Dan Loeb Goes Bargain Hunting: 3 Stocks He Just Bought

The active management industry has been shrinking while the passive investing craze has continued to take share. But notable active managers are still worth following; in fact, in a world where so many are going passive, one could make the argument the remaining notable active managers may have more importance as they buy and sell popular stocks.

Meanwhile, one of the best ways to follow these notable money managers is through their 13F filings, which show each manager’s buys and sells over the most recent quarter. As they come out 45 days or so after the end of the quarter, the most recent disclosure period is for Q1.

Still, these can help glean what the best active managers are thinking. One notable manager, billionaire Dan Loeb of the hedge fund Third Point LLC, continued to buy into the artificial intelligence story in the beginning of the year — but perhaps not in the way you’d expect — with these three buys.

Amazon

Amazon (AMZN 0.03%) was already the third-largest holding for Third Point as of Q4 2023, but Loeb decided to increase that stake even further, making Amazon the new second-largest holding in his portfolio, finishing the first quarter at 11.72% of Third Point’s holdings.

Amazon has been a relative laggard among the Magnificent Seven stocks for the last couple years, having the second-lowest three-year return of the bunch.

AMZN 3 Year Total Returns (Daily) data by YCharts

Loeb is probably correct in thinking that is unwarranted. While much of the attention has gone toward Microsoft and its partnership with OpenAI, the parent company of ChatGPT, Loeb actually lowered his stake in Microsoft slightly and bought into Amazon.

This is because, more likely than not, artificial intelligence won’t be a winner-take-all industry. Customers will likely use a variety of models for a variety of different applications, while using different kinds of AI chips.

In a notable move last quarter, Amazon inked a partnership with OpenAI rival start-up Anthropic, investing $4 billion in the start-up, with Anthropic making a commitment to use Amazon Web Services (AWS) as is its primary cloud provider and committing to building, training, and deploying its models on AWS’s in-house designed Trainium and Inferentia accelerators. Amazon was actually one of the early cloud companies to bring chipmaking in-house, giving its customers the option for lower-cost in-house alternatives to expensive Nvidia GPUs.

For those thinking OpenAI has the first-mover advantage, consider that Anthropic’s CEO Dario Amodei used to be a vice president of research at OpenAI, and actually left with several other OpenAI employees to form Anthropic, because of differences of opinion with its current leadership.

And as with its market-leading cloud, Amazon is committed to hosting the widest variety of large language models for use on its Amazon Bedrock platform.

Amazon’s relative underperformance comes despite all of this as well as an impressive turnaround in its e-commerce business over the past year, where revenue has increased and profitability has surged. This all bodes well for the future of the stock, even after it surged another 10% already in the second quarter.

Meta Platforms

Another incremental addition Loeb made was to his fifth largest position in Meta Platforms (META 0.13%).

It’s unclear when Loeb added to his stake in Q1, but hopefully it was early in the year. Meta soared to begin January and then took another big leg up after reporting impressive Q4 results. While the stock saw a big sell-off after its Q1 results that were reported in April, due to the company’s higher AI spending forecast, the stock is still up over 49.5% for 2024.

And if leadership wants to spend more on AI, this investor is all for it. Artificial intelligence has helped Meta bounce back from weak results in 2022 and early 2023, as its investment in GPUs allowed it to reengage users with relevant Reels and posts, growing Daily Active People across its apps by 7% last quarter. Additionally, AI has helped Meta better monetize those users. Last quarter, revenue surged 27% even as Meta kept a lid on costs, enabling its operating margin to expand by 13 percentage points, from 25% to 38%, in just one year!

But the AI story doesn’t stop there, as Meta has also begun building its Llama family of large language models that aim to compete with OpenAI and all other AI start-ups. Therefore, its not just that Meta’s core social media platforms remain tremendous high-margin cash cows, but Meta also stands a chance of becoming a leader in third-party AI models that are currently generating so much excitement.

Meanwhile, Meta still trades at a reasonable 27 times this year’s earnings estimates, despite losing over $10 billion on its long-term Metaverse investments and not much revenue from Llama yet. No wonder it made up a strong 7.65% of Third Point’s portfolio at the end of Q1.

Alphabet

Meta’s digital ad rival Alphabet (GOOG -0.02%) (GOOGL -0.03%) actually became a new position for Third Point in the first quarter, with Loeb immediately making it a 5.77% position, good for the seventh-largest position in the fund.

In fact, Loeb devoted a large part of his quarterly letter to investors to discussing Alphabet. Somewhat like Amazon, Alphabet had been a relative laggard among the Magnificent Seven over the past three years, with the third-lowest performance. This is probably due to the perceived threat of AI on Google’s core Search business.

However, Loeb thinks these concerns are overdone. After all, the word “Google” is synonymous with search, and when the company’s name has become a verb synonymous with its activity, that’s usually a good sign of a distribution advantage. In the Q1 letter, Loeb states this very clearly: “Alphabet, however, has both a substantial distribution and technology advantage over competitors and is positioned to use its AI capabilities to unify, enhance, and better monetize the entire suite of its products,” Loeb wrote.

Google is also working on its own LLM called Gemini, which had a few notable blunders in its first public appearance. However, Alphabet invented some key technologies used in all AI models today, and it has substantial financial resources to put behind its AI efforts in the combined Google Brain and DeepMind teams, which Google merged last year. Thus, Gemini should be able to hold its own in the AI races over time.

With an entrenched user based and significant resources to put behind its own AI efforts, Google still looks relatively cheap to the other Mag-7 stocks, at just 25 times this year’s earnings estimates, despite having appreciated some 25% since the end of the first quarter.