On this week’s edition of the Trivium China Podcast, host Andrew Polk is back with a gaggle of Trivium analysts to go deep on various aspects of the recently released 15th Five Year Plan (FYP).
The group covers a broad range of issues including:
Key macroeconomic, development, and economic security goals
The plan for industrial upgrading – and key priorities in upstream, midstream, and downstream industries
The nexus of technological and industrial ambition
How climate goals and industrial policy interact
China’s latest efforts to solve its compute problem
How innovation factors into goals for the healthcare and agriculture sectors
Just like the FYP itself, this conversation is wide-ranging and hits on a bunch of key policy themes that matter – both now and in the years ahead.
Transcript:
Andrew Polk: Hi, everybody, and welcome to the latest Trivium China Podcast, a proud member of the Sinica Podcast Network. I’m your host, Trivium Co-Founder Andrew Polk.
And we’ve got another special pod for the listeners today. We are still on sort of Two Sessions outcomes, looking at the broad takeaways, both from the Government Work Report that was released last week, as well as the 15th Five-Year Plan. As I mentioned last week, we did sort of a quick rundown of high-level takeaways from the group at Trivium, and we’re going to go a level deeper on some of the analysis this week.
So, we got a bunch of Trivium folks on. We’re going to break the pod up into two parts, kind of like last time, mainly due to time zone issues. To start off, I am joined once again by a couple of Trivium pod regulars, and that is our Head of Markets Research, Dinny McMahon, and our Head of Supply Chain and Critical Minerals Research, Cory Combs. Dinny, how’re you doing, man?
Dinny McMahon: Good, mate. All the better for seeing you.
Andrew: Yeah. Always love to hear that. Always love to hear that. Cory, how about yourself, man? We missed you on the pod last week. You’re performing your civic duties at jury duty.
Cory Combs: Yes. It was quite fun to get out of a full day and then be asked, “What’s happening with this thing that happened while you had no access to a phone?” So, it’s been a fun week of catching up.
Andrew: Yeah, yeah. Well, it was a good experience for you. Glad to have you back, though, ahead of your trip over to the East here soon. I think you’ll be in China here soon, and I’ll be in China here soon. So, exciting stuff is coming down for both of us. As I mentioned last week, we’re going to, like I said, do a little bit of a deeper dive on key takeaways from all the things from the Two Sessions. We set the stage last week. Unfortunately, we don’t have Kendra on this week to talk about some of the tech stuff, which is hugely important in the 15th Five-Year Plan, but we will have tech angles when I talk with the Asia team on the latter half of this pod.
We will talk about, again, sort of macro goals that China has set. And then we will also talk about sort of the main ways China is looking to achieve those, which is technological upgrading and industrial upgrading. So, we will talk about this stuff today, specifically from kind of Cory’s angle. Obviously, listeners will understand the tech, the critical minerals, industrial upgrade, and this stuff is all layered together in China’s view. So, we’re going to talk about it from sort of mostly Cory’s angle today, but we’ll be sort of examining all these policy dynamics on an ongoing basis.
And I think that the thing to understand is that technological upgrading tech self-sufficiency remains as the sort of the core component, but there are growth elements of that, there are industrial elements of that, there are critical minerals materials, all of those stuff, it all fits together, and so we’re going to get them to different aspects. But before we do, I bet Dinny and Cory may have forgotten about this, but we have to start with our customary vibe check. Cory, I hear you laughing, how’s your vibe, man?
Cory: I’m excited. We just wrapped several big projects right ahead of travel. I’ll be in Hawaii, then I’ll be in… for work, by the way, for anyone who’s just like, “Oh, he’s just disappeared to the beach.” Yeah.
Andrew: So he says, yeah.
Cory: Just for a few days. And then Japan, where I’ll actually be on vacation a little bit. So, a bit in Japan and then on to China for the next through April. So that’ll be very exciting to be back.
Andrew: Yeah, good stuff. We’re going to overlap in Beijing for a few days, which will be nice. You have to fly to the other side of the world just to see each other. Dinny, how about yourself, man? How’s your bad?
Dinny: Oh, mate. I mean, it’s the change of the seasons. And so, the kids have been bringing bugs home. And I thought I managed to avoid it, laid out everybody else in the house. But yeah, there’s no avoiding the plague. So yeah, it’s my turn. So, once we’re done here, I’m climbing back into bed. So, my vibe is stupefied.
Cory: It’s a good vibe.
Andrew: I love it. Well, yeah, stupefied. That’s why people tune into the pod.
Dinny: Oh, no, no. I’m bringing my best material today. I’m hopped up on Advil.
Andrew: Oh, I believe it. I believe it. It’s like I always talk about Jordan with the flu, that game, I forget which late ‘80s, early ‘90s, when Jordan had the flu and just played lights out. Sometimes when you’re sick, you have your best performance. So, we’ll be expecting that today, Dinny. My vibe’s also pretty pumped. I’m heading off on a three-week trip starting on Monday. I’m going to be in Spain and then China and then in Indonesia. So, doing the whirlwind tour as well. Excited for that and excited for this conversation. But of course, before we get fully into the content, we also have to do the quick housekeeping. First, of course, a reminder that we’re not just a podcast here.
Trivium China is a strategic advisory firm that helps businesses and investors navigate the China policy landscape that, of course, includes domestic policy in China, which we will be talking about today, as well as policy towards China out of Western capitals like D.C., London, Brussels, and others. So, if you need any help on that front, please reach out to us at hq@triviumchina.com. We’d love to have a conversation about how we can support your business or your fund. Otherwise, if you’re interested in receiving more Trivium content, check out our website, again, triviumchina.com, where we have a bunch of different subscription options for China policy intelligence.
We’ve got paid options, free options, updates that look at the technology space, the market space, general China watcher space for executives who need to stay on top of what’s happening policy-wise on the ground. So, check that out. You’ll definitely find the option you need in terms of staying on top of China policy developments. And finally, please do tell your friends and colleagues about Trivium and about the pod. Word of mouth recommendations really help us out both to grow the business and to grow the reach of the podcast. So, we appreciate any efforts on that front from our listeners.
All right, let’s get into it. Dinny, I’m going to start off with you again today. You talked a little bit last week about how both the government work report and to some extent the 15th Five-Year plan are a little bit listless when it comes to figuring out exactly what it seems that Chinese leaders want in terms of the overall pace of growth. What are we trying to achieve here? We talked about the growth target sort of being 4.5% to 5%, or above if we can. But I just wanted you to talk today, after having had a week to dig into this stuff a little bit further, beyond just the headline growth target. What are the big economic goals you feel like leadership is trying to achieve both this year and beyond? If they’re wayward on the growth target, is there some other sort of clear economic framework they’re trying to achieve?
Dinny: Yeah, so I mean, the starting point has to be this growth target, right? So, as I spoke about last time, the Government Works Report supposed to… it exists to give guidance to the whole economy about the direction that they expect the state, local governments, state-owned enterprises to kind of be moving lockstep towards. And this 4.5% to 5% range, but we’ll be striving for better. I mean, I saw a government advisor say that this is a target with a floor, but without a ceiling. I think it kind of sums it up really well. It’s like when you’ve got a floor, which is 4.5%, which is 10% lower than the target for the last three years, and then at the same time, the Premier is saying, but actually, we’d like it to be higher than the target for the last three years, what are you telling people?
I mean, what are you striving for? And I think ultimately, the only thing that really matters from that sentence, the target is 4.5% to 5%, but we’re striving for better, the only thing that matters is the back half. They are striving for better. They want something faster than 5%. But it kind of begs the question, how are they going to get it? Because prior to the Government Work Reports, all the provinces have their own get togethers, their own legislative meetings and their own work reports. And I think we had 22 provinces that put out their growth targets, and 14 had set targets, growth targets slower than last year. And then all the others, most of them were kind of setting the target the same as last year.
So, prior to the premier saying, “Hey, we want faster growth,” most of the provinces were expecting things to slow down this year. And there was good reason for that. And domestic demand has been a drag. Investment last year was a drag. And the thing is the Government Work Reports or the NDRC report or the Ministry of Finance budget report doesn’t really address any of these issues in a meaningful way. So, if you look at the reasons that demand is being a drag, I won’t get into the details because we talk about them a bunch, domestic demand is dragging because of property, the impact of property on people’s wealth, because local governments are in a fiscal hole, and so they’re pursuing austerity policies.
Because overcapacity is undermining job security, and to a lesser extent, we’re starting to see the impact of China’s souring demographic profile impacting certain income groups, which is sort of undermining their spending. So, you put all of that together, and you’ve got weak domestic demand. And there wasn’t really anything in the work reports that kind of take any of that on head-on. I mean, Beijing has passed, seems to be, when it comes to property, has seemed to sort of resign itself to the fact or to the position that there’s no point in artificially inflating this market. The best thing they can do is let it burn out, run its course, and then rebuild from there.
And up until the point that happens, then the burden on softening the blow and lessening the pain falls on local governments. Local government financial distress. Beijing said exactly the same thing — This is a local government problem. No bailouts are coming. We’ll transfer a bit of money, but you’ve got to work out this thing yourself. So, in terms of turning around the weak domestic demand issue, there really wasn’t anything going on in the Government Work Report. And that was the same thing with investment as well. Investment last year, property investment declined 17%, and it accelerated in the second half of the year. Fixed asset investment for the manufacturing sector rose, which is good, but only 0.6%. Now, that’s been a major driver of investment over the last few years. I mean, the sheer scale of money that’s gone into expanding China’s industrial capacity since COVID, even during COVID, I mean, that’s been a big driver of growth.
Last year, 0.6% growth. I mean, that was a big change. And then infrastructure investment declined 2.2%, which is the first time in a decade. Now, usually, Beijing treats infrastructure as the lever it pulls if growth in other parts of the economy is weak. So, usually what you’d expect is it sees weak domestic demand, and it sees weak property growth and it sees weak investment in manufacturing, and it goes, “Right, investment in infrastructure. We’re going to do more of that.” And that’s what we were expecting from the Government Work Report. I mean, it’s been making all these noises over the last few months that there’s a big gap in the stock of infrastructure between China and developed economies, and that gives them plenty more runway to invest in infrastructure.
And we thought it was a pretty clear signal that we were going to see more money going this year. And we might, but you can’t tell that from what was disclosed in the government work report. They said, “Oh yeah, more of the money that local governments raise from issuing special-purpose bonds will go into infrastructure.” That’s great, but how much more? They said that the policy banks will spend an additional 300 billion RMB on infrastructure. That’s great, but what the policy banks invest in, it’s just not normal credit. What the policy banks do is they provide seed capital.
They’ve been doing it for the last few years because the local governments haven’t been able to do it themselves. So, there’s a question here is, is that extra $300 billion just trying to fill a hole that the local governments can’t be relied on to fill anymore or does this represent a real expansion in infrastructure? So, this is kind of where we are.
The reports aren’t giving us any indication of greater support for domestic demand or investment. And so, how and why does Premier Li Qiang think growth of faster than 5% is reasonable? I think this comes straight down to exports. I think Beijing looks around and goes, “You know what? What we pulled off last year, we can pull off again this year.” And that’s partly because of the trade relationship with the U.S. has improved because they handled trade conflict with the U.S. so well last year with the way that they used rare earths. And I think they look around and they see demand globally for some really important Chinese exports is rising. I mean, global energy demand is expanding and China has been central to helping anyone sort of build out their energy generation capacity.
Demand for Chinese cars, for semiconductors, for ships, it’s all on the up and up. And so, you’re in a situation where demand is rising, and that will support investment as well. And I think Beijing looks around and goes, “You know what, we can achieve 5% plus growth mainly on the back of global demand for our products.” And I think that’s where we are.
Andrew: Yeah, well, and that view would be supported by the fact that data this week, just I think yesterday, showed that exports for the first two months of the year grew at over 20%, right?
Dinny: Yeah, a mic drop.
Andrew: And so, exports are absolutely humming at the beginning of the year. And, as Joe Peissel, our lead macro analyst wrote, it’s very diversified, right? That’s still in the face of an 11% drop in exports to the U.S. You’re getting gangbusters growth to Africa, to large parts of the Global South, also to the EU. You’re also getting primarily large growth in high-tech exports to high-value-added goods. And so that diversification builds, not only offsets the weakness from the U.S., builds resiliency to future trade shocks. Now, I think a lot of people looked at that data and said, “Well, this isn’t sustainable.” Well, guess what? We’ve been saying that Chinese export growth is not sustainable for, I don’t know, five years now.
And just every single year, it seems like, oh, they can’t pull that trick off again. And yet they clearly intend to. So, it’s a great observation. I think that makes sense that that’s the plan. Hey, if it ain’t broke, don’t fix it. Let’s stick with what’s working.
Dinny: I think the Germans think it’s broken.
Andrew: Well, yeah, fair enough. I mean, but the German government might, but you know what? The German people are still buying all this stuff, right? Like the Europeans are buying all this stuff. And yeah, the European Commission doesn’t like it. And the various state governments in Europe don’t like it. But it’s happening. And there’s really no steps they’ve taken concretely to stop it. We keep saying geopolitical tensions are going to rise, people are going to be pissed off, but the consumers are voting with their wallets, right? And I think they will continue to, and China will sort of stick with the path until some of these other governments decide to really do something about it, if they ever do.
And so, that’ll remain a question. It certainly remains a downside risk, right, if these countries start pushing back. But one thing I want you to touch on a little bit further, I don’t know if you have some of the numbers at hand in your head or on your screen, but we’ve written a little bit about the mixed picture on the fiscal side. You talked about the special purpose bonds, how the Government Work Report said we’re going to spend a bigger proportion on investment, but we don’t really know what that means. We wrote a little bit today about how there’s more money for infrastructure, but it’s not clear where it’s coming from. Some from the policy plate, some from others. We’ve written about, Even talked through the fiscal deficit that hasn’t officially increased.
And the point is, which you kind of already touched on, and this is all before we move to Cory, the fiscal stance doesn’t align with this ambition to hit 5 or surpass 5% growth. And I’ve talked with Gerard DiPippo, who’s been on the pod before, who’s over at RAND, who’s an econ analyst, global econ analyst, focuses on China a lot. I mean, the numbers just aren’t there on the fiscal side for a lot of this stuff. So, I don’t know. How do you think about all of that in terms of the official fiscal stance when you look at the hard numbers versus just Li Keqiang’s like, we want to surpass the growth target? How are those at odds? Or how should we be thinking about it?
Dinny: Yeah. So, in terms of borrowing, special purpose bonds, which are local government, special treasury bonds, which are central government, same as last year. In fact, I think there’s marginally less special treasury bonds than last year because they’re not doing a particular… Anyway, for the most part, we’re not seeing an increase in borrowing. The budget deficit is the same as last year, 4% of GDP. So, of course, if the GDP increases 5% or let’s say it’s 3.5% in nominal terms because of deflation, then you know the amount of translations are higher borrowing. I think perhaps the most interesting additional source of revenue growth will come from taxes because last year, tax revenue only increased 0.8% whereas this year they’re targeting 2.9%.
So, that’s meaningfully more than they brought in last year except, of course, they want the economy to be growing at 5%. So, an increase in 2.9% of tax revenue, that’s pretty poor. The other thing to watch will be, of course, dividends, remittances from SOEs. Really interesting detail that our colleague, Wenye, dug out in Shanghai is that last year, the tax revenue target for the full year was 3.7% growth. They got nowhere near that. They got 0.8% growth. And it seems as though they made up the difference with remittances from state-owned enterprises, specifically centrally-owned state-owned enterprises.
There was a really big increase in the amount of money that went into… it’s complicated in China, the government has four different budgets. And so, the budget that takes in dividends and then sort of disperses it mostly back to SOEs as capital injections and subsidies, that really grew quite aggressively. And then rather than all of that being distributed back to SOEs, because usually that’s what happens, not all, but a big chunk of it, a significantly larger portion was then moved over to the Ministry of Finance and used in ordinary budget expenditures. So that’s kind of how they squared the circle last year. Tax revenue growth was far slower than expected. And so, they leaned on the standard enterprises to do that, to sort of fill the hole.
So, the question is, do they repeat the trick this year, right? So, they’re certainly expecting higher tax revenue growth than they delivered last year. They are also talking about an elevated level of SOE remittances, lower than last year, but certainly higher than what’s typically been the case in the past. So, we’re getting more sources of revenue here and there, but ultimately, the key thing to compare it against is what compared to last year. And so, I think So there’s definitely going to be revenue growth, but it’s not the sort of increase in scale that you would expect if you’re trying to deliver 5% growth. So, yeah, I’m not quite sure how this is all supposed to come together. But this is usually not how they go about delivering on an ambitious growth target.
Andrew: Yeah, well, lots of questions up in the air. I mean, in the past few years, they have left themselves a little bit of room to add fiscal firepower later in the year. It just doesn’t seem like that’s sort of their approach this year. And I guess, you know, the interpretation is just, well, we’re going to sort of keep the settings where they are and hope exports continue to pick up the slack. So, we’ll see if that comes to fruition. I think it’s pretty clear from our side, the economic plan in the short term is a little bit muddied. And maybe that’s because they are spending more time planning for this longer term industrial upgrading, the big economic transition to a new growth model, as we’ve talked about many times.
And we’ll just see kind of how it all plays out. But while we’re watching that, I do want to bring in Cory to talk about sort of a level deeper in terms of some of the macroeconomic goals, we’ll call them industrial goals. And here I also have to say, Cory did ping me on Slack while Dinny was talking to point out that I called the premier of China Li Keqiang, which is obviously incorrect — with the Premier being Li Qiang. So, before all of our listeners email me about that, we clocked it. So, hopefully, you can accept this as a mea culpa. But with that…
Dinny: Sorry, quick quiz, Andrew. Who’s the president?
Andrew: Yeah, um Hu Jintao. So, yeah, this is the danger of podcasting on a regular basis is your capacity to misspeak comes to the fore very often. But anyway, Cory, at this point, I want to bring you in. Talk to me about the next level down. We’ve said from the beginning, like tech goals are very clear here. Industrial goals are also very prevalent. Talk to us about the main themes that you’re seeing on those fronts from the Five-Year Plan in particular, especially since we didn’t get to have you on this discussion last week. What are you seeing?
Cory: Absolutely. So, it’s great that we’ll also have colleagues from the tech practice in later, so I won’t steal too much of their thunder. But the headline is really just the degree of focus on technological self-reliance. There’s a huge focus. And then this cuts across both, I mean, obviously technological upgrading and more advanced capabilities in that space. But there’s a heavy, heavy overlap with industrial upgrading. So, just as a basic example, in order to achieve tech self-sufficiency on the chip side, you need a lot of industrial capabilities that China doesn’t currently have.
And also to advance manufacturing in certain ways, especially in some of what were called the future industries. You need a lot more technological capabilities to actually achieve those industrial goals. So, heavy, heavy overlap. And I’d say the biggest, just at a high level theme, takeaway, whatever you want to call it, there is a whole section in this FYP dedicated to self-reliance. And the language that I think is going to keep a lot of analysts up at night is the call for the state to take “extraordinary measures” to reduce reliance on foreign technologies. And what does that mean?
Kendra and others have noted, we saw this language back in October of last year in a draft, in a preview, and didn’t know what it meant and was like, okay, maybe the FYP will clarify. It kind of, sort of clarified, but didn’t really. So, still a lot of open questions there, but what we do know, the specific calls for, again “decisive breakthroughs,” which, again, just means getting better at the innovation side, really, in integrated circuits, industrial machine tools, high-end instruments, software, and then something I’m going to focus on in particular is advanced materials, as well as biomanufacturing.
So, if you’re thinking about what spaces to look at, if it’s in one of those areas, it’s probably a good focus for the next little while. When it comes to these extraordinary measures and what might China actually do that’s different, again, this is not to steal tech team’s thunder, but just a little bit of lay of the land of how the innovation side is supposed to work in this Five-Year Plan period. Previously, I mean, really since 2018, but especially since 2021, 2022, 2023, we’ve seen a centralization of financing and of government coordination of science and technology R&D.
That is partly to help overcome some of the issues of fragmentation, like four different leading small groups running things back in the day. And now there’s the Central Committee on Science and Technology. So, it’s a better coordination system. But now the question is, okay, how do you actually make that more streamlined financing useful? How do you actually drive basic science innovations, bring them through the ecosystem from moving from an idea to a technology, moving from a technology to something that’s industrially viable? And how do you move that to something commercially relevant?
So, how do you actually get through that chain? And a lot of talk, a lot of policy thinking about that, that the tech team will talk about, obviously, chips are a major focus, but again, while they’re the top focus, I would argue, they’re not the only. There’s a much broader focus in terms of industrial upgrading. So, that’s the piece where I’d like to come in in more detail. But long and short, China has all of these upstream capabilities. We’ve seen it in terms of, you know, the moniker of the world’s factory, all the way through to, oh, this is where China’s asymmetric leverage lies in terms of export controls pushing back on U.S. tech control policy, etc. It’s all been this kind of upstream story. And by upstream, for listeners less familiar with the vocabulary, all I mean is things you pull out of the ground, purify, and then make ready to be useful for industry.
Things like antimony, your purified tungsten precursors, APT, whatever. And then midstream is more the materials that actually get used, your substrates, your gallium arsenide, your gallium nitrate, things like that. And then downstream talking about the things that people actually buy, your batteries, your cars, whatever. So, a lot of the focus has been on the upstream, both in terms of what China has invested in for a long time, and depending on your verbiage, dominates in many sectors, in many industries, I should say. And China wants to do two things.
It wants to, and it’s going to sound ironic to people who are used to thinking of China’s degree of control over industries so that the rest of the world is vulnerable, China also has its own upstream vulnerabilities. And so, one of the focuses of the last five-year period was really to shore up those upstream vulnerabilities. So, for example, getting lithium from sources that are not Australia and Chile, U.S. Treaty Ally and Strategic Partner, things like that.
Now, a lot of that same focus is being carried down to the midstream and downstream. So, the things you need to make semiconductors, right? So, that’s more of a focus now, increasingly downstream. But the other side of the same coin is, again, China’s had all this upstream industrial capability, but not a lot of value is created up there. The value is created in the midstream and downstream, like turning your rare earths into magnets. That’s something China’s done very effectively. But then turning that into motors, right? That’s where the value creation really comes in. The upstream sectors are really small in terms of financial revenue and total asset value.
And especially when it comes to increasing margins, like China needs more than just batteries and solar panels. It needs to start commercializing other downstream industries and not just dominating the supply chain, but saying, “Hey, if we have A, B, C down through W, we need to start making X, Y, and Z where all the value is created at the end of that chain.” So that’s kind of what we’re seeing is why the focus on technological upgrading, it’s both to shore up vulnerabilities, but it’s also to capitalize on existing advantages but move forward to where more value can be created. So, that’s the high-level story.
Andrew: Yeah, that’s super helpful. Why don’t you talk to us a little bit more about those upstream, midstream, and downstream aspects of what you’re seeing happening, break it down along that industrial chain for us.
Cory: Yeah, so upstream, this is where China’s really focused on hardening. Obviously, it works very differently than the rest of the world is trying to harden against China’s leverage. But specifically, China’s focus is really about domestic governance in many ways. There has been a focus on diversifying its source of imports, trying to get domestically some of the things it imports even at higher cost, what you might call a security premium, paying more for something just because it’s domestic. So that’s been in focus. But now there’s an intensified focus in the NDRC report and also the 5-Year Plan itself, the outline, the key phrase being ‘full chain management,” quán liàn tiáo guǎn lǐ.
What does that actually mean? And I think there’s a few things that it means, but the first is where previously there’s been a huge focus on China better governing its rare earth extraction and processing sectors, the next question is, okay, how does that relate to the downstream? Where is all of the different rare products, where are they all going? So, building out the map of that, figuring out what companies what industries are using it where. If you’re really thinking about export controls as your main focus, one of the natural questions would be, will this help China be able to impose downstream export control or midstream and downstream export controls? Will it help them understand the global ecosystem to better pursue extraterritorial jurisdiction, if and when it pursues that?
Again, absolutely. It’s also, though, about understanding where the value is being created and shoring up its own cross-supply chain needs. So that includes, for example, we’ve seen a lot of efforts over the last few years for geological survey, understanding where resource is. Now there’s going to be an increased focus as well on treating production site reserves as part of strategic assets, not just, oh, there’s a mine with some untapped resource, but actually calculating that as part of your strategic assets pool. Production capacity reserves, product reserves themselves, mapping that out more fully than China has in past. All that’s a security thing for China to see... And I should back up a step.
The key issue is no one knows exactly how fast the downstream demand will grow. I mean, China is banking on all the industries that use these things growing. But no one knows how fast. The last thing China wants is a shortage that bottlenecks that downstream growth. They’re going to make sure it has plenty of material, right? So that’s part of what all this is, is just mapping the whole upstream. Midstream is quite different. There’s a lot going on in midstream, but I’m going to boil it down to the most interesting piece in my view, which is a focus on what’s called new materials.
There’s a lot of ways you could read that, but what China really means is the types of, we call them advanced materials, specifically materials that are not your go-to commodities. Things like high-quality super alloys, ultra-high purity metals, high-performance fibers and composites. This includes bio-based materials. When you talk about bio-manufacturing as an industry, this is kind of more of a future-facing industry. So, what’s bio-based materials? We talk about recovery of materials from unusual sources, which is part of security, but also a value-added, a question of value-add.
All of these things, the difference between commodities and these things is that these things are incredibly hard to make. The IP is very strictly controlled. Very few people can actually do it. And so, you have scarcity by nature of how difficult it is to make these things. So, that means, one, you can sell them at high value. Two, you can make things that are also themselves high value and you’re not dependent on someone else. So, for example, if you see a military, you’ll frequently talk about advanced polymers, and advanced aerospace-grade alloys and supermaterials, the reason for that is obviously strategic, but it’s very hard to make them.
You don’t see them talking about steel quite as much. Everyone can make steel, right? China wants to basically be much more self-reliant in these advanced materials, but also start dominating the innovation side. It’s not enough to just make the things that everyone’s making, but more cheaply. They want to make things that no one else can make for a lot of reasons, right? To do that, you also need a lot of control upstream or a lot of security, I should say, upstream, right? So that ties in directly.
Andrew: Can you define, maybe we should have done this earlier a little bit more clearly, what you mean by midstream just for listeners? You’re talking about, let me see if I can do it, which is upstream is basically raw materials, with midstream being some level of processing. So, these are intermediate industrial inputs, but not finished goods. Is that sort of the way to think about what you’re describing here?
Cory: Yeah. And I will note, some people will use, there’s some gray boundaries. Like I consider purified minerals to be upstream. If you’re in the mineral industry, that’s going to be slightly different, right? But generally speaking, exactly, your minerals, your metals, the first things you make where, if you’re going to make a semiconductor, for example, you’re probably not extracting gallium from your bauxite processing chain. Someone else is doing that. So that’s upstream of you. And then you might make gallium arsenide. You make some chip substrates. That’s going to be midstream for the chip industry.
And then you have your thin film deposition, you start inlaying circuits, right? And then you make an actual semiconductor, you make an actual chip. That’s going to be a good you can actually sell. So that’s further downstream. So, when I’m talking upstream versus downstream, just talking about direction of travel. And there are certainly some great boundaries there, depending on which industry folks you’re talking with. But yeah, that’s exactly right. And there are different stages of the value chain. That’s critical.
In a lot of cases, why does the West not do much mining? Because mining tends to be very low margin. And so, if you’re going to make money, it has to be very, very high volume, just how iron ore works, for example, or copper. But it’s not how rare earths work. There’s a surprisingly low volume actually being used at any given time. And so, it’s low margin and low volume, which is why it’s not particularly attractive. Magnets, more attractive, motors, much more attractive, and so on and so on, right? This is true of a lot of industries.
Andrew: Yeah, we were talking about this, you and I, with some other folks on a call yesterday, and I didn’t get a chance to say it, but I’ll say it now. I mean, low margin, low volume is the absolute worst business you ever want to be in, right? If someone is trying to pitch you to invest in a business and says, “Hey, our business is low margin, but it’s also low volume” – run. Do not invest in that business.
Cory: That’s why it depends on government support. I mean, who’s going to pay for it? Basically, someone who has, again, this isn’t necessarily standardized language, but what I would call a security premium, someone willing to pay for it just to not get it from a particular source, right? Someone’s got to pay that.
Andrew: Yeah, totally. Well, keep going. I cut you off earlier. Were you about to go to downstream or did you have another point you’re going to make on the midstream piece?
Cory: Yeah, downstream is split. And I think this comes more fully back into kind of Dinny’s territory where there’s really two big downstream stories. And I should also clarify. For any one of these, there’s like 15 things going on. So, if you’re like, “Hey, you didn’t talk about X,” email us. There’s a lot more going on we can talk about. But just kind of picking a few high-level pieces or the primary pieces here. So downstream, I’d pick out two particular pieces. There’s managing overcapacity. Again, old story. Everyone’s heard about overcapacity to death. Still an issue. Surprise. Frankly, in one reading, there is an intensification of language around Beijing’s intent to deal with overcapacity.
On the other hand, frankly, I just don’t see it being that meaningful at this stage. I will wait to see the sectoral plans and other kind of specific industry siloed plans, but I’m not convinced at this stage that much is going to change on overcapacity issues until corporates just do more M&A activity and consolidate themselves. The other side of the equation, apart from overcapacity issues is industrial upgrading. Industrial upgrading space is, obviously, there’s a lot going on there, But I want to pull out a couple of interesting pieces. I should actually back up on one piece. On the industrial upgrading, there’s a set of industries known as the emerging industries, others called future industries, right? And so, this is the idea of batteries. It’s a thriving industry now, but it’s very young. It hasn’t fully matured. There’s still a lot of innovation.
We haven’t quite reached a steady state. So, it’s still emerging. And that’s a critical source of value creation and growth for China, right? So, those are strategic industries that will continue to get support. There are future industries that are kind of in principle, we’re on the cusp of them, that they’re not really big industries by themselves yet. Low altitude economy is growing. Biotechnology is very nascent. That’s going to be kind of a pillar for the future. So, we have all this stuff. Specifically connecting that to the capacity story, there is a specific line in the NDRC report, which I think is very, very telling.
It specifically says the government will allow appropriate surplus capacity while also encouraging competition and innovation in emerging industries. What they’re saying, to be very clear...
Andrew: This is good. I’m glad you’re bringing this back up because Ether mentioned this on our last podcast. And he was like, this is a little bit of a strange wording. And they talk all the time about shutting down capacity. And they mentioned that as well, shutting down overcapacity. But also they said appropriate excess capacity. So, I’m glad you brought this back up.
Cory: Exactly. Like, taking down overcapacity in low-grade things, like some of the lower-grade steel. And for those who don’t know, some steel is more valuable than others. A lot of the stuff coming out is just not that valuable. And so, there’s an effort to kind of pull that down and drill that down. And part of the way China will do that is through ostensible climate goals, carbon. So, carbon is being used as an industrial policy tool. I’ll get to that in a minute, but not necessarily at the scale that I think is going to really matter in the next year or two, maybe later. But then this other side, exactly.
It’s like, well, if this is an emerging industry, one, they don’t want to bottleneck it. Two, I don’t want to put words in Beijing’s mouth because this part’s not explicit, but I would venture to guess that, you know, looking at some of the industries with the worst overcapacity, such as batteries and solar panels and elsewhere, and electrolyzers too, what we’ve seen is that, yeah, companies take a huge hit. But at the same time, they have put so much money into competing with each other after bringing the cost curve down. How do you compete after you’ve run out of space to compete on price? You start making new technologies. You start making new versions of a battery, a slightly different one.
Half of the stuff won’t work, but it’s a very, very rapid prototype kind of system. It’s focused on a rapid prototype and getting new things to market. Is that sustainable? In no way shape or form is that unsustainable, right? But at the same time there are some short-term benefits as much as it’s painful for most of us observing the outside and anyone investing in the space. So, I wonder to the extent to which it is just a calibrated gamble that’s apart from not bottlenecking the space, let them fight it out and the best technology wins. Now that is very risky, I want to be very clear, because what you’re doing is cannibalizing returns that could be invested in more long-term assets or R&D for short-term turnaround.
Andrew: Well, it’s risky if you’re an individual business or an individual innovator, right? It’s less risky if you’re a government because that’s the whole point is the government’s assuming the risk of multiple technological pathways to future innovation and saying, “Go for all of it at once. We’ll see which one wins out.” I mean, we talk about the Chinese government becoming more like a venture capital investor where you back several different… you know, you build a portfolio, I don’t know, 20 portfolio companies, you expect nine of them or 19 of them to fail. But as long as one of them hits the moon, you’ve accomplished your goal, right? Is that a way to think about this?
Cory: Yes, but with a caveat, I think that was kind of the model, especially when local governments were picking their local champions, investing in each thing. And that was a mess from a debt perspective, local governments putting, you know, failing debts. That’s been somewhat remedied. Somewhat — there’s still ongoing issues. I think one of the issues here is that the question is how much of the investment is productive. And a lot of, especially when you’re looking at more future industries like advanced materials, robotics, you need many years of R&D to actually culminate in something useful. And you don’t have many years to wait while surviving in this kind of pace of turnover. So, there’s a trade-off there. If you have a relatively mature thing like a battery, and everyone’s like, “Let’s make a slightly different lithium-ion chemistry,” it’s worked terrifically.
If you’re like, let’s build something new in five years that no one’s ever seen before, a little harder, I think. That’s my guess, at least at this point. So, I’ll wrap up because I don’t want people to get bored with my voice, but the last kind of pieces here are those who know, I got on this from the climate perspective, and seeing the industrial transformation around the energy transition, climate transition. And it’s very interesting how China is and is not using carbon as an industrial policy tool. And I mentioned one, the strictest form of overcapacity control in kind of policy terms is really around the emissions of steel factories. And it’s basically a way to selectively cut out old capacity. That’s what it’s really doing.
But what we don’t see is any kind of absolute emissions cap, right? That is off the table. And critically, the headline climate/carbon target relates to carbon dioxide intensity reductions. For context, the 14th Five-Year plan had a goal of, over the five-year period, reducing China’s overall CO2 intensity by 18%. Partly because of the COVID recovery effort, China got nowhere close. If you’re being generous, maybe 12% reduction as opposed to 18%. We did not expect China to try to make up the difference in the 15th Five-Year Plan, But we also did not expect it to actually reduce the headline target. It’s now 17% reduction.
Despite not meeting the last one, you would think that it had more space to catch up. It’s basically, at a headline, what it’s saying is, this is a relatively weak headline carbon target. And I think that indicates they don’t want to add extra unnecessary costs. They don’t want a bottleneck industry. They don’t want to basically increase pressure that isn’t explicitly productive toward driving new value creation. There’s an element of compliance with CBAM and the European carbon market, right? But that’s not what this is really about. So, in that side, it’s quite weak.
On the other side, we have at a much more micro level that frankly is… I mean, we don’t have a strong take yet on how impactful this will ultimately be, but really worth noting, there’s a lot of focus in the 15 Five-Year Plan documents and in other policies that we’ve seen come out just ahead of it. On zero-carbon industrial parks, what’s really interesting about this? They’re already green and low-carbon industrial parks. And the idea there is make stuff you’re already making, but make it greener. The idea is hopefully kind of gradually commercialize lower carbon, lower-carbon-intensity industrial processes, start using that as offtake for green energy.
If you’re not aware, China has a surplus of green energy in most places. So how do you use it? Well, let’s send it to industry. And a good way to do that is have industrial parks that are literally built to offtake the green power and to have all these lower carbon technologies piloted and commercialized at these parks, which get a lot of policy support. The difference between that and a zero-carbon industrial park is this. You always generate some carbon, even if you mitigate everything you can.
So how do you get to zero carbon is you use offsets or you do carbon capture. Carbon capture is fixed asset investment of a very specific emerging technology. Emerging in terms of we know how it works, but it hasn’t really been large scale adoption. We haven’t seen large-scale adoption yet. And the other is to start participating in offsets. That is currently a voluntary market in China, carbon offsets, that is really weak and not doing much. The idea is that you want to transfer financial resources from high-emissions, typically older industries, to greener industries that can sell certificates for carbon offsets. That’s what zero carbon industrial parks are partly intended to help drive is, again, trying to help transfer those resources through carbon offsets in other markets, while also helping commercialize this carbon capture.
Now, that said, how much actual investment is going to go into that? Huge question mark, right? I still have to go through more details on the budget. But this is one of those things that could be really, really interesting if, say, several cities in Guangdong were just like, “You know what, let’s go full bore. Let’s fully support these parks and let’s make the best zero…” They can go into that. Companies have a lot of potential to profit from this. But will it be relevant at a higher macro level? I have no idea at this stage, but it’s one of the new pieces that I think is going to be very interesting. NDRC is like, let’s use green energy to manufacture green products and extract value from it. The big question is, can they create value from it? And that’s an open question at this point.
Andrew: Yeah. Well, thanks for covering that. The climate aspect of this, I think from our standpoint, often gets lost. As you kind of said, the climate goals are a little bit less ambitious than we thought they might be. But this is all sort of, I think it’s important to remind listeners, this is all in the Chinese policy mind, right? This is all part of the same coin, right? Different industrial policy, climate policy is technological and innovation policy. I just want to ask you, I probably should ask this at the beginning, but I want to step back a little bit and then bring Dinny back in. How do you think of all of this? So, you talked about upstream, midstream, downstream, we talked about climate aspects. How does all of this sort of industrial policy efforts, which is what I’d kind of categorize this as, fit into the wider economic upgrading strategy and economic security strategy.
And then, Dinny talked about this on past pods, you know, previewing the 15th Five-Year plan, want to get your thoughts on, if anything, has evolved in terms of how you see all of these efforts in terms of China’s plans to fully refashion its economic growth model. So, Cory, let me start with you there. Big question.
Cory: Yeah, no, I love it. The first thing I’d say, actually, maybe a little comparison is helpful here. I think the big takeaway, whether we’re talking about the upstream, midstream, downstream, whether you’re talking about climate and tech and kind of traditional industry, the major theme is that Beijing is very clearly trying to gain as many synergies as possible. Everywhere we look in terms of these policies, it’s trying to attack multiple goals with every given policy’s focus. Two big ones are resilience, so making sure that industry has what it needs and making sure that China isn’t facing the vulnerabilities that its own export controls pose on others, but from a tech perspective as well. And the other is to create value, to create new value from what it does have. Increasingly, there’s a point in the same direction. China needs the same technologies to shore up and to be more resilient and to create new growth, right? When we look at climate, it’s like, how do we get higher-value goods and materials?
It’s how do we move out the value chain there? It’s also, how do we use green energy? Huge storage, just to tie in Iran, for most of the world, fossil fuels are not energy security if you import them. Part of what China did, but China has coal as a backstop of energy security, but a central reason that it built so many green power, renewable energy sources, is not just because of climate goals. It’s because it was a form of resilience, right? It also led to...
Andrew: Yeah, I talked about this with Ilaria on last week’s pod.
Cory: Oh, perfect.
Andrew: About how a lot of developing economies are trying to invest more in purchasing EVs so that they don’t have a balance of payments in crisis when the price of oil skyrockets. So, anyway.
Cory: Exactly. And so you have three things there. You have domestic resilience. You’re not just dependent on oil and gas to the same extent that a fuel-driven economy is. Obviously, there’s still disruptions in China, but much less than others. And just to confirm for viewers, or the listeners, China has a lot of coal, a lot of renewables, not much domestic oil and gas, a bit. They refine a lot, but they’re not really independent there. And you have, it’s cheaper. Once they built it out, it’s now cheaper, which supports their industry. And now they’re exporting. Some of their biggest exports are these green energy goods.
And so, that I think is the prototypical case of climate, industrial, tech policy overlapping to meet multiple goals at once. That’s the ideal case. And what we see in a lot of these focuses on new materials and everything, it’s obviously a little bit more convoluted because it hasn’t happened yet. But a lot of what we see China pursuing seems to be attempting to do the same thing. It’s not one policy or another. It’s all of these grouped to multiple ends at a time. And I want to contrast that, for example. And this is not supposed to be a full-blown thesis of U.S. governance or policy or anything like that, but just one specific point.
When we look at the U.S. diversification effort for critical minerals, you see resilience. A thing happened, it caused disruption, let’s fix the thing that caused disruption. There’s no new value creation. There’s no tying that to other goals. It’s just fix a particular problem, that’s it. China’s goals are a lot more integrated, and they’re drawing a lot more synergies trying to basically maximize the return on every dollar spent on industrial policy.
Andrew: Yeah, there’s a lot more there to dig into. Some of that reminds me of the conversation you and I were having yesterday with that other group, but we can dive into that certainly on future pods. We’ve touched on a lot of different aspects of your portfolio there, Cory. I really appreciate all of that. Dinny, I want to bring you back in quickly to tie this into this bigger idea of China trying to transition to this new economic growth model, then we’ll wrap up this part of the discussion. What are your thoughts on how this all fits in?
Dinny: Yeah, well, as Cory said, it’s all about value creation. I mean, this is Xi Jinping’s model of new quality productive forces. This is, we are going to drive the economy through productivity gains. And what that’s all about, I mean, because productivity is an awful idea to try and explain, but what it’s about is it’s straight up about wealth creation, right? Which is one of the reasons they’re not doing consumption-led growth, which is just wealth redistribution. They see that future economic prosperity needs to be grounded in wealth creation. The idea is that this will allow them to continue to raise living standards, even in the face of a declining population and a rapidly aging population, they’ll be able to raise living standards such that China continues to converge with the living standards of the EU in the US.
And that, ultimately, value creation, productivity gains among firms generates high profitability, which generates higher-paying jobs, which those two things combined expand the tax base. And that allows Beijing to then be able to provide all the services that it will need to provide for a rapidly aging population, specifically pensions and healthcare. And so, you and I did a huge report on this for CSIS last year. Put it together by just pulling the strands, reading what China was saying, all the different places. The first time ever, I actually saw an official publication make that argument explicitly.
Earlier this week, the People’s Daily had a column talking about economic development in various places. And specifically, it was talking about one city where productivity gains was translating to higher wages and higher profits and was allowing the city to be able to spend more on aged care services. It just drew the link. A leads to B leads to C. And that’s what they’re doing. I mean, ultimately, the focus on value creation has a whole lot of different goals. As Cory was saying, we’re talking about resilience, we’re talking about security. But when it comes down to the new economic model, this is what it’s about.
Value creation is so important because it’s wealth creation, which is going to be increasingly difficult to do as the population shrinks and ages.
Andrew: Well, and I’ve got to say quickly here that as we describe Beijing’s goals and approach here, we’re neither endorsing it nor saying they’re going to be successful, at least 100% successful. I mean, my view would be they will at least be partially successful. But I just always want to put that out there that the goal here primarily is to describe how Beijing views this stuff. We’ll see if they’re successful over time. We’re not here to prejudge. We’ll be kind of evaluating that as we go. But also, I think for Western governments, the idea is, well, this is what the avenue Beijing’s clearly going to take by their own policy documents.
And hopefully, that kind of gives the U.S. government and others a chance to compete against those efforts. Guys, this has been great. Perfect place to end it because we’ve kind of pulled back up and talked very high level. I’m going to talk with the rest of the team in Asia here momentarily around other aspects of all of this. So, we’ve got more to dig into for listeners, but I think for this part, we’ll wrap up here because this has been really helpful. Cory, thanks a bunch for the time and all the thoughts today.
Cory: Cheers. Thank you both.
Andrew: And Dinny, thank you as well, man.
Dinny: Pleasure as always.
Andrew: All right. Listeners, stick around for my conversation with the rest of the Trivium team on all this stuff. Lots of 15th Five-Year Plan analysis to go through, and I hope you enjoy the rest of the conversation coming up.
All right, I’m joined now by, once again, by two of our analysts from the other side of the world over in Asia to talk like we previewed last week a little bit more in depth about some of the key themes sectorally within the 15th Five-Year plan and to some extent the Government Work Report over all the documents from the Two Sessions just to kind of give a little bit more color now that we’ve had a bit to ruminate on what all this means for various targets, various policies going forward. So, I’m once again joined by Even Pay, who once again is a Director at Trivium, a jack of all trades, covering ag, biotech, trade, and a few other things. Even, welcome back to the pod. How are you doing?
Even Pay: Yeah, it’s good to be here. It’s been a long week since all of these documents dropped, but my eyesight is coming back again. So, really excited to read some more policy documents today.
Andrew: Awesome. Well, I appreciate you getting on early for this. Trouble with running a small company that’s also a global company is dealing with time zones. So, it’s good to see you in your morning, my evening. And also from Shanghai, Bao Linghao. Linghao is one of our, really our lead analysts on the semiconductor space, who’s going to get a little bit more into the chips and compute side of things. How are you doing, Linghao?
Bao Linghao: I’m in the same boat with Even, final gas on. So, we can do this pod.
Andrew: Yeah, yeah, I know, I know. I appreciate it. The listeners definitely will appreciate it, I’m sure. I definitely appreciate it. And it’s also a good chance to step back from the screen. I guess we’re still on screens, but at least you’re not typing. All right, I’m going to start with Linghao today. Linghao, last time you talked us through some of the just quick hits on what you’re thinking in terms of how the Two Sessions address the issue of compute, but you wanted to dive more deeply into both the chips piece and the compute piece. So, start us off. And now that you’ve had a few more days to go through these documents, what are the big things that you think policymakers are thinking about in your space?
Linghao: First of all, there is a sense of urgency on chip progress coming out of this document because you see a language like calling for extraordinary measures to achieve breakthroughs in sectors like semiconductors, right? And they don’t use these languages lightly. They typically say that when they feel like they’re hitting a wall or they feel like they need to accelerate progress. And the reason for that is pretty obvious because a huge part of the AI race is a compute infrastructure race. The key pieces to AI development is basically compute, talent, and data, and energy infrastructure. I think China got all the pieces already except for the compute infrastructure.
And there’s a joke that the U.S.-China AI race is really a competition between Chinese engineers in Silicon Valley versus Chinese engineers in China, right? So yeah, China got all these smart people, and they can definitely do that. But they need chips and compute infrastructure to do R&D, right? It’s basically your R&D budget. But the thing is, I don’t think there’s a lot that Beijing can do at this point to help chip founders like SMIC to make more GPUs, right? I feel like at this point, any incremental policy support is like pushing on a string. It’s just a massive engineering challenge among the ecosystem chip companies.
So, it takes time. But there are other things that the government can do to help. I mean, The logic goal is if we can’t increase the volume of the leading edge chips, we should at least make the best use of existing chips we have. So, the name of the game has become about increasing computer utilization. And one way to do that is to push ender process towards cloud platforms. I think this policy push actually deserves more attention than people are paying attention to because cloud is the best way to achieve the highest utilization possible. The difference is like owning a car versus taking an Uber, right?
A primary-owned car sits idle most of the time, where an Uber car is always on the road, right? And when you have a limited amount of cars, you should tell everybody to take the Uber, not to buy the cars themselves. And I’ll give you this interesting piece of data. China’s cloud adoption rate was only 28% in 2022, where it’s 60% in the United States. And the one main reason for this low adoption rate is that government agencies and SOEs, which, by the way, is a huge part of the enterprise market, I think SOE revenue is like 30% of the total. But anyway, they’re a sizable part of the market. But SOEs prefer on-prem over cloud.
On-prem basically means you buy the servers and you use them for yourself. It’s the same with the private-owned car analogy. And they do it out of data security and privacy concern. To make the matter worse, in recent years, local governments have been rushing to build GPU data centers. But the problem is that they have no idea what they’re doing. These GPUs are super complicated to set up and very easy to break. And on top of that, you also need to run very complicated software and then manage it in a way so the customers can use it. So, anyway, not everybody can be a cloud provider. It’s a non-trivial task.
That’s why the biggest cloud platforms are the biggest companies in the world. So, obviously these local government-built data centers have huge poor utilization problems. And we saw, you know, last year NDRC tried to step in and try to stop that. Shanjie Zheng spoke out and they need to stop these poorly built data centers. Because if you think about it, like this kind of behavior, it actually exacerbates China’s compute crunch by taking those companies off the market, right? Which could have been run by, you know, AliCloud, ByteDance, or Tencent more efficiently.
I think the push towards public cloud is a recognition that there’s a huge amount of waste of computing resources in China, and they really need to fix it.
Andrew: Okay, this is super helpful. I’ve got a few questions for you. Some relate specifically to the Five-Year Plan, some that don’t. So the first one, I’m going to start with the devil barrel question. One is, I mean, weren’t just a few years ago policymakers saying the exact opposite about public cloud? Weren’t they trying to push users away from public cloud? So, tell us sort of, to the extent you remember it off the top of your head, the history and about face there, if that’s indeed correct. And then the second part would be, I know we’ve been having a debate internally about whether this push for use of public cloud will be good or bad in the end for Chinese hyperscalers. So, walk me through sort of the debate as you currently see it and kind of where you currently land on that debate. Does that make sense as a two-parter?
Linghao: Yeah. First of all, you’re right. A few years ago, when we were still in the whole tech crackdown campaign, we saw incidents where local governments or state-related agencies that were previously AliCloud and Tencent Cloud, tried to move themselves off that because there was this big tension between big tech and the government. I guess they were worried, if we’re about to launch a probe into these companies, would they actually know? Things like that.
So, there is some tension there. But I would say like that tension has largely, like the tension has ease a little bit. But the thing is like fundamentally these local governments and SOEs, they just don’t have the habit to being on cloud. To be clear, like they do a lot of cloud deals with the telecoms. But these are private clouds. It’s more like on-prem deployment. They’re not sharing the resources with other tenants. So the utilization is still very low.
Andrew: Right. So, it’s not that they need to move from private cloud use to public cloud use. It’s just that they’re not on the cloud at all.
Linghao: Right. So, to your second question, whether it’s good or not for the Chinese hyperscalers, the honest answer is I don’t know yet. I mean...
Andrew: Well, walk us through the two sides of the argument. Why might it be good? Why might it be bad?
Linghao: So because of the two ways to do it. One way to do it is actually let hyperscalers like AliCloud and ByteDance to do their job, right? To help them to succeed, to push the SOEs towards their cloud platforms. I think that’s going to be huge for increasing computerizations. And they can go even further than that. They can even let these hyperscalers to buy the residual assets from these poorly run data centers. They can get those chips back on and increase the utilization. If you want to go full force, that’s going to be huge for Ali and Tencent ByteDance, right? And the government can even, you know, help them get a cheap credit and to do the expansion.
But another way to do it is, you know, you could also let a telecom giant to do it, although they don’t have as much of expertise as the private sector hyperscalers, right? And I think if they do that, like it could be a mix though. And fundamentally, like the tension and the debate we’re having right now is that Beijing views compute as a public utility. And usually in these public utility sectors, SOEs dominate. So, we’re not totally sure if policymakers are willing to or comfortable with letting the public sector SOE hyperscalers to dominate the infrastructure of the future. If you’re thinking AI is going to be a huge part of the future economy and this infrastructure is basically supporting the future economy, then that is a gigantic and influential powerful sector.
And how the government is going to deal with the control, how they’re going to balance between growth and control is going to be a very interesting problem to watch going forward.
Andrew: Yeah, and I would say if we look at past examples of areas where the government has seen some sort of service or infrastructure become what is a utility in their mind, it doesn’t usually bode well for the participation of the private sector, generally, not even to speak of the profitability of the sector, right? Because the government tends to use the companies that are providing those public goods as kind of buffers, right? When things go badly, they can sort of increase costs through lower margins. Like this is why the government wants SOEs in those places, to sort of serve as a ballast in bad times in particular, and as a reliable partner, both in good and bad times. Is that kind of how you view it?
Linghao: Yeah, I think that’s a really good point. Also, there’s a parallel policy push to increase compute utilization, where it sort of makes this exact point where the state won’t control that layer. The policy push is actually called the National Unified Computing Power Network. That actually predates ChatGPT. And the key idea is that they want to build a platform on top of the cloud platforms. Basically, a master orchestration layer for all the available compute in China. I mean, if they can do that, I mean, the idea is similar, right?
It’s aimed at increasing compute utilization. Then when you can aggregate these supply together. But the problem is it essentially this intermediates the hyperscalers. Because if the master layer is customer saving, then what the hyperscalers are really doing, right? They’re basically setting up them centers that are not like customer-facing. So, that’s a major conflict of interest, right? And the hyperscalers aren’t really interested in being commoditized, although they publicly support it.
But if you look at one of the closed-door policy discussions that they had that got reported a couple weeks ago before the Two Session about the computing network, the think tanks were there, the telecoms were there, but the hyperscalers were not there. So, that’s very telling.
Andrew: Conspicuous. Yeah, yeah, for sure. And on that National Computing Network, correct me if I’m wrong, but my understanding is sort of, it’s almost like they’re trying to create a version of a smart power grid that would enable certain areas of the country effectively can use compute from other areas of the country that might have idle compute at certain times, just like you would store power on the grid from one area and try to ship that power to another area so that you kind of have various parts of the country absorbing or using power, or in this case, compute, when they need more and when other places need less. It’s not a very eloquent description, but is that generally kind of how you think about it?
Linghao: Yeah, I think that’s exactly right. I feel like they are treating compute as power, as one of those commodity market. But just as I mentioned before, building a cloud platform is non-trivial. It takes a lot of work. It takes a lot of expertise. That’s why we have the poor utilization problem in the first place because there are only a few players on how to build it. Maybe this market can be commoditized, but at this point, the companies building that should be rewarded for what they’re doing. It’s not a commodity market yet, and takes expertise to build that.
So I’m not sure if viewing, although they view it as a public utility, it does have some certain characteristic to that. But I’m not sure if that sort of fits with the reality of the market.
Andrew: So, I mean, last question, just based on your reading of the Five-Year Plan, I mean, do you think this is sort of one of the core elements of what they’re going to pursue in this space. Do you think they’ll be successful in, you know, alleviating their compute issues? Or at least maybe a better way to put it is making what compute resources they have more efficient or maximizing those?
Linghao: Yeah, I’m a little bit skeptical to how much success they can have in the short term. Because we just mentioned all these conflict of interest and the problems, like how they think about these issues. I mean, the best way to achieve the outcome is to let the hyperscalers to succeed, and help them to succeed. But I’m not sure that we’re fully there yet. So, yeah, we’ll see. Like, I think, you know, we’re going to see some signals maybe this year, in the next few months, to see how they’re going to go about doing it. Yeah, but I think, you know, they have recognized this is a huge problem.
And it is a huge problem because if you look at what has happened in the last few months, Chinese models has improved significantly, and also especially on the coding side. And coding is the first vertical that has proven to be a huge market. When we saw Zhipu launch their GLM-5 model, which is a really coding model, they just couldn’t service their customers. They launched the model and on the same day, they raised the prices. That’s unheard of in China. They raised the prices because they couldn’t like deal with the influx of the demand.
And we said like compute was an R&D budget, but you also need to compute to service customers. And when we are heading into this agentic AI era, which is going to consume several order of magnitude of compute, when we have these scaler apps in the future, China is going to have a problem to service them. That’s going to be a huge problem.
Andrew: Yeah, totally. Well, it’s obviously a core issue that they’re facing, as we said last week and you mentioned again today. It’s good that they’re zeroing in on this issue as what seems to be kind of one of the top priorities in getting the chips and compute space right. And so, we will be following that in the days ahead, in the weeks ahead, in the five years ahead, and beyond. So, Linghao, appreciate you kind of diving deeper into that for us. We’re going to bring in Even Pay to talk about a few different areas of her portfolio, as well as we continue to go through different aspects of the Two Sessions and Five-Year Plan. So, Even, why don’t you come in as we pivot?
I know we’re going to continue to talk about tech in some respects because tech and innovation was so central to the entire five-year plan, 15th Five-Year Plan, but we’d love your thoughts on kind of how you read the document generally and how the tech and innovation piece specifically kind of showed up in the areas that you follow most closely. So, what’s the top of mind for you, a week out from having had a chance to go through this document?
Even: Yeah, yeah. Good strong pivot. Last time I was on here, I chatted a bit in, I think, a really granular fashion about how exciting the bioeconomy contents were, which was a pretty kind of granular look at the plan. And a part of that was, you know, it’s 140 pages of Chinese text and it just takes us all a little while to get through. But over the last week, we’ve all spent a lot more time with the plan, and I have a much better sense at this point of where that fits into kind of the broader vision, right? And like you say, this is in no way a surprise, but innovation, you know, the innovation agenda, the new quality productive forces agenda is a major, major priority across every aspect of the plan.
It’s probably the most prominent priority in terms of where there’s ambition, where there’s adjustments to top-line targets, and where to the best of our ability to tell there’s likely to be funding commitments, policy movement over the next five years, right? And that shows up really clearly in the first place in those top line targets at the beginning of the plan, which are going to, in many respects, going to be the core areas of focus for local policymakers in terms of when they’re like, “Oh, what do we have to get done? What are we going to be graded on? You know, what is our report card going to say at the end of five years?” those top-line targets are really central, right?
So, there was an increase of over 80% in terms of the expectations for creation of high-value patents. In the 14th five-year plan, it was 12 high-value patents per 10,000 people. Now, it’s been raised up to 22 high-value targets per 10,000 people. So, really, this is an anticipatory target, but Beijing is looking for a lot more impactful innovation that’s translated into actual patent registration over the next five years. And policymakers also want to see R&D spending continue to grow at over 7% annually through 2030. So, there’s been a pretty healthy R&D spend growth for many years now, and that’s continuing to get juiced at a level that’s like nearly double GDP growth.
So, that’s a pretty firm commitment. But beyond those sort of the top line target items, when you dig into the plan, what you see is that basically every kind of problem or challenge or target or sector or domain is approached from an innovation frame, right? It’s all about, what are we going to do in this place to ensure we’re making breakthroughs, to ensure those breakthroughs are aimed at the industry’s actual challenges, and to make sure that if breakthroughs are being made, they’re actually going to get adopted, disseminated, and scaled into the real economy?
So, for example, in agriculture, the top line target for staple crop production was raised pretty significantly. The last five-year plan asked for 650 million metric tons in annual staple crop production by 2025. China had already exceeded that level in 2018, long before the plan came out. So, that was not an ambitious target. This time, the new plan targets 725 million metric tons. And China is still a little over 10 million tons short of that target. So, unlike the last time, it’s actually a reach target. And what the plan describes, all of the efforts for getting there, 100% of the efforts for getting there are squarely focused on adoption of ag tech to improve productivity.
That’s rolling out more farm drones and robots. It’s adopting using biotech and traditional breeding, whatever it takes to adopt new higher-yielding crop varieties. And it’s stuff like adopting AI models for forecasting and controlling pest and disease outbreaks or severe weather so that farms can prepare for anything that might impact yield. So, instead of expanding how much farmland is being used to grow crops, they’re just saying, “Look, we’ve got this amount of farmland or maybe even a little less by the end of the next five-year plan. You need to make it count.”
Andrew: Yeah. It does seem like, obviously, productivity and new protective forces at the heart of all this. And so, it’ll be interesting to see the ways in which they apply those to some of the more specific areas that are kind of beyond just industry, the big macroeconomic kind of macroeconomic drivers, right? Talk to me a little bit more about what you’re seeing in terms of… Dinny kind of flagged this, Dinny McMahon, our Head of Markets Research.
On the whole, it kind of combines a few things. One, the idea of consumer spending, but also healthcare and elderly care, and sort of welfare, social security spending, the whole idea of investing in people. It’s less on the innovation front, but I know you’re also following this closely. It’s just something that we haven’t touched on yet in all of our analysis of the five-year plans. I just wanted to get kind of your take on what’s going on in that sector, at a high level, but then also, like I said, for listeners, you do a lot of our work in the health space. So, what are some of the specifics that you see there as well?
Even: Yeah, absolutely. So, this is, again, another area where you’ve got some pretty significant challenges where there’s been this incredible drag on consumption that’s come from households’ willingness and ability to spend. And at the same time, you also have sort of mounting kind of demographic challenges where you have the birth rate falling pretty precipitously. You have that dragging down total population. You’ve got this growing population of older folks. And because almost everybody who is currently a working age adult came from a one child family or really, you know, the exceptions to the rule might have had a single sibling, you’ve got a situation where a household, two adult working age, parents who might be thinking about having kids, they also have to care for both of their parents.
There is not another sibling. So, each household has four grandparents and whatever number of kids they’re thinking of having. So, you can imagine the sort of financial sort of pressure as these households main asset, right? Their homes have lost value, right? And I think finally, there’s been a pretty clear recognition in Beijing that it’s dragging on the entire system. It’s contributing to the falling fertility rate. It’s contributing to sort of really lagging consumption because adult working age folks are just freaked out about the burden of care that is falling to them. Right?
So, at the top line level in the plan, we see a few relevant targets. There’s a brand-new target that’s calling for a six percentage point increase in child care enrollment rates. The 14th Five-Year Plan talked about opening up child care slots, new places that kids could fill into, but they didn’t actually talk about whether those kids showed up in childcare, which obviously makes it maybe less of a useful target. You’re not getting all the way to the goal that policymakers want. So, by 2030, Beijing wants there to be a lot more kids under three that are actually able to access childcare.
We’ve also been following closely a lot more subsidies, both at the local and central level, for childcare and early education, kindergarten enrollment. There’s another new target aimed to raising the proportion of nursing care beds that is relative to like a standard residential care bed in an elderly care facility. So, by 2030, Beijing wants 73% of those beds to be places in elderly care to be set up for people who need like really serious medical attention. And those are the folks where it is really important for them to be in a facility because if they’re having to live in a home, the adult children caring for them are probably going to be really struggling with what is a medical care provision on a day-to-day basis.
And then there’s also a target to raise, to expand the health care workforce. Previously, this target was applied only to practicing physicians. But in the new plan, they’ve expanded that target to include both practicing physicians and a separate target for registered nurses. And that’s really a recognition that the entire healthcare system is understaffed and making healthcare, especially primary care, work better is going to require more people. And at the same time, across at least two of these three areas, we also see some commitments to innovation in the space. And for better or for worse, there is a big call for increasing the uptake of AI in hospitals.
So a lot of that is, at least in the initial, I guess the initial push is going to be focused on trying to reduce administrative burdens. So, to free more staff up for more patient care while the AI deals with paperwork. We’ll see how that goes. We’ll see how much progress can be made over the next five years. But there’s also, in sort of the text of the plan, some other innovation relevant contents about how we’re improving care is a push to do more preventative care. And that includes efforts like rolling out more vaccines within the National Immunization Program, which makes those vaccines both free to patients and also highly, highly recommended prescribed by doctors.
So, what something like that does is, first of all, it creates a market for people who are developing vaccines. Of course, anybody in the medical innovation space will be interested and excited to see that. But more importantly, from a public health perspective, it reduces how sick people are getting. If you’re vaccinating against diseases, for example, most of China’s national immunization program has focused on kids, like most places in the world, you know, getting everybody’s basic vaccines done when they’re little.
But increasingly, there are a few things we know we should be giving adult booster shots for. Or, for example, the annual flu vaccine that many people get, older folks really should be getting in order to reduce how sick they get if there’s a bad flu year. And just including those as something that’s recommended can have a really huge impact on how hospitals are functioning, you know, through the winter. So, they’re looking to do that, and that’s really kind of a landmark development in this five-year plan.
Andrew: Well, we covered, man, a lot of ground. I mean, you didn’t even yet get to hear my conversation with Cory and Dinny, but obviously we touched on sort of tech aspects, but then within Cory’s portfolio, it was kind of the critical mineral stuff. Dinny, of course, talked about, a little deeper on the macro stuff and around infrastructure in particular. Linghao just covered the chips and compute piece. You’ve gone through some of the biotech, the healthcare. What else do we need to touch on here in the 15th Five-Year Plan that you think listeners need to know about that really jumped out in terms of some of the things that have not yet been on our list of discussion items?
Even: Yeah, it’s been hard getting through this past week in the context that elsewhere in the world, there’s this massive global crisis unfolding in Iran and the Strait of Hormuz. And I think one thing that we haven’t talked about is sort of Beijing’s stance on or positioning with regard to energy security and raw material security more broadly. And stepping away from the Five-Year Plan for a moment, one of the things that we’ve been watching really since 2018, when there were institutional reforms that brought a few various sort of dispersed responsibility around state strategic reserve management under one regulator.
And then again in 2023, when we saw sort of a post-pandemic, like a big uptick in spending on reserve management and a few more expectations for SOEs. We’ve been tracking a pretty big increase in the annual spend on strategic reserves, that over the past six months, we’ve seen a pretty substantial amount of sort of increased stockpiling specifically in the crude and petroleum product space in response. And when we look at this year’s Ministry of Finance proposed budget report, it flags yet another over 8% increase in spending on state reserves. And we’ve talked a great deal about innovation and how sort of forward-looking and ambitious and maybe exciting some of this direction is.
But I think it’s really important to remember that, ultimately, a lot of what Beijing is doing is preparedness, effectively, for gray rhino types of crises. You know, the innovation agenda is about breaking chokeholds and gearing up, not just to break chokeholds that exist today, but to anticipate and prevent any future chokeholds from emerging, right?
Andrew: And I don’t want you to lose your train of thought, but just for listeners who may not be familiar, the gray rhinos in the Chinese lexicon is there, that idea is kind of the opposite or analogous to black swans, which are sudden events that are catastrophic that you can’t see coming. The gray rhinos, as I understand it, are sort of obvious, slow-moving risks that you might not pay attention to because they’re so obvious and they’re slow-moving. But unlike black swans, have a very high potential to impact you negatively. And so, there’s things that you absolutely need to be in front of. So, for any listeners who aren’t aware, I just wanted to drop that in. Yeah.
Even: Exactly. Exactly. And I would contend that this is one of the strengths of China’s system, for better or for worse, is that with the current crop, at very least, of top leadership, there is a great deal, maybe even too much focus on preparing for managing future kind of gray rhino scenarios, whether that is another pandemic or another trade war or another choke point technology or a new conflict in the Middle East. We’ve seen years and years, we’ve seen how the system after the first Trump administration, spent years doing scenario planning for another trade war.
And at this time, as tensions mounted, that they had a really robust toolkit that has been very effectively used to get to the negotiating table much faster and, you know, much more kind of aggressively than the last round. And as sort of the conflict in Iran is starting to look more complicated and more intractable and like it may drag, initially, there’s been a lot of focus on, oh, China might be knocked off balance by this because they’re import dependent for oil and gas. And a lot of that is coming out of Iran and elsewhere in the Gulf. And then practically this incredible sort of long-term effort toward preparedness in the reserve system, you know, has kind of come into play.
And now it’s like, well, China’s going to experience kind of relative stability, actually. As long as this is a relatively short-term issue, those reserves and other aspects of how price management works is going to be able to tamp down on that risk. So, we haven’t dug too deep into how that is showing up in the 15th Five-Year Plan, but it’s certainly showing up in the annual documents, showing up in the annual budget line coming out of the Two Sessions that China isn’t going to start backing off reserves. It feels very vindicated, actually, in most ways, about building up those massive reserves and that is useful. It enables China to avoid global markets at times of crisis and shock. So, that’s not going anywhere.
Andrew: Yeah, that’s a good point. I was reading today about how Southeast Asian countries like Thailand are rationing fuel and maybe even going to a four-day work week because they’re absolutely going to be hammered by this. I feel like we could go on and on about the Iran situation and how it will affect China and Asia more broadly. But I will say, I think that’s a great place to kind of wrap it up because for two reasons. One is you hit on something that’s always like an evergreen theme in Chinese policy making, which is the idea of economic resiliency and economic security and kind of pushing forward on these long-term agendas year by year, five-year plan by five-year plan, but on an ongoing basis to ensure against these gray rhino risks that they see always kind of lurking.
But also, it’s an important point on China’s response to Iran, which we will get into much more in the coming weeks. We’ll next week have a segment on that. We’d love to have you join us, Even. We’ll definitely have Joe Mazur as well. So, that’s something we can and definitely will explore further on upcoming pods. So, I think great place to leave it. We’ve covered a ton of ground. Thank you, Even, so much for joining me early in your morning today.
Even: Awesome. Always a pleasure.
Andrew: Thanks, everybody, for listening. We’ll see you next time. Bye, everybody.
