You may think that this year's outperformance from artificial intelligence stocks means they're due for a correction, or at least relative underperformance in 2024.
But I think you would be wrong. With the introduction of OpenAI's ChatGPT chatbot just one year ago, we are still just at the beginning of the AI age.
For long-term investors, getting in at the relatively early stages of powerful secular long-term trends at reasonable valuations can lead to astounding long-term returns.
That's why Super Micro Computer (SMCI 8.89%) deserves a spot in everyone's portfolio, highlighted by these four underappreciated attributes.
1. Artificial Intelligence hardware expectations keep going higher
Sure, artificial intelligence stocks have climbed a lot this year, but this appears to be based on actual revenue and earnings growth, with AI use cases expanding and becoming, "table stakes" for any company hoping to compete in the 2020s.
Looking at one data point, look no further than the escalating predictions for AI chip growth from Advanced Micro Devices CEO Lisa Su. Su has an amazing track record as AMD's CEO, and is certainly one to be taken seriously by the investment community.
Back in June, Su predicted the AI chip market would grow from $30 billion in 2023 to $150 billion by 2027. But at AMD's recent December AI event, she greatly upped her own outlook, to a $45 billion market in 2023 growing to a stunning $400 billion market by 2027.
While some of that increase is due to non-server markets such as AI PCs, one can safely assume at least half of those chips will be in AI servers, which means Super Micro will have an awful lot of room to grow if it executes.
Super Micro has underrated advantages
Many investors regard server-makers as just "dumb box" assemblers of highly proprietary chips and networking technologies. But the higher demands of artificial intelligence mean more and more complex parts have to work together, requiring new innovations in server construction such as liquid cooling to dissipate heat from AI's intense computing demands.
That plays into the strengths of Super Micro's business model. Super Micro began as a producer of motherboards back in the 90s, before moving onto chassis, backplanes, and other components, before moving to making entire servers and now entire server rack systems. Most recently, SMCI has developed proprietary liquid cooling systems that will be required of more servers going forward in the AI age.
Basically, Super Micro started out similar to today's ODMs (original design manufacturer) which make parts and components used by other companies. ODM clients include large companies that construct their own servers (like cloud giants) or other OEMs (original equipment manufacturers) that incorporate third-party parts into their own branded servers. But Super Micro evolved from that model into its own vertically integrated system manufacturer, albeit with all its components designed and optimized in-house. That has enabled a "building block" architecture for its systems.
The result is that Super Micro can customize unique models for various needs, rather than offering a limited number of standardized models, which is the model of most OEMs. And its use of entirely in-house components allows SMCI to have incredibly fast time-to-market. It's basically the best of both the ODM and OEM worlds.
So in order to compete with SMCI, OEMs would have to change their current ways of doing business to manufacturing their own parts that they currently outsource. And ODMs specializing in one or more components would have to learn how to build entire end-to-end systems. So, it would be difficult for competitors to reconfigure their current models to compete.
Image source: Getty Images.
A reasonable valuation
While many AI-related stocks have climbed a lot this year, Super Micro only trades at 25 times trailing earnings and 15.8 times 2024 earnings estimates, with the 2024 fiscal year ending in June. That's actually a below-market multiple based on 2024 estimates, which seems quite odd given current revenue growth estimates for Super Micro of 49% in 2024 followed by another 26% estimated growth in 2025.
In contrast, rival Dell (DELL 0.12%) trades around 10 times forward earnings, but its revenue is expected to contract 13% in fiscal 2024 and only grow 4.7% in fiscal 2025 -- significantly lower than Super Micro. And Hewlett Packard Enterprise (HPE 1.18%) only trades around 8.6 times forward earnings, but its revenue is only forecast to grow 1.6% and 3.6% in each of the next two years.
So while Super Micro is slightly more expensive than its two larger rivals, it's also forecast two grow five to 10 times faster. Moreover, both Dell and HPE have relatively high debt loads, whereas Super Micro has a net cash position.
A highly motivated founder who's also a large owner
Perhaps the most overlooked part of assessing a business is looking at inside ownership, as well as the criteria for management's compensation.
Super Micro is fortunate to still be led by its visionary founder Charles Liang, who as of Super Micro's recent proxy report still owns 14.3% of the company.
That's a large amount, but if that wasn't motivation enough, Super Micro's new 2023-2028 CEO compensation plan offers even more. Over that time, Liang will only earn $1 in salary and no cash bonuses. Instead, Liang will be eligible for options to purchase up to 500,000 shares across five performance-based benchmarks, with options for 100,000 shares awarded after hitting each tranche before 2029.
What's so encouraging is that the two benchmarks encompass revenue growth and the share price, with revenue targets spanning between $13 billion and $21 billion, compared with last year's revenue of just $7.1 billion and this year's guidance for $10.5 billion. Meanwhile, share price targets range from $450 to a whopping $1,100 per share, in contrast with a roughly $273 share price today.
As a final exclamation point, all of these options have a strike price of $450 per share, which means SMCI's stock has to clear this hurdle by 2029 for Liang to earn anything!
All in all, given increased AI growth expectations, Super Micro's unique business model, and Liang's motivation to generate revenue growth and a much higher stock price, Super Micro belongs in everyone's portfolio today, in spite of its massive two-year outperformance.
