Regular readers have probably been following our recent quest to find the best semiconductor stock. We’ve dropped down quite the rabbit hole trying to identify a company in the mold of NVIDIA (NVDA). Its graphic processing units (GPUs) have helped it take a dominating position in multiple industries and sectors related to computing, including AI chips, data centers, gaming and the metaverse, and (at least at one point) crypto mining. But not every investment play has to be a pure play. Sometimes the best option is to go with the pick-and-shovel play. Or both.
We originally went long on Teradyne (TER), which automates testing for semiconductors and an array of other electronic and digital gizmos, because it was going long on industrial automation by acquiring four robotics companies in the span of four years. Our thesis was that Teradyne would eventually offer significant exposure to the robotics theme. Unfortunately, it’s not happening as quickly as we’d like. While we continue to wait, we’ve been distracted by the fact that the Baawston-area company is also a pick-and-shovel play on AI and automation. It is a major service provider to big-name customers like Samsung, Qualcomm, and Intel that power things like supercomputers, smart appliances, and smartphones. Like Nvidia, it offers both direct and indirect exposure to multiple emerging technologies.
It’s been a little more than a year since we last checked in with Teradyne, in an article we optimistically called Teradyne Stock: The Forecast Looks Better Than Ever. At the time, the company had just posted record profits, but the stock had taken it on the chin because management actually offered some pretty transparent guidance for the first half of 2022 that suggested revenue headwinds were blowing. While we don’t have the 2022 full-year SEC report yet, Teradyne held an official presentation at the end of last month. Let’s see where things stand with Teradyne stock in 2023.
The Chips are Down and Customer Concentration is Up
And let’s start with the bad news. Annual revenue dropped about 15%, from $3.7 billion to $3.2 billion, in 2022. The money drain was mainly in the system-on-chip (SoC) testing market, which the company estimated dropped 6% globally to about $4.6 billion. Based on that estimate, Teradyne accounted for about 37% of the market share with $1.7 billion in revenue. Bottom line is that testing revenue – Semiconductor Test (includes SoC testing and memory), System Test (something to do with data storage and defense systems), and Wireless Test (connectivity and all that) – was pretty lackluster.
The company warned investors of more the same in 2023. For example, it predicted the market for SoC tests would contract between 10% and 30% below last year’s $4.6 billion level, so expect another drop in SoC testing revenue. Management did try to put a positive spin on things, of course. The mass production of 3-nanometer chips is supposed to pick up steam this year, so that could help offset softness in other parts of the company’s testing markets. Specifically, Teradyne said its “historically largest end customer” is expected to lead this transition. That’s either Samsung or (more likely) Taiwan Semiconductor. While the latter is the world’s largest semiconductor manufacturer, the former got a jump start on producing the first-generation of 3-nm chips last year. Regardless, that customer will move from less than 10% of 2022 revenue to a low double-digit percentage of revenue for this year.
Customer concentration is an ongoing problem for Teradyne. In 2021, the company’s five largest customers accounted for 33% of its total revenue. Taiwan Semiconductor alone represented almost 19% of its 2021 consolidated revenues and was as high as 25% in 2020. Last year, Qualcomm was the only company to account for at least 10% of total revenues, according to management, though we don’t know the full picture until the 10K hits the newsstands. Regardless, Teradyne is reliant on a distressingly small number of customers.
The Roundup on Robots
Now, on to industrial automation (IA), also known as the robotics stuff, the reason why we invested in Teradyne in the first place. Revenue was up just 7%, from $376 million in 2021 to $404 million in 2022. That’s a big letdown after Teradyne’s revenue from its robotics division jumped 34% between 2020 and 2021. Blame a strong U.S. dollar for dragging down revenue, according to the company’s bigwigs, who said revenue growth was actually 15% in constant dollars. That makes sense, because 65% of 2022 sales were outside of the United States, including 40% in Europe, 11% in China, and the other 14% spread across the rest of the globe.
The biggest piece of the pie comes from Universal Robots, which Teradyne acquired in 2015 for $285 million. Last year, the subsidiary contributed $326 million. The robotics company also launched its strongest cobot to date, the UR20, which can move payloads up to 45 pounds, despite only weighing about 140 pounds. Mobile Industrial Robots (MiR), which Teradyne added to the fold for $148 million in 2018, accounted for $77 million – a 19% increase in revenue from the year before.
Like its testing markets, Teradyne expects the IA division to face an uphill battle in the first half of this year before rolling to victory in the second half. Why the optimism? Well, the company doesn’t believe currency exchange will be such a problem in 2023. In addition, Teradyne is building its customer base outside of traditional distributors and into manufacturing, which grew 26% in 2022, among other verticals. Finally, the company believes the new UR20 will drive sales, leading to more than 20% growth in 2023 revenue, mostly loaded on the back end. (More on this in a bit.)
The Bull Case for Teradyne Stock
One thing we like about Teradyne stock is that the company provides lots of data and analysis, and management clearly explains what’s up with what. Here we see the company has revised its mid-term outlook through 2026 based on the current market weakness.
Based on the near-term conditions, the 2026 numbers seem pretty optimistic. Management is bullish for a few reasons. It believes markets like AI and cloud computing, mobile processing, and automotive – especially advanced driver-assistance system ADAS and electric vehicles – are driving up the numbers and complexity required of semiconductor hardware. Ditto for markets serving wireless standards for supporting “ever higher data volumes and the pervasive deployment of edge AI.” Teradyne has automatic test equipment for all of these technologies. We would also add that Teradyne likely stands to benefit from the passage of last year’s so-called CHIPS Act, which is pouring $280 billion into all sorts of emerging technologies, including $52 billion in subsidies and additional tax credits to companies that manufacture chips in the United States, over the next five years. The EU also passed similar legislation last year.
Teradyne leadership also said IA would help drive revenue growth in the next few years. The thinking is that all of the global slackers labor shortages and wage pressures will spur more companies to automate. They also claimed that market penetration for collaborative robots, including the little mobile autonomous ones that scoot around factory and warehouse floors, is less than 5%. The big announcement was that the company believes revenue from robotics will hit $1 billion, representing about 20% of total revenues by 2026. The new CEO, Greg Smith, was most recently president of the IA group, so presumably, he knows what he’s talking about. To hit this target, the Industrial Automation segment would need to grow by a 25% compound annual growth rate (CAGR) over the next four years (reflected below in green).
After waiting four more years, provided the company hits that aggressive target, we’d only have 20% exposure to industrial robotics which is why we’re holding the stock in the first place. Our last piece on Teradyne talked about the company promising us that by 2024, industrial robotics would be 18% of the mix (an implied $887 million in revenues). Now we’re being spun a much less attractive story. It’s awfully tempting to find a better way to play the exposure of robotics and swap out Teradyne.
The Bear Case for Teradyne Stock
The last time we checked in with Teradyne stock, the forecast model then went out to 2024 certainly looked stronger. We hope this doesn’t become a case of forever moving the goalposts. In addition, the total addressable market for its core market in semiconductor testing is only about $5.6 billion between SoC and memory. Based on the company’s dominant market share, we wonder if slowing revenue growth will be an issue in the future.
However, our biggest concern is with China, and we’re not talking about balloons. Well, in a way we are. While we refuse to indulge in political banter on Nanalyze, politics does have a way of barging into the conversation with regulations. Specifically, the U.S. Department of Commerce has put new restrictions on exporting semiconductor-related technologies to China, including sales of semiconductor testers. The regulations will definitely affect Teradyne, though the company is still assessing the impacts and looking for workarounds such as applying for a bunch of licenses and waivers to continue some operations. Still, this could be especially disruptive, given China accounts for 16% of total revenue. From the Q3-2022 report: “[T]he regulations may have an adverse impact on certain actual or potential customers and on the global semiconductor industry.”
So, add regulatory risk alongside customer concentration to the list of red flags we need to watch moving forward.
Conclusion
These are good reminders that no matter how well-positioned a company appears, it has its own unique vulnerabilities. The question, as usual, boils down to risk. While customer concentration for Teradyne is a concern, there is no reason to believe its biggest customers are going away. In fact, given the global need for semiconductor technology to power everything from our toaster oven to our electric car, long-term business prospects look pretty good. The argument for accelerated industrial automation in robotics also makes sense to us, though even if Teradyne manages to hit its $1 billion target in three years, our exposure to industrial robotics will still be minimal. Where’s the robotics exposure we had hoped for?
Tech investing is extremely risky. Minimize your risk with our stock research, investment tools, and portfolios, and find out which tech stocks you should avoid. Become a Nanalyze Premium member and find out today!
