Transcript
Speaker: Welcome to Under the Banyan Tree, where we put Asian markets and economics in context. I'm your host, Harold van der Linde, in the Hong Kong studio here at HSBC. On the podcast today, the blockbuster IPO putting Chinese chipmakers in the spotlight. CXMT, that's the name of the company, has been crowned China's most valuable listed company after its explosive debut in Shanghai on Monday.
Speaker: Is this the moment Chinese chip makers have been waiting for? How did it happen? And what does it mean for the established giants like the Korean players? Joining me to help answer those questions are Global Head of Tech Hardware and Semiconductor Research Frank Lee and Asia Economist Ines Lam.
Speaker: Let's get the conversation started right here under the Banyan Tree.
Speaker: Let's start with some numbers around this massive IPO. CXMT, the Chinese memory chip maker, saw its stock skyrocket 466% on debut this week. So it went up almost five times in what has become the largest Asian listing of the year. Its market cap was equivalent to around 85 billion US dollars during the IPO process. After Monday's rally, it was close to 490 billion u dollars. George really serious player in this industry. Ines, where does this company come from? it hardly existed about 10 years ago, right?
Speaker: Yeah, that's right. CXMT was largely unknown until very recently. There was a supply gap in China's domestic memory chips.
Speaker: Global memory market has been traditionally dominated by Hynix and um Samsung in Korea and Micron in the US. But then about 10 years ago,
Speaker: A Chinese entrepreneur, he identified this supply gap in the memory chip in China and he has decided to go all in in making memory chips. And what's ah what's making this story so remarkable is that this company has suffered years of financial losses. But thanks to the global AI boom that has pushed up memory prices, this company has reversed its losses and is now making you know a lot of money. it's It's not just this entrepreneur who's been willing to put his own savings into this because it seems like the government was behind this as well. It's called the Hefei model, right? That's what some people refer to. Can you explain what that is? Yeah, absolutely. Hefei model is now gaining attention again because... is the primary supporter behind this company that has contributed to its massive success. And the primary supporter being the local government. That's right. So what's unique about this Hervé model is that the local government acts as a venture capitalist. It provides financial support to a new company at its early stage in a sector that is identified as a strategic industry in China. And not only financial resources, but it also helps build the supply chain ecosystem around it and helps connect or link up suppliers and customers for the company. So it's like all-round support. Just for those who don't know, Hefei is basically a city, I would say right spot in the middle of of China, correct? Yeah, that's right. Hefei is the capital city of the Anhui province in China. yeah um
Speaker: This sounds like a story that we've seen in many other industries as well. EV, solar, you name it. Is that fair to say? Yeah, it's fair to say. And the Hewei model has created a number of success stories already. For example, NIO, the EV company, was struggling a few years ago. But thanks to the support from the Hewei government, it's managed to survive and thrive. And another example is BOE. It's an industry leader in LED and OLED display space. So come from the same province, basically. That's right, yeah. I guess what is a little bit different, though, here is that, for example, in EVs, every province had its own EV. So at one point in time, we had like 75 electric vehicle companies in China. And they all had to compete. And a couple of these players emerged as the winners,
Speaker: NIO being one of them, and there's a couple of other companies like BYD, and they're now exporting, right? But I guess in the memory space, it's been really one company here, right? Yeah, but it's mostly CXMT. It has another company that is in the same memory space, but CXMT has proven to be the leader here. Now, we've seen in many industries, I'm going to go to Frank here, we've seen in many industries where the Chinese emerge from shipbuilding and steelmaking to...
Speaker: Now we see memory, but also in the car industries, there's a lot of government effort behind it. You create domestic champions, they then start to export, and very often in some of these industries they've taken over. So does this mean that companies like Samsung and Heinex are going to be out of business in the by the end of the year? I think that's a little bit premature. I think yes i think what we're seeing here is just that it's a perfect storm for CXMT's listing. You're coming off basically one of the biggest memory super cycles in recent history because of AI. And I think you know if you think about why are people so excited and talking about this, because...
Speaker: Chinese investors have been seeing global investors participate in this memory trade for the last nine months, whether it's the Korean, you know, obviously the Korean market, you know Micron in the US. So finally, you have, you know, their own version of a memory company that they can participate in, which was what I think drove this huge interest in the IPO. yeah But I think if you look at more carefully at the nuance of specifically what's happening CXMT is, we think, going to be a significant player when it comes to the commodity memory, the memory that goes into smartphones, that goes into handsets, right? And as we know, the end market isn't good there. It yeah hasn't been strong. But they've benefited because you know the strength in ai servers, the HBM memory, has resulted a shortage even on the commodity memory.
Speaker: So this has allowed CXMT to post really strong earnings. Yeah. So I think if you're talking about where they could impact the market, it's probably in the consumer-related memory, right? um you know Are there going to be a challenge to the ah Koreans on the very high-end HBM memory that goes into servers? I think it's not going to be that easy. I mean, there's going to be an effort to try to localize and do that, but it's it's not going to be easy. So I think their bigger impact is going to be on the lower-end commodity, which for the Korean makers and some of the guys like Micron it's not such a big part of their revenue.
Speaker: Yeah. So it's not a near-term risk, and we'll just have to figure out to see how this works out with with this company in the longer run, right? I mean, you could say, well, look at BYD. These companies, they're now taking over and could become the biggest car company in in five years' time, so don't rule them out. On the other hand, I remember being in Taiwan 25 years ago, and people said, oh, my God, you've got SMIC making...
Speaker: similar products as TSMC and by 2010 they will have overtaken TSMC and they haven't. So it's in certain industries it's not as easy as you think to take over from the leaders. from fear i think I think if you think about what's different about semiconductor manufacturing is that the pace, right, the new product development, the next generation is really quick and it's hard to make that leap to keep up. So you brought up a good example. SMIC, they're still very behind TSMC, right, because the One, it takes a lot of money to invest in each generation. right So you know you're not going to have 50 companies doing memory in mainland China because the cost the barrier to entry in terms of cost capital is just too high.
Speaker: But I think the main thing is it's a race. right By the time you catch up, the market has moved on to the next generation, yeah which incrementally is harder. right I think EV is fast paced, but definitely not at the pace that semiconductor is. Yeah, it's taking place. Still, if I look at the sheer prices of, for example, some of the yeah leading Hyundai Samsung semi-universe, but also in the US, it's come off quite a lot over the last, I would say, since late June or something like that. They've come off quite a lot. Is that because of worries, you think, of of of this? i I think new competitors coming in and changing the landscape. I think it's um i think what's hit the memory makers, ah two things. The first thing is that they started entering these long-term agreements a couple of months ago. So that basically gives the market a message that the upside in terms of further price increases is going to slow down.
Speaker: Long-term agreement, you mean the demand for the memory is so strong that they said, okay, you want to buy our stuff in 2029, we're willing to sign a contract now. Something like that, right? Yeah, but also I think the more important issue, especially in the last couple of weeks with you know all the tech stocks being down... is this you know focus on CapEx spending again, on an AI CapEx. And you saw that Google last week, first time we've seen negative free cash flow. no So essentially that means that you know these hyperscalers, which have been driving the entire AI trade, they only start to tap the outside and debt markets, equity markets. right and And I think that is the bigger impact. on the sale off. I don't think it's really a fear of the Chinese you know coming coming in. Okay, so we have worries about the capital that this hyperscale has put in place in the US. We have the price increases, which were dramatic at the beginning of the year, starting to peter out. But ah i guess a third factor in here is that we've seen the proliferation, in particular in Korea, of all sorts of leveraged products. So you can buy
Speaker: in Korea, a leveraged ETF that is two times. So that means that if the Heini share price goes up 5% or 10%, your product will go up 10% or 20% on that particular day, depending on the movements. um And they do that These funds, they borrow money. and But it also means that if the share prices go down, they've borrowed a lot. They need to sell a lot, therefore, as well. So that amplifies share price movements in in the markets. And I guess you get fear in the markets and therefore people wanting to unwind these positions. And it's it's these sort of dynamics that have just added to the volatility. that fair to say? Yeah, i think for sure. i think one of the reasons that people willing to take on this kind of risk in terms of these leveraged ETFs is that fundamentally the view was that memory was going to be tight all throughout 2027. We didn't really see much new capacity coming into the market at least until the end 27, sometime in 2028. So there's this confidence that you can't go wrong yeah with these memory stocks right now. right the The industry outlook looks bright. um content growth, no supply coming at least and until the end of 2027. Let's up the mortgage and put the money in there because it's a good bet. That's that's the sort of situations that have that have unfolded, in particularly in Korea, but in Taiwan as well, right? Yeah, I think Taiwan also, loan margins were all-time peak. And I think part of the reason if you look at Taiwan is interesting is that
Speaker: The property market, I think it's tied to that because if you look at right now, um the government's trying to slow down the property market. So if you're gonna buy a property in Taiwan that's more than 70 million NT, you need to put 70% down payment. yeah And so no one's really buying houses in Taiwan. So what they're doing is that they're basically leveraging their house, using the house as collateral to borrow money from the banks, because the banks aren't lending out mortgages. And put that then into the stock market. yeah yeah So that amplifies these situations. Do you also see a split in these markets? We see split in certain markets whereby you have US technology and a Chinese technology. You have the EVs coming out of China and then you still have existing models that are being sold in the US. We see this in other industries as well. Internet is maybe a nice example. um Do you think this is going to happen in in memory as well?
Speaker: Yeah, that's a really good memory question. I think it's probably a bit of both, right? because On the high-end memory, you know, the HBM memory that goes into servers, into the data centers, I think that's probably right. I think you're going to see polarization, you know, U.S. market, China market. I don't see U.S. hyperscalers buying memory from Chinese makers in this market.
Speaker: Where you could see, I think, that kind of the the overlap is on a lower-end commodity memory that goes into smartphones, that goes into PCs. This is not an area where there's a lot of sensitivity, right? I mean, we've already seen, i think, media reports that Apple is petitioning to use Chinese memory for their smartphones, yeah right? And so I think it's so it is probably going to be, i think, a bit of both in the memory space.
Speaker: Ines, I just want to circle back to you again. So we've seen the surprise emergence, you could say, over the last decade of this company CXMT. Are there more surprises coming out of China? Are there other provinces or companies that are emerging and will list and also potentially impact global industries? I mean, I'm sure there are plenty of Chinese tech companies like working in the background, but they are they're all a bit secretive. so But I think we'll see more surprises coming out of um China's tech space because the Chinese government is very aggressively pushing for self-sufficiency, especially in AI and semiconductors. So yeah, I would say watch this space. so So basically what we've seen, the Chinese government has put its weight behind a new memory player. It has emerged very rapidly. And as Frank, you rightly pointed out, it attracted a lot of attention in China because those people couldn't buy it anywhere else. The worry in the market is that, hey, are they going to push the existing players away? Are they going to take over? It does seem, least Frank, that's what you're suggesting, that this probably will take some time. They're still at the lower end of what is called memory, not the very high end of what the stuff that the AI needs. But these stocks have started to roll over in semi-stocks ah around the globe. And that might well be also related to factors such as
Speaker: The price increases are not as spectacular anymore. And the amount of money that's being spent is, yeah, and they need to borrow for that in the US to to build these hyperscalers, these data centers. And that there's a bit of a question on that as well.
Speaker: And that might all be amplified by some of the sort of financial instruments that have been built around this as well, leveraged ETFs, et cetera. So it will remain a sector that is going to be very dynamic and super interesting. So hopefully we can ask you to come back at some point in time to give us an update on this.
Speaker: Thanks a lot, Ines and Frank. Thanks, Harold. Thanks, Harold. Thanks for having us.
Speaker: And we're going to have to leave it here, ladies and gentlemen. Great having you with us, as always. And do join us again at the same time next week. Until then, you can listen to our sister podcast, The Macro Brief, bringing you insights and analysis on the major global stories driving markets.
Speaker: Under the Banyan Tree is an HSBC Global Investment Research production. Our producer is Graham McKay. Take care for now.
