whether public or private, is mined. That includes the people with whom you associate, the events to which you are invited, down to your location and the device you use to access the service. This data is used to improve the effectiveness of direct advertising, but it’s also sold off to third parties.

    The provision of the social network enables the above. As the old saying goes, if you’re not paying for the product, you are the product. Social media companies sell you and any information you choose to provide, knowingly or not.

    To date, people seem happy with this arrangement; the growth of Facebook, Twitter and smaller platforms such as Snapchat are testament to that (Instagram is owned by Facebook). But they all use variants of the same model. As do the likes of Google, which data mines your searches and even your email (do you use Gmail? Or send/receive to people using Gmail? Google’s bots scan it all).

    But I’m not so sure this model is viable in the long term. Do I know what, if anything, will replace it? No. But as more and more information about the nefarious nature of how these companies generate revenue is revealed to the public, there will be pushback. Fewer new users will sign up, existing users will give these companies less face time and rivals - for example, encrypted and/or decentralised alternatives (e.g., diaspora*, Friendica, GNU social, Signal and Wire) - will slowly capture the more privacy conscious among us.

    I’ll leave you with the following quote, written about a certain social media company:

    “Well on the way to becoming what economists call a ‘natural monopoly.’ … Users have invested so much social capital in putting up data about themselves it is not worth their changing sites … Its massive user base will help maintain its dominance.”

    That wasn’t written about Facebook or Twitter. It was about MySpace, which has since faded into obscurity. Not because of antitrust regulation but because a new competitor, Facebook, was simply better.

    Does the same fate await today’s social network providers? I’m not sure. What I do know is that we live in an ever-changing world where business models need to constantly adapt just for a firm to maintain its status. Today’s advertising and data mining social media model seems antiquated to me, but I’ve been barking up that tree for years.