Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Under the Banyan Tree Live – China, EVs and the next steps in AI image

Under the Banyan Tree Live – China, EVs and the next steps in AI

HSBC Global Viewpoint
Avatar
359 Plays8 hours ago

Recorded in front of a live audience on 21 October 2025, Fred and Herald are joined on stage in Hong Kong by Jing Liu, Yuqian Ding and Frank Lee for a comprehensive round-up of the major macro and micro talking points in Asia.

Disclaimer: https://www.research.hsbc.com/R/101/zJCkN77

Stay connected and access free to view reports and videos from HSBC Global Investment Research, just search for #HSBCResearch on LinkedIn or click here: https://www.gbm.hsbc.com/insights/global-research.

Recommended
Transcript

Podcast Introduction

00:00:02
Speaker
Welcome to HSBC Global Viewpoint, the podcast series that brings together business leaders and industry experts to explore the latest global insights, trends, and opportunities.
00:00:13
Speaker
Make sure you're subscribed to stay up to date with new episodes. Thanks for listening, and now onto today's show.
00:00:23
Speaker
This is a podcast from HSBC Global Investment Research. Analystifications, disclosures and disclaimers must be viewed on the link attached to your media player.

Focus on Asian Markets

00:00:42
Speaker
Welcome to Under the Banyan Tree, where we put Asian markets and economics in context. My name is Harold Vendelinde, head of Asian Equity Strategy here at HSBC. And I'm Fred Newman, chief Asia economist here at HSBC as well in Hong Kong. And we have a live Banyan Tree recording today, a special event. Harold, who's joining us here

Panelist Introductions and Topics

00:01:02
Speaker
live on stage? Well, sitting next to me is Jing, our esteemed China economist, who obviously is going to talk about what is happening in mainland China.
00:01:11
Speaker
We have Frank, Frank Lee, who is our tech analyst and looks at companies such as Nvidia. So we're going to touch upon the AI bubble. And we have Yu Chen, who is the analyst who looks at electric vehicles and companies such as BYD.
00:01:26
Speaker
So EVs, technology, China, that's going to be the topic for this week. Three key areas to discuss here, three key topics. We're not going to talk about tariffs today. We're not going to talk about what's in the newspaper. We're going to dig deeper right here under the banyan tree. So let's jump right in.

China's Economic Rebound

00:01:43
Speaker
Good. Jing, you look at the Chinese economy. Some of the data that's coming out is not overly exciting. Things are slowing a little bit here and there. So what makes you excited about China?
00:01:55
Speaker
I think it's fair to say China is rebounding. I think the most concerning data is the contraction of investment. But there's a reason behind that.
00:02:06
Speaker
For example, we know that the local governments may have some trouble with their own financial situation. ah So that data may reflect the fact that the local government did not have enough money to go for the capital requirements so they cannot leverage the um other money for investment.
00:02:27
Speaker
That's changing. And on top of that, the government also front-loaded the future quota for the local government debt swap. This point has been under-appreciated by a lot, ah but this is working because we see a lot of companies get paid, they start to normalize their CAPEX and many other things. Let me press you a little bit on this because you say there's actually a lot positive happening as well, yet the economic data is very, very challenging.
00:02:57
Speaker
And i think what people point to in particular is very weak consumer spending. Now, we had just had the golden week holiday, for example. per capita spending actually still down compared to 2019.
00:03:11
Speaker
um Is there any hope that the consumer will start to come back? Because there's so much riding on the Chinese consumer here. ah Are you really seeing the consumer coming back? Or is it more local government spending, more but more infrastructure? Yes, we're going to see a bit of

Chinese Consumption Patterns Shift

00:03:26
Speaker
a rebound. The last couple of years, right? The consumer is missing here. When is the consumer going to step up?
00:03:33
Speaker
I think um basically consumers are not missing here. But what I'm seeing is there's a change of the consumption pattern. There's a change of mentality.
00:03:44
Speaker
For example, now a lot of Chinese, they probably prefer the Banffo Buck, not necessarily go with the designer brand name, which is more expensive.
00:03:55
Speaker
Another thing which is happening is they prefer a lot on experiencing things. So that has made the service-related consumption ah probably more promising going forward.
00:04:07
Speaker
And actually, a big focus of the next five-year plan will put a lot of focus on rebalancing the economy. But what about investments in in technology and these sort of things? That remains a focus, presumably, as well, right?
00:04:20
Speaker
Yes, indeed. ah We have seen, for example, in the past five-year plans, they always set some target in terms of R&D spending and also how to build the ecosystem.
00:04:30
Speaker
And it is really working because in the latest World Intellectual Property Organization's ranking, of the most innovative economies, for the first time China was ranked and number 10.
00:04:43
Speaker
And 10 years ago in 2015, China was ranked at 29th. So in just 10 year time, we see China actually advanced from 29th to 10th.
00:04:55
Speaker
And at the same time, when we measure the input-output for innovation, China is getting in much better shape in terms of the efficiency for innovation. So there is a civil lining and we want to talk a bit more about technology also with our other two panelists here because there's a lot of exciting stuff happening in China.

Real Estate and Equity Markets Impact

00:05:14
Speaker
But let me ask you again, and I'm sorry to come back to consumption on this, because you have technology investment on the one hand, you get great companies coming through, but unless there is spending at some point, you're not going to essentially generate revenues.
00:05:29
Speaker
Now, for consumers to spend, there is negative and positive news. And negative news is the real estate market is still struggling, but the equity market is doing well. Is the equity market rally, which is driven by hopefulness on technology and optimism about technology, is that going to happen a wider impact on consumer spending and the economy? Is that enough to lift everything up?
00:05:53
Speaker
I think it has started to have some wealth effect, especially those who have the stock market account. They start to actually get more money and more willing to spend. At least this is the feedback I have got from our luxury analyst in his recent tour from China.
00:06:11
Speaker
And on top of that, when we look at the consumption, I think in the short term, stimulus matters a lot. In the medium term, it's probably more about ah structural, like ah whether we have a big ah middle income group.
00:06:24
Speaker
and in the longer term it's all about the income outlook and the wealth accumulation. Well, this is a bit important, right? Because it's very nice the stock market goes up and therefore the consumer spends more, but that's a sort of cycle that goes into reverse as well because as the stock market will come down, people wouldn't consume as much anymore, and then we're flat on our face. Harold, stock markets never go down. I learned that from you over the years. I told you that 10 years ago, is that right? Yes.
00:06:48
Speaker
So um what can they do in order to make sure that it's not just a wealth effect, but income growth and these sort of things are, yeah, are supporting that? Yes, indeed. and So we hear China talk a lot about the new urbanization plan.
00:07:04
Speaker
And actually a core of this plan is to improve the social safety net. So to lift the disposable income. So imagine the migrant workers, they used to need to save a lot just in case they need to go to hospital, they need to send their children to school ah who are not um eligible to go to public school maybe previously because of their lack of household registration.
00:07:29
Speaker
But now imagine they have equal access to this kind of public services, then they might be willing to spend more. So this is the direction we're seeing. Another thing that China is doing is what they call in a convoluted way the anti-involution policy.

Anti-Involution Policy in China

00:07:44
Speaker
What is that and where do we stand with that? Right, so anti-involution actually refers to this ah destructive competition. And actually, if I ask my friend who are practicing in competition law, they will tell me this is not even fair competition because oftentimes we see some of the large players abuse their market dominance by squeezing their suppliers. So they're selling under cost and stuff like that, squeezing everybody in the market. Yeah, boycotts, et cetera.
00:08:14
Speaker
And in China, there's a characteristic which may be absent in other countries, is that local government intervention, like the local protectionism, which has prevented the market to work in a more efficient way.
00:08:29
Speaker
but But, Jing, jing so sorry, I need to i press you again. And I know we're friends. We sit next to each other. But is it enough to eliminate deflation in China but purely focusing on essentially eliminating excessive competition? Or is it also excess capacity because we're investing too much?
00:08:48
Speaker
is it Is it enough what they're doing to eliminate deflation just focusing on competition policies? I think these two things are um actually related. Because there is overcapacity, we end up seeing the cutthroat competition.
00:09:02
Speaker
um To resolve the problem, we need to see the boost on the consumption side and also at the same time normalize the production side. Now, I was on a holiday in Uzbekistan in the last two weeks.
00:09:16
Speaker
And what I saw there, the moment I arrived on the way to the hotel, BYD, BYD, BYD, and Geely as well. And then all sorts of other Chinese cars with brands I'd never heard of, EV cars. So if there's one industry where there's oversupply, that's EVs,

Chinese EV Market Insights

00:09:33
Speaker
right? What is happening in the EV sector?
00:09:35
Speaker
Do we see consolidation? There's this anti-involution, this rationalization. Is that actually taking place? Yeah, I would say the game of ice and fire. So obviously, of course, on fire because the China EV product in terms of the product strength is leading the globe.
00:09:50
Speaker
So can you imagine in China with about 10,000 US dollars, you can get a robust EV and it can also self-driving in a highway and do the auto parking. That's the the value and price you are getting.
00:10:03
Speaker
So domestic market, of course, you get over a thousand car models available on market. So because of the overcompetition, the pricing environment being challenging, the market is called for a consolidation. So it's a little bit painful.
00:10:17
Speaker
um On the overseas market, because the product and the technology and the cost structure is so advanced, China would be able to sell the EV model in overseas with twice the price. So basically if you if you go to Europe, you're paying twice as what I'm paying in in China.
00:10:34
Speaker
Yuqin, when we talk about a Chinese EV market, it's it's a phenomenal success. And we all know this seed, even on the roads of Hong Kong, more and more Chinese EVs, fantastic. But there are two elements when it comes to the technology. One is that the cars themselves are really, really good.
00:10:50
Speaker
but it's also the production process which is very, very advanced. Can you speak a little bit to that? Because it seems to me that it's not just that the technology in the car is good, it's also the way they're being produced. It's just phenomenally competitive.
00:11:04
Speaker
is that Is that a right observation? how How good are they actually at manufacturing this stuff? Yeah, I would say ah two things. First is the fastest faster product cycle. Let's say a typical car cycle from a global standard is five to seven years.
00:11:18
Speaker
But in China, if you don't refresh the car every two to three years, your peers would and your market share will be taken, ah which force the model to be refreshed at a much much faster speed.
00:11:28
Speaker
So you get to load all these advanced technology, all these fancy fast charging, flash charging into um into the vehicle. And also, I would say, economic scale. China is one of the largest, I would say, the largest car market ah across the globe.
00:11:43
Speaker
It has the supply chain from producing fancy chairs, massaging chairs, to fridges. Fridges. Karaoke, too. Karaoke, too. Oh, look at that. Yeah, the massive supply chain advantage. So before China EV being leading the globe, China battery is already leading the globe. And you know, battery one third of the bomb.
00:12:03
Speaker
Therefore, China can reach the cost parity much earlier because the supply chain scale and technology innovation because we have a big market and iterate faster. Now, I have a question here. So they are extremely innovative.
00:12:18
Speaker
They got scaled, so they're being cost down. That's all nice and well. But this anti-involution policy means they can't discount that much anymore. That's only helping the smaller companies that that struggle to survive because the bigger ones are not really discounting that much anymore.
00:12:34
Speaker
So it's actually not helping the industry. Is that a fair observation? Yeah, I would say, Jing and I was debating about this when when we're coming upstairs. ah So anti-involution initiatives saying auto in auto industry means better pricing and capacity discipline.
00:12:48
Speaker
That's good for the near term. It's like a painkiller to the over-competition China auto fever. But the problem about that is are still yet before the consolidation point. It has to get worse before it gets better. So I would say ah for the temporary chapter, the consolidation is taking a pause.
00:13:05
Speaker
But eventually it will do. One question on automotive driving. um I went to Shenzhen and I took one of these robo-taxis. It's quite amazing. You plug in where you want to go. i actually went on a colleague of mine who did that.
00:13:18
Speaker
And there you go. And it's it's it's quite an experience. And suddenly you realize, well, this car can be completely ah redesigned. You don't need a steering wheel anymore. And actually you can go further. The whole city can be redesigned.
00:13:29
Speaker
So when is this going to roll out? When are we going to have this in Hong Kong? When is it going to go very fast? yeah I would say now if you will travel to Shenzhen, you will get the real commercial paid autonomous driving mobility service.
00:13:43
Speaker
so real Actually, i yeah I used it for 15 minutes and cost me eight Hong Kong dollars or ten Hong Kong dollars. It was equivalent. I mean, it's really cheap. Yeah, I would say, first of all, it's a real commercial service with no drivers sitting on the driver's seat.
00:13:57
Speaker
Currently only available in a geofenced area, for example, only in the Nanshan district, you can get that. But eventually, we're at the cusp of the ah inflection point of the robot taxi. The fleet will grow tenfold next year.
00:14:11
Speaker
So wellre well we're looking at a couple of thousand this year, but next year it will be tenfold. But still, don't don't be afraid. It's only a fraction, only after um even we have explosive growth, it's still a fraction of the current taxi market, which is over 10 million. But Yuchin, you're an analyst. i need to press you on the forecast here.
00:14:29
Speaker
In Hong Kong, when will all of us here be driving to work in an autonomous vehicle without having to think about driving anymore? Is that is that going to happen this decade, next decade, 2050, 2060?
00:14:40
Speaker
When is this going to happen? when is this going to happen I would say if it would be easier if you don't specify the geography as in Hong Kong area. So in China, I would say daily commute relying on the car who can drive itself from your parking lot at your residency to your office will be happening in the coming six to 18 months. And for the robot taxi, it's already there.
00:15:03
Speaker
and The only problem is about a scaling, and it will be a government-controlled, responsible scaling for the coming decade. But for Hong Kong, I would say, yeah, if the government allow, it will be much easier.
00:15:15
Speaker
Last question. Chinese ah cars, fantastic, cheap, innovative. Are they going to conquer the world? Is there any space left for German automakers, for American automakers, for Japanese automakers? Or are the Chinese going to essentially have a lock on the industry globally?
00:15:33
Speaker
I would say consumers want a good, cheap, reliable product. um That's the ultimate driver. But we're looking at a very complicated global trade environment. Therefore, many of the Chinese OE has localized their production, like being the world's local car, not a being a Chinese EV, but being a global EV.
00:15:51
Speaker
So I think trade barrier obviously is one of the near-term challenges. But eventually, hopefully, the good product will find its way. Good, so we have cars communicating with each other, city planners that can figure out where these cars gotta communicate with traffic lights and with hospitals, everything.

AI Market and NVIDIA's Role

00:16:09
Speaker
This is AI, we're getting to the area of AI and we have Frank here. Frank, you sat on exactly that chair, if I'm not mistaken, two years ago, and you talked about the chipset developed by Nvidia, it was a sort of a gaming thing, cost $200,000.
00:16:26
Speaker
This seems to be the the linchpin in that AI story, right? The technology that's been developed by this particular company. um How is that progress? Where do we stand with that and Sure. I think if you think about the AI story, ah instead of two years ago, we talked about product,
00:16:41
Speaker
how good the chip is, you know why NVIDIA was such a dominant player. This year has been a very interesting point, because you start to see the total addressable market of AI is potentially a lot bigger than I think people expected a couple years ago.
00:16:54
Speaker
If you think about the first two years of it was really about four or five huge major US hyperscalers spending all this money, everybody's still questioning about monetization.
00:17:06
Speaker
But since beginning this year, I think what we've seen is a lot more talk about enterprise and sovereign AI. We had the Stargate project got announced in beginning the year. And a huge difference from beginning the to now, because now we actually have some ideas of when Stargate is going to start to build out. And you start to see you know these open AI announcements.
00:17:25
Speaker
So now we're looking at a framework where the market potential could be 2, 3x of what we originally expected. So I think in some ways, while the product, the chip is important, it's really about how big the market opportunity is now. It looks a lot bigger than people expected just two years ago.
00:17:41
Speaker
So when when you talk about market opportunity, you mean the AI hardware that's needed to build the infrastructure. That's ultimately what we're talking about. but But in that, there's still just a handful of companies that really are essentially hating it dominating, right? I mean, that that dominance is not... ah decline I mean, we have a broader application, people investing in it, but we still have a very small set of companies that are really at the heart of the technology. Is that a misinterpretation? i mean, NVIDIA must be the poster child of the year, but who else is is is really driving this?
00:18:13
Speaker
Well, I think what you've seen is that, you know, well, yes, NVIDIA is the center of this. um There is a lot more opportunities, especially you know on the hardware side in the last probably 12, 18 months.
00:18:26
Speaker
You've seen more competitors coming in. ah You have companies like Broadcom, AMD now who are competing more. than You also have more component companies. on the hardware side. and And even in China now, CameraCon has been a big focus this past year because of the ability to make chip for the China market. So I think you are seeing a much more broader-based universe than what we saw probably you know the first year, which is basically one company.
00:18:49
Speaker
So NVIDIA is still the industry leader, sets the standard, if you want to put it like that. The story is that actually over the last 12 months that China has caught up very, very well. Where do we stand on that progress? ah Would you say 12 months ago, they would be a year behind and now it's...
00:19:04
Speaker
six months I think if you look at this, it's a different question because China is definitely making progress but in the grand scheme of investment in AI and the market size, China's not that big. right um And I think if you think about you know the size of the deals right that are being announced right now, the total spend on China is still not a big number.
00:19:27
Speaker
you know The spend outside of China is much bigger. Is that because DeepSake is much cheaper than OpenAI? I think what it is is that you have the scale of these investments in the US is so much bigger, especially now with sovereign.
00:19:43
Speaker
I think what China does is that China is actually more profitable because they think about monetization, i think much earlier in the stage. I think with the US, monetization has not been as critical because the access to capital,
00:19:55
Speaker
has enabled these companies to get so big without really thinking about profitability so quickly. So I mean, you feel like an open-minded is the best. That sounds like story we've heard before. I need to pick up on that. Yes. Because here's a big question.
00:20:09
Speaker
Yes, you're saying the addressable market is bigger and suddenly we're talking about hundreds of billions invested and these numbers keep rising, rising, rising. But there isn't necessarily the profitability yet. There is a question about what the return is.
00:20:24
Speaker
Isn't this a giant bubble at the end? I knew that question was coming. yeah I think you can definitely say that there are certain things that, to be fair, do make people a little bit more nervous or questionable about the ah potential bubble, especially the last few months with some of these deals that have been announced.
00:20:45
Speaker
But I think you can also look at it from a different point of view is that If you look at some of the companies that are at the heart of ai investment, they are generating real money. If you look at NVIDIA, they're making real money. So I think, is there a bubble? Yes. The question is, are what stage or inning are we in? and Are we in the seventh inning, so you know something that's going to crash right away? Or are we still you know a few years out in terms of this? Because the one thing that everyone has definitely underestimated is how much demand has increased.
00:21:12
Speaker
ah How much computational demand has increased? Just think of this example, SOAR 2, that OpenAI has launched, the video of people playing with that, but the amount of computation that goes into that is exponential.
00:21:24
Speaker
So that's what drives, I think, this continuous investment is that you know The computation you need just keeps getting bigger and bigger. But is that is that real demand? Because you you're talking about there's a lot of demand for these chipsets, right?
00:21:36
Speaker
It's the service that you need. But it looks to me that companies are very careful in rolling out AI in their own businesses, unsure how that impacts their business and also if what what the impact of AI is on their business.
00:21:49
Speaker
So it looks like there's caution in demand, on the demand side, and there's just a lot of supply coming up. I think you know this year has been interesting, like I said earlier, because it's not just four or five huge companies, these huge cloud companies that have lots of cash and a strong balance sheets that's funding it.
00:22:06
Speaker
Now you're talking about governments getting involved, right? It's a sovereign market, right? and And this has always been very difficult to, and number one, think about how you commercialize it, and two, how big the scale is.
00:22:18
Speaker
I think what you're starting to see now is some idea of scale because it's based on all this gigawatt announcements have been made. But this brings up another interesting point is, do we have the grid system? Well, this is my second question. So you can have, so maybe I'm wrong with demand. There's actually a lot of demand and there's a lot of supply as well.
00:22:33
Speaker
But then the limiting factor is power because you need an enormous amount of power to run these servers, right? Well, I think that's that is a fair question. I think that question will definitely become a more relevant one probably in the next 12 months, because 12 months from now is when a lot of these sovereign projects really

AI Infrastructure and Power Constraints

00:22:50
Speaker
kick off.
00:22:50
Speaker
But I think the thing is, the one thing that's interesting, though, is that you have a lot of guys who are in crypto mining are now looking at converting some of that power, which they take a lot of power, into driving the data centers.
00:23:03
Speaker
um Frank, maybe a last question for you. um Now, Harold talked about some of the risks, which is energy constraints, for example. Another risk must be the speed with which technology evolves. And um if you think about the dot-com boom, some of the companies that were you know hot back then weren't around a few years later.
00:23:23
Speaker
Is there a risk? I mean, we think of... companies like Nvidia being a complete leader, but doesn't a technology evolve? Could we talk in two or three years about another chip company that suddenly has much better technology?
00:23:35
Speaker
Or is this different? is it Is it so hard to catch essentially the know-how, catch up with this know-how? How how how dynamic is this market ultimately when it comes to the top-end chips?
00:23:46
Speaker
I think if you look at, and another way to think about this perspective is this decade is very clearly about AI data infrastructure built. So I think it it feels as though like this will continue probably until the end of the decade.
00:24:00
Speaker
The question is beyond that, what happens, right? is it Is that really becomes the time for robotics to come in? Because you need like a certain maturity on the AI level before you can really implement the AI and the robotics together, right? So that that could be something the next big thing is is into the next decade.

Optimism in China's Economic Future

00:24:16
Speaker
Now I want to wrap it up a little bit here. So Jing is quite optimistic that some turnaround is taking place ah in the yeah in the Chinese economy and that the consumer actually could well be leading this instead of investment. So there's a change there.
00:24:30
Speaker
In the EVs, we see significant changes as well going towards autonomous driving, Battery technology is at the forefront of that as well, which, funnily enough, might actually be at the forefront of ai as well, because demand is substantial, supply is substantial.
00:24:45
Speaker
But one of the limitations is, can we actually manage that and empower these things? And therefore, the Chinese might, to certain extent, play a role in ah in in that particular field in AI as well.
00:24:56
Speaker
Fred? I think you said it beautifully. I think despite all the uncertainty, I think to me in this world with tariffs, geopolitical risk, this conversation was actually a lot about optimism about the future. but and Even if you only buy half of Frank's story about the potency of AI, that's still very bullish.
00:25:14
Speaker
um Even if you buy half the story about electric vehicles, that's very bullish. And even if half of China starts to consume, well, that's still very bullish as well. So maybe it's a good time here to wrap up. We want to thank our panelists for joining us here under the Banyan Tree Live. ah Jing, Frank, and Eugene, thank you very much for your insights.
00:25:35
Speaker
Thank you very much. Thank you. Thank you. Thank you. Thank you, guys.
00:25:59
Speaker
Thank you for joining us at HSBC Global Viewpoint. We hope you enjoyed the discussion. Make sure you're subscribed to stay up to date with new episodes.