Manager: Nvidia Will Be the Biggest Company in the World

Christopher Johnson: Welcome to Morningstar. My name is Christopher Johnson, and today, I’m joined in the studio by Stephen Yiu, the managing partner and CIO of the Blue Whale Growth Fund. Stephen, thank you so much for being here with me.

So, my first question to you is about Nvidia NVDA. It is the top holding of the fund. An investor from Liontrust was recently interviewed on CNBC, and she argued that, despite Nvidia’s high valuation, it is currently undervalued. Do you agree with this statement?

Stephen Yiu: Yes, we do agree. It depends on which time period you’re looking at Nvidia. So, on a one-year P/E ratio basis, if you follow them using the market consensus, it’s actually trading at a similar valuation to Microsoft MSFT and Amazon AMZN. So of course, then if you believe that Amazon and Microsoft are overvalued at over 30 times earnings, then so is Nvidia. But if you believe that 30 times earnings for those businesses are OK, then Nvidia is not overvalued. But of course, over the medium term, in the next three to five years, we do believe that Nvidia is going to make a lot more money in terms of selling more of the GPU in revenue terms and just because the business in itself is very high margin in terms of operation, so they’re going to deliver a lot of free cash flow back to shareholders.

And to our mind, I mean, we have been invested in Nvidia since 2021, and we have increased our position quite a lot in 2022 when the share price came back down a lot. And we still maintain a very large position—just short of 10%. So, to our mind, we have no doubts that Nvidia is going to become the biggest company in the world in time, surpassing both Apple and Microsoft, which is not a long way to go. But of course, from an investment manager perspective, if Nvidia is going to do that within the next 12 months or so, I think we are very happy with that. But if Nvidia is going to take another three years to do that, we will be happy, too, but then we might have a smaller position. So, it’s one of those things that we want to optimize the performance of the fund in time. But as far as the prospect of Nvidia is concerned, we have no issue with that.

Johnson: At the Morningstar Investment Conference in London, there was a panel discussion about AI, and people were talking about the ramifications due to it maybe using the intellectual property from other content out there. Do you think this is a risk that maybe investors are not seeing with AI?

Yiu: This is the direction of travel. So, there’s going to be more regulation coming in, and the regulators are coming to challenge some of this. I think the one company that’s slightly questionable today is OpenAI because they do not disclose their source, how they come to do the training. But if you look at, for example, Adobe Firefly, which is a competing product to Midjourney—Midjourney doesn’t disclose their source as well—if you were a creative professional to use Adobe Firefly rather than Midjourney to generate an image using the GPT capability, they guarantee you that it’s copyrighted in a way that there’s no issue. They would be paying whoever generated the images or the video that have fed onto the AI to do the learning and then gave you the output. So what Adobe is doing is to assure the enterprise customers that you’re not going to just generate some images and then you suddenly get a lawsuit. So hence, for the very large enterprise customer, they would be still using Adobe Firefly rather than Midjourney for maybe just an Instagrammer.

On the other hand, if you look at maybe the music industry, I think last year there was some music that’s been generated by AI using like an existing artist archive ,and it was very popular. But of course, what has happened is the record companies, the labels, have then said to the likes of Spotify SPOT or Amazon Music that you need to take this down because it’s copyrighted—which they did. So, I think the direction of travel is like you would continue to see this two way of coming up with the output on the back of generative AI, but I think for the more sophisticated market like the enterprises, like if you’re a big company, you would rather not take the risk. So, you need to basically do what you need to do to satisfy the current regulations.

Johnson: At the end of December 2023, according to Morningstar data, you sold 780,000 shares in Universal Music Group UMG, reducing your position. What led to this decision? And did it have anything to do with the months-long licensing fee between the group and TikTok?

Yiu: So not so much about TikTok. So basically, we invested into UMG about two years ago, and we actually recently exited our position completely. So, we sold out from the shares completely only on the back of two things. One was on valuation grounds. It’s done very well for us. I think the thesis that we had at the time was that the group was undervalued for how well they monetized the music, partnering with Spotify, Amazon, Instagram, and some others, of course, TikTok was also one of the players. So that has played out quite well. We have no concern about that part of the market. We had not so much concern about AI music being generated using the existing artists or the kind of the label archive because they have very strong relationship with the distributor, which are Spotify, Amazon Music, so they can force them not to embrace that.

What is slightly questionable in terms of the AI threat to the music industry? We don’t know whether it’s going to materialize or not. It’s a speculation based on our take today. It’s how we come to know about a pop star like Taylor Swift or some others is that the record label has spent a lot of money to build up that sort of character or the artist in itself and of course then they do a lot of concerts. I mean, it’s exceptional artists and other stuff. And because in terms of how we receive the kind of the music and all that, it’s also quite controlled in a way that, OK, if that got recommended to you, if they have more money to spend doing tours and you come to learn about them, if they have money to spend on Instagram, then you come to know about them, etc. But in time what we don’t know is whether AI is going to change that in terms of how we receive music, not in terms of the generation of music, is whether then suddenly maybe the big star, the big pop star or artists are no longer for everyone, that maybe it’s the smaller artists that doesn’t even have a record label to back them up, that because of the recommendation system that they get to know a lot more about you, let’s say Spotify, know about what you like and then they start recommending new stuff that are not the kind of the artists being backed by the record labels.

So then whether that would change the pattern in terms of the business model of the record label because at the moment of course they have all the archive, which they are making a lot of money from, but then we’ll talk about what’s going to happen in the next 20 years, are they going to have the next Taylor Swift or is this Taylor Swift going to be coming from somewhere else that’s not backed by a record label? So that is just a question mark that we are debating, but because of valuation ground, it’s delivered what we expected them to do and then we’ve made quite a lot of money on the back of that. So, we exited the position.

Johnson: According to Morningstar data, your fund returned 29.5% over the three years to May 2024 versus the Morningstar Global Growth Index, which returned 13.36%. So, what do you think has been the key to the fund’s outperformance apart from Nvidia?

Yiu: In Q4 2022, we initiated our first and only energy company, called Canadian Natural Resources, which is currently in the top 10 [of our portfolio]. That company actually has gone up quite a lot—about 50% return between Q4 2022 until now, so it’s done very well. That is on the back of our view that oil prices are going to be well supported at a higher level than before just on the back of the political uncertainty, geopolitical uncertainty in the Middle East, in particular. And at the same time, if you look at like Mastercard MA and Visa V, you would think, OK, these two companies sound quite boring, everyone knows about these companies, but we actually would categorize these two companies as the new regime beneficiaries, so they’re going to benefit more on a high level of inflation.

How are they going to do that? It’s just on the back of them charging us a fee on our nominal spending, so not on our discretionary spending. So, if you go to supermarkets, the inflation at 5% or 10% before, you just end up paying more using your debit card or credit card. And then if you think about going back to the prepandemic period, the inflation number for consumer spending is 1% to 2%, going forward it’s likely to be 3% to 4%, so they just have a natural tailwind of making more money on top of the digital transformation from cash to contactless to digital. And of course, these companies are very high-margin businesses. So, any way of making more money on the top line would then get translated onto the bottom line because they are operationally geared, being very tech savvy.

At the same time, one area that we haven’t talked about, which is very exciting for us, which is semiconductor equipment companies. So, currently within our top 10, we have Applied Materials AMAT and Lam Research LRCX, and last year we had ASML ASML, which we exited on valuation grounds, but in terms of the end-market, dynamics will be quite similar. So, why is it so exciting about these companies? And we actually categorize the semiconductor equipment companies under the new regime beneficiaries, so not within AI because they don’t actually do a lot in AI. It’s on the back of the geopolitical uncertainties that with the Chips Act in the US, with a similar mandate in EU, there are over $100 billion worth of tech subsidy today to incentivize companies like TSMC TSM, Samsung SSNLF, and Intel INTC to build new foundries in the Western world, because currently, Taiwan controls over 90% of the high-end semiconductors production. And if tomorrow, let’s say, Taiwan is no longer autonomous, let’s say, somehow it becomes part of China and then China ended up weaponizing semiconductors, I mean, we will be in trouble, right? So, the Western government has recognized that two years ago back in 2022 that we need to have more foundries being built so that we can produce the chips that we need ourselves.

And then, if you look at the numbers today, there’s over $300 billion to $400 billion worth of commitment to build new foundries in the US, in Ireland, in Germany, in France, in South Korea, in Japan, so then if you’re selling the mission-critical equipment into this foundry like Lam Research, Applied Materials, or even ASML, then you ended up selling a lot more of your equipment into these foundries. So, this is what is very exciting that like you would think typically the sector in itself is quite cyclical because the end markets are consumer-facing: PCs, smartphone, or autos. But because of the tailwind they’re experiencing on the back of the government backing this dramatically because we think this is the most important reshoring in terms of silicon sovereignty—like we need to be sovereign in terms of the access to silicon—then the government is putting the full force behind the scene. And of course, when you talk about building a foundry, I mean you’re talking about billions of dollars, and it would take at least three to five years in order for a foundry to be completed. So, then we’re actually very excited about this area that we think is going to benefit on the back of this new regime.

Johnson: Stephen, thank you so much for taking the time to speak to me.

Yiu: Thank you.

Johnson: This is Christopher Johnson for Morningstar UK.