Mark Casey is a Capital Group equity portfolio manager who studies new technologies and their potential long-term impact on society. We asked him about some powerful trends emerging in the digital world and the tech sector’s recent share price tumble.
Most tech stocks went up a lot during 2020: The world shut down, and people were locked into their homes. They started using computers, TVs and video games way more often. A lot of these companies saw their fundamentals accelerate, and investors rewarded them with higher stock prices.
In the last third of 2021, as talk of higher interest rates started to kick in, highly valued stocks in all sectors started to come back down. Tech was disproportionately affected because many of those companies had really high valuations. Looking at it now, I think this has been a discriminating selloff. Not every stock is down equally. The companies that have the best fundamentals and stock prices that were rationally related to them have generally fallen less than average.
The stocks that went down the most were typically companies that warned of meaningfully slower growth, that said they won’t get as many new customer sign-ups as they used to or said their margins probably won’t go up this year. The market has been pummeling those stocks. We’re probably not through it all yet, but a lot of the froth has come out of the valuations, and most of the underlying trends driving the growth of some of these industries seem like they’re very much in place to me.
I think there’s a lot of opportunity in some of those stocks. The valuations of companies that offer infrastructure as a service (IaaS) aren’t statistically cheap based on near-term results, but looking forward, IaaS has a very long way to go from here. I think there is a lot of room for leading IaaS companies to continue to draw customers and IT workloads away from on-premise data centers. The three big players — Amazon Web Services, Microsoft’s Azure and Google Cloud Platform — add up to roughly $120 billion of run-rate revenue. That’s about one-tenth of the total revenue generated in on-premise data centers.
The same goes for software as a service (SaaS), which makes up most of the revenue growth in the software industry but is not yet close to making up the majority of software industry revenue. Another thing that’s got a long way to grow is the digitization of payments. I think almost all payments will be digital at some point.
I think it’s important to note that Facebook, or Meta Platforms, as it’s now called, isn’t dropping everything to pivot to the metaverse. The company will keep running the businesses that generate nearly all its current revenue — namely, Facebook, Instagram, WhatsApp and Facebook Messenger. I think what it’s doing is layering on a gigantic R&D project, Reality Labs. It’s like when Google started Waymo to do self-driving cars. It’s just that Reality Labs is very large. Waymo, plus all the other developing initiatives at Google, are losing $5 billion a year, while Reality Labs lost $10 billion last year.
Honestly, I love aggressive investments like this. Amazon CEO Jeff Bezos says if you tackle projects that take about three years to complete, you’re going to have lots of competition because lots of companies are willing to invest for three years to complete something. But if you start a project that takes nine years to complete, well, almost no companies are willing to invest on a nine-year time horizon. You don’t have much competition, and therefore you’re really betting on yourself much more. I think there are smart people at Meta, and they’ll either make the technology work or they’ll stop spending money if it becomes apparent it’s not viable.
I think NFTs are going to be really big, but I don’t think they need blockchain. Bitcoin is the first blockchain project, and it was created to let strangers who didn’t trust each other exchange digital coins. The blockchain system is like a Word document, and it continually adds new pages listing all the new transactions — only, unlike Word, once a page has been added, you can never delete it, you can never edit it. It’s a good architecture if you want to create a system that nobody can influence the rules of. But it’s very cumbersome, expensive and difficult to work with.
Bitcoin takes exceptional advantage of the blockchain and its immutability. Bitcoin’s capital supply is capped at 21 million coins, and there are no middlemen who can block, delete or reverse transactions. The rules can’t change. I think Bitcoin is the most exciting financial asset in the world.
Many other projects that say they use blockchain architectures — NFTs and many non-Bitcoin cryptocurrencies are examples — seem to me to be doing something contradictory. They’re using blockchain architectures, but the projects are run by people who can and do change the rules of the project. If you’re going to change the rules, why are you using a blockchain? Why not just use a regular client server database? It’s so much cheaper and easier.
As for NFTs, I think there is going to be huge interest in owning digital items. They’re already popular; people are buying things like new hair for their character in the game Fortnite. One reason people buy expensive things is to impress their buddies and hold the esteem of their peers. Last I checked, a Birkin bag doesn’t hold your items any better than a Coach bag, but people pay $35,000 for a Birkin bag. I think digital goods could serve the same purpose. If you buy an NFT from a well-known collection like CryptoPunks or Bored Ape Yacht Club for a million dollars, you can link to your OpenSea account and prove you own that. But you don’t need the blockchain. As long as issuers can credibly show there’s a limited number of the items, I think they’ll be able to use their own databases.