Below is a complete transcript of the Sinica Podcast with Paul Triolo.
Kaiser: Welcome to The Sinica Podcast, a weekly discussion of current affairs in China, produced in partnership with SupChina. Subscribe to SupChina’s daily newly designed China Access newsletter to keep on top of all the latest news from China from hundreds of different news sources, or check out all the original writing on our website at supchina.com. We’ve got reported stories, essays, and editorials, great explainers and trackers, regular columns, and of course, a growing library of podcasts.
We cover everything from China’s fraught foreign relations to its ingenious entrepreneurs, from the ongoing repression of Uighurs and other Muslim peoples in China’s Xinjiang region, to Beijing’s ambitious plans to shift the Chinese economy onto a post-carbon footing. It’s a feast of business, political, and cultural news about a nation that is reshaping the world. We cover China with neither fear nor favor. I’m Kaiser Kuo coming to you from Chapel Hill, North Carolina.
Semiconductors have assumed a place of enormous importance in conversations about China in recent years. To be sure, it’s become quite cliché now to say that silicon is the new oil. Okay, sure. There’s other people who say water is the new oil or data is the new oil. But just as Americans are occasionally made acutely aware of just how much the economy can be shocked by oil supply shortfall as we were in 1973, or more recently, during the Colonial Pipeline hack just last year. Or, of course, after the Russian invasion of Ukraine in February, we have felt the impact of a shortage of semiconductors very acutely recently.
This resulted in big problems for sectors where, not so very long ago, we didn’t immediately think of semiconductors as being quite so vital, like the automotive sector. The fact that so much of the world supply of advanced semiconductors comes from Taiwan, and that tensions across the Taiwan Strait have ratcheted it up so much in recent years, has of course added to concerns. And not just here in the U.S., but certainly in China, too. China, after all, has been on the receiving end of increasing pressure from the U.S. and some of its allies, who have sought to prevent it from getting its hands on core technologies necessary to fabricate the most advanced semiconductors, let alone just being able to import those semiconductors directly.
At a rare joint FBI and MI5 press briefing, the chiefs of the two services, just last week, warned of stepped up Chinese espionage and influence operations. The FBI director, Christopher Wray, said that if China were really to take Taiwan, it would “represent one of the most horrific business disruptions the world has ever seen.” In recent weeks and months, Taiwan’s top trade negotiator, also U.S. Senator Mark Warner, the Democrat who heads the Senate Intelligence Committee, and many other people have also warned that the disruption of the world economy of war across the Taiwan Strait would dwarf the impact of even the Ukraine war.
Aware of the vulnerability of supply chains, eager to avoid political risk, and really hoping to benefit from government largesse, American semiconductor companies are now backing legislation aimed at reshoring or onshoring chip production, most notably the CHIPS Act, which we’ll talk about. That would earmark $52 billion for American chip makers to move more production state side.
Meanwhile, more and more companies have already been added to the entity list published by the Bureau of Industry and Security, BIS, under the Department of Commerce. These include, of course, giants like Huawei, which was added some years ago now, and had been a major buyer of chips produced using American IP. Beijing, understandably, has come to believe that all this is aimed at throttling China’s rise as a technology power. And China’s leadership has opened up all the taps in an effort to catch up as quickly as possible with the U.S. and its allies.
Joining me again on Sinica to help us all better understand how semiconductors fit into broader U.S.-China relations is Paul Triolo. Paul is now Senior VP for China and Technology Policy Lead at Dentons Global Advisors ASG, formerly and probably better known still as Albright Stonebridge Group. He spent 25 years as a research analyst for the U.S. government at a certain three letter agency. And until recently, he served as Practice Head of Global Technology Policy at the Eurasia Group. Paul Triolo, welcome back to Sinica, man.
Paul: Hey, Kaiser. Great to be here and to talk about one of my favorite topics: semiconductors.
Kaiser: Yeah, yeah. And you talk about it well, so I’m really excited. Paul, why don’t we just start with a broader landscape here. What’s stoking our current anxiety over semiconductors? Is there any one issue that dominates, or is this more sort of a confluence of several issues?
Paul: Great question, Kaiser. I think the latest anxiety over semiconductors sort of depends on where you’re sitting, right? For the general consumer, there was no anxiety over semiconductors, or even until the chip shortage, really. So people just took it for granted that their next smartphone would have more power or more memory, be a lot faster, and do more cool things. Then all of a sudden, as a result of the shortage, people could not buy the things they wanted. So all of a sudden, semiconductors and TSMC microcontrollers were on everybody’s lips.
Several interesting things happened here. The general consumer and public became aware that semiconductors mattered. All of a sudden, everybody realized they were important, and that the semiconductor industry supply chains could be disrupted. They were a little bit fragile, could be disrupted by pandemics, by fires, by freezing weather.
And also, as you noted, that Taiwan and China were somehow tied up with all this. There’s a lot of confusion on this — the roles of China and Taiwan in the bigger semiconductor… Now, for industry and Congress, for example, anxiety over semiconductors began later in the Trump administration when people realized that 92% of the advanced semiconductors were manufactured in Taiwan by one company and that this was a problem in a geopolitically and geographically fragile region.
And that anxiety, of course, spread quickly to Europe, and to a lesser extent, Japan, which was well aware of all this. Part of the anxiety here is the growing realization that semiconductors underpin all current, and importantly, future industries. Particularly things like artificial intelligence, quantum computing, wireless networks, robotics, and the metaverse. And so, this was something that again came to the fore, and also, of course, the idea that semiconductors are used in military systems also became on the radar.
For U.S. trade and economic officials, subsidies around semiconductors have been a concern for a long time, going back to even 2016, and also, sort of China’s growth in this area and desire to acquire companies. So, in 2016, for example, former commerce secretary Penny Pritzker called China’s national IC Investment fund an attempt to appropriate the global supply chain for semiconductors. So, in this sort of trade and economic arena, China’s subsidies and sort of how they’re operating have been a concern.
For Chinese leaders and industrial planners and tech companies like Huawei and ZTE that you mentioned, anxiety over semiconductors arguably began on March 6th, 2018, when the commerce department reactivated a suspended denial order on ZTE, blocking it from importing U.S. components, mostly semiconductors, from over 200 U.S. suppliers. And basically, the company was on its knees because it couldn’t really function, couldn’t support existing or future systems.
And then of course, that anxiety grew in Beijing as we had a May 2019 entity list action against Huawei, which we’ll discuss. And in August, 2020, an expansion of the Huawei action to include the companies that were providing semiconductors using U.S. technology, sort of an extra territorial expansion. That really set off alarm bells in Beijing. And as you noted then, China now has sort of pulled out all the stops to figure out how to reduce dependence on the U.S..
And then finally, if you’re a Russian leader or business person, your anxiety got really hot in February after the invasion of the Ukraine and when allied export controls really ramped up pressure on your country and its industrial base because… Largely cutting off semiconductors and other products that use semiconductors. So, I think by now it’s a secret to nobody in all these areas I mentioned that, yeah, semiconductors matter and their supply matters, and where they’re made matters.
But yeah, three years ago or two years ago, you could argue people didn’t know what TSMC was, for example, and now everybody’s writing on TSMC and how important it is. That’s probably the poster boy for the whole thing.
Kaiser: So, the anxiety is quite global and the semiconductor supply chain is really globalized as we are often reminded. Paul, this may be a tall order, but you did such a good job sort of laying out where the anxiety is from all the different actors around the world, but give the listeners a sense of the complex geography of the value chain. Where are chips designed? Where are they actually fabricated? What companies dominate in fabrication or in the technologies that are necessary for fabrication of the most advanced chips? Where are they packaged and assembled? And where are the major centers of their final integration into finished goods?
Paul: Wow. That’s a tall order, Kaiser, but I’ll see what I can do here.
Paul: Well, no. The other thing that I think people realized, those people I mentioned in all the different groups, is the complexity of the global value chain or supply chain, whatever you want to call it. I think I saw one estimate, there’s something like 120 countries have some parts of the supply chain on the upstream side.
Paul: People were sort of shocked to learn, well hey, Ukraine produces neon and palladium and all these critical gasses and other things that go into semiconductor manufacturing, so there’s lots of countries that have had pieces of this. And then on the downstream side, of course, I think the estimate I saw was 170-some countries were estimated to have been affected adversely by the global semiconductor shortage. Obviously yeah, it’s a complicated supply and then demand issue that we’re talking about here. On the supply chain side, despite a lot of countries and companies being part of the mix, the industry is really concentrated in a much smaller set of countries.
And those are really the U.S., Taiwan, South Korea, Japan, Europe, more broadly, but really Germany one could argue, and some other smaller countries like France and Belgium have key players which we can talk about, and China of course. And China is sort of the new kid on the block here. Probably if you’d asked me this question 15 years ago, I probably wouldn’t have included China there. All these countries have a major concentration of key supply chain links. Because of this concentration, which can get close to, in some cases, a single key supplier or a few suppliers or producers in one country or area, people have come up with the idea of choke points in the industry.
And this became a major issue during the pandemic as shortages backed up. But it’s not new. I mean, there have previously been some cases where pieces of the supply chain experience natural disasters and impact things. But I think in previous choke point cases, the industry was able to recover pretty quickly. I think, for example, in the Russian invasion of Ukraine in 2014, there was a big shortage of neon and the industry adjusted to that. That was really a small, again, a small but important part of the supply chain.
If we look at these six areas, let’s just briefly go through them. So, U.S., okay. The U.S. has sort of the origin of the semiconductor industry, and it certainly leads in the area of semiconductor design. Think Nvidia, think Apple, think AMD, think Qualcomm, basically all the chips in your iPhone or your Android device, all of those really are designed primarily in the U.S. And then there are also, importantly, the design tools, the so-called electronic design automation tools are all really dominated by U.S. companies. In this case, Cadence, Synopsys, and Mentor, which is a Siemens company, but most of the IP is U.S..
Then, of course, the U.S. also has one of the three top manufacturing companies, Intel, that has off-and-on been a global leader. Intel is what’s called an integrated device manufacturer, so most of what they manufacture, or at least at one point, all of what they manufacture goes into their own devices or is sold as a commodity and is not sort of specially designed. Then, of course, there’s lots of other really good companies like Micron and Western Digital in the memory space. So of course, the semiconductor industry has many, many subsectors and specialty areas. And then Texas Instruments, for example, which has long done analog semiconductors and analog devices are very good companies in this space. So, the U.S. has a lot of players in this.
Paul: And then finally, in addition to those EDA tools, the U.S. companies really dominate in the semiconductor manufacturing equipment area. So, companies like Applied Materials, Lam Research, and KLA-Tencor come to mind there. Now Taiwan — very different position in the supply chain. It’s home to so-called leading foundries or pure-play foundries. These are companies like TSMC that focus only on manufacturing semiconductors. They don’t design them. But they have a really critical role because they work with the design companies, and they work with customers to figure out what they want.
The other critical piece of Taiwan’s position in the supply chain is assembly, packaging, and testing. Something like 54%, or well over 50%, of that capacity is in Taiwan. Although it’s a much smaller part of the value chain. Now, South Korea, of course, has the second most capable foundry with Samsung, but in that game, a much smaller player than TSMC.
And it also, in the memory space is really dominant with companies like Samsung and SK hynix, really leading in the DRAM space and also playing in the memory sector. And a lot of those factories, of course, are right near the military zone within artillery range of North Korea. Then you had Japan, which was once dominant in some areas of the industry, but it’s really strong in things like specialty chemicals and some of the raw materials needed to manufacture semiconductors like photoresist.
And they also have some individual companies that are also key players in niche sub sectors such as lithography. And of course, Japan is now making a big push to boost some of its manufacturing capability by attracting TSMC to come to Japan. But Japan is not a significant manufacturer of semiconductors at this point.
Europe now is really interesting because Europe boasts these niche players like ASML in the Netherlands, which is the sole supplier of advanced equipment in the industry that is needed to make semiconductors at the most advanced levels. This is called lithography equipment. We’ll talk about that some more. They have a monopoly on the most advanced stuff and also dominate in the sort of the second-tier level of equipment. And then Europe, of course, critically for Europe, they also have IP. There’s an IP part to this. ARM designs a lot of the cores that are used in the designs to make billions of semiconductors. Every device you look around you probably has an ARM IP in there.
And that’s a UK-based company owned by SoftBank right now. So, it has Japanese owners, but it’s a UK company. Then finally, in Europe, they have very, very good R&D centers like IMEC in Leuven, Belgium. The Fraunhofer Institutes in Germany and CEA-LETI, in Grenoble, which are really critical to help push the R&D and develop new manufacturing equipment in the industry. So, Europe has a lot of advantages. And then finally, China. Of course, China being a new kid on the block, China is a big player in that outsourced semiconductor assembly and test market (OSAT).
So, Chinese companies, it’s kind of a low-margin business, and Chinese companies got into that early on. China does also, as you mentioned earlier, does a lot of the integration of semiconductors into final products using a lot of contract electronic manufacturers. Also, Taiwan is also very prominent in that area, with companies like Foxconn, Pegatron, a whole bunch of companies that are Taiwan companies, but they operate in China because they can find the talent and the costs are lower.
And then China, of course, has been trying to, as we’ll talk, has been trying to move into a lot of other areas like manufacturing, like foundry. So, they have a big company called SMIC, which we’ll talk about, which is a growing and important player in the foundry business. And then in the memory sector, they have a company YMTC, Yangtze Memory Technologies, which now has around 5% of the 3D NAND market and is an upcoming player, and they’re trying to break into the DRAM space.
China, across the board, you could argue, has companies that are players, but it’s very uneven. China is the source again of a lot of that anxiety we spoke about because their companies are starting to become major actors in many sub sectors of the industry. And the question is sort of, on what basis do they do this competition? Are they being subsidized by the Chinese government or not? Does that affect things? And that’s also part of the anxiety, because there’s a sense that China has benefited from access to U.S. technology, but its companies have not always played by the rules because the government is chucking money at them at an increasing rate.
Finally, on the advanced side, really here, we’re talking about Taiwan and South Korea and the U.S. — so that’s TSMC, Samsung, and Intel. Those are the companies that are competing at the cutting edge. A lot of other companies decided, like GlobalFoundries for example, decided that the R&D costs and the CapEx costs of competing at that level were just too much. Because you have to have a very large customer base over which to amortize that R&D to really compete here. And so, at this point, we’re down to TSMC at the very, very cutting edge. Samsung is also competing there. And Intel is trying to catch up. That’s why the concern again, about the concentration in Asia because of the presence of TSMC and Samsung.
Kaiser: Yeah. Paul, in case some of our listeners just aren’t clear about this, what makes these most advanced semiconductors, this cutting-edge technology so advanced? I mean, why is getting microprocessors smaller and smaller so important?
Paul: Great question. It’s really about sort of density of packaging, and also power consumption, and also of course performance, and allowing your iPhone to run all sorts of cool applications like facial recognition and other things. So what’s driving the industry to more dense packaging and more dense semiconductors is really this insatiable need of people like you and me for more functionality on our iPhones, and also more functionality and capability in some cloud-based applications, for example.
Also, the other thing that’s driving it is the need for specialty semiconductors, so-called ASICs, application-specific integrated circuits, to do really specialized tasks like run AI algorithms in the cloud, or like on your autonomous vehicle like on your Tesla, there are some very specialized semiconductors made by TSMC that do all the video processing, for example, of all the data that a self-driving or autonomous vehicle needs.
So, the industry has sort of migrated towards more and more complicated and dense packaging and higher tech and lower nodes. We’ll talk about seven, three, two nanometers. This is largely driven by the need for more performance and the need to really process bigger data loads, one could argue. The big thing is that the data workloads, particularly for AI for example, have gotten so huge that you just need more and more and more capable silicon. To some degree, of course, the semiconductor advancements have enabled things like AI to happen because before you couldn’t really run these big data sets to train AI algorithms.
And now you can do that in a minute as opposed to many, many, many months. So, there’s a sort of a symbiotic relationship between some of the key applications and the need to keep driving that performance, and also, of course, things like power consumption for mobile devices to new levels.
Kaiser: Yeah. When we’re talking about seven, three, two-nanometer scale, we’re talking about really one company here, and you mentioned them, this Dutch company ASML. And the U.S., for quite some time, has prevented them from sharing the most state-of-the art extreme ultraviolet or EUV lithography machines, which are amazing, truly, truly amazing devices. Prevented them from sharing that with China and is now pushing them actually to refuse to even sell its earlier generation DUV, or deep ultraviolet lithography equipment, to Beijing.
Reuters reported last week that the U.S. Commerce Department, BIS I suppose, is planning to further restrict access to chips of 14 nanometers or smaller. What do you make of this policy, I mean, if indeed the reports proved to be correct? I mean, this is relatively old tech, as I understand it, so why does the U.S. care so much about this? What is the national security gain of cutting off China from this older technology?
Paul: That is a great question. And so, I think it’s an issue around, sort of a justification for denying China this most advanced technology. That’s sort of where the crux of the matter is. So EUV, for example, is controlled by U.S. Export Controls, and then by the sort of broader multinational organization, the Wassenaar agreement. Those controls were added fairly recently. And so, that’s why, for example, the U.S. is able to control Chinese company access to that equipment.
And also, interestingly, other companies like SK hynix, which wanted to use ASML equipment, the EUV equipment, in its Wuxi factory to make memory was also denied the permission to do that because of concerns over that the technology could leak, for example, to Chinese companies. Now, the justification here is really, in my view, a little problematic. At one level, there’s a lot of concern within the U.S. Department of Defense, for example, about the potential for Chinese domestic companies to produce advanced semiconductors that can be used in high performance computers to model weapon systems.
And that’s sort of really the justification that’s being used for denying China this advanced equipment and the ability to produce domestic semiconductors at advanced level. Of course, some of those companies that are doing that can all still produce those in Taiwan. So, we’ll have to get to that issue because Taiwan, of course, factors in here. So, there’s that justification, although the Chinese, of course, view this very much as the U.S. trying to contain China, freeze China’s technological capability.
And if you talk to people in the administration and in the academic community, I mean, there is this sense that the broader justification is there’s a desire to keep U.S. dominance or leadership in the manufacturing arena. Of course, the leaders are TSMC and Samsung, not necessarily Intel now. But that China should be kept some number of generations behind, two to three generations behind, just as almost an economic security issue, and that there should be an attempt to maintain Western dominance in this arena. So, that also is out there as a justification.
Now the latest thing that you mentioned here, this recent attempt, and they’re related, the two things you mentioned are related. There’s an attempt at this point within the administration and working with allies in the Netherlands, of course, in Japan. Because to control this equipment, for example this deep ultraviolet lithography, which has been around for a long time and covers many many mature nodes all the way down to potentially seven nanometers, the reasoning behind that again seems to be to freeze China at 14 nanometers. And again, this is a complicated issue because the Commerce Department, for example, put SMIC on the entity list.
And they specifically said that there would be a denial of licenses for equipment that was uniquely designed to produce semiconductors below 10 nanometers. Now, the problem is none of this equipment that we’re talking about is uniquely designed for specific nodes. It can be used across a whole range of nodes. But the Commerce Department, probably with some egging on by the Defense Department, has identified this 14-nanometer node as a way to be a little clearer about where you’re drawing the line.
The problem is, where do you draw the line? EUV is one thing. There’s a whole bunch of other advanced technologies that are really part of the equation here. Once you get below 10 nanometers, it’s really not really about the node so much as it’s about a whole suite of technologies, including EUV, but including chips and other advanced packaging techniques that really will define where the industry is going here. So, there seems to be an attempt, “Hey, let’s draw a clearer line here. And 14 nanometers may be clearer than 10 nanometers, and let’s determine what’s going on here in China.”
Now, the problem with that is they’re gonna try to do this by looking almost at a facility by facility basis. So looking at companies in China that are producing mature nodes, you probably don’t want to cut off Chinese companies that are producing automobile semiconductors for example at more advanced nodes that are already in short supply, if you can help it.
If it goes forward, and there’s lots of hurdles here still, if this sort of trilateral arrangement goes forward, it would attempt to draw new rules around export controls around semiconductor manufacturing equipment, and do this on a probably case by case basis in China. So, if you are a memory company that wants to manufacture memory, maybe that’s deemed as a commodity, and that’s not going to be of concern.
Paul: But if you are a company like SMIC, which would like to go down to some of those lower nodes, you’re gonna be denied equipment to do that. But this is gonna be a very complicated process. And industry of course, and ASML included, is not really all that excited about this because China remains a huge market. For example, right now there is something like 15 fabs under construction that might want to buy that deep ultraviolet lithography equipment from ASML and more maybe to come.
So, it’s a huge market for ASML. Also, if they can’t get that ASML equipment, they won’t probably buy other U.S. equipment for other parts of the manufacturing process. For industry, many of these companies are generating lots of revenue in China, and they’re using that to plow back into their R&D. The argument is you’re sort of shooting us in the foot here. By denying Chinese companies access to this technology, you’re sort of undermining the whole virtuous cycle of how we do R&D and how we generate revenue.
There are no other markets in the world really, on the scale of China, that can sort of replace those 15 fabs under construction, for example, that would use DUV. There’s no easy way to replace that revenue in other markets, because building and equipping these big manufacturing facilities takes a huge amount of money, a lot of CapEx, lots of planning ahead. The argument is, again, with some of these regulatory or export control issues, you’re messing with an industry that, as we talked about earlier, is global. It has these complicated supply chains, and when you go in and mess with that, you’re going to have a lot of unintended collateral damage. That’s the argument the industry is making.
Kaiser: Just so we’re clear, our ability to go in and mess with that, the U.S. government’s ability to do that is predicated on the fact that there’s a lot of U.S. owned IP in even these DUV machines by ASML, right? That’s why they’re able to tell them that they can’t export. I mean, otherwise it seems like-
Paul: As part of the Wassenaar agreement, the Dutch government agreed to this idea that this specifically EUV technology is controlled. So, the Dutch government agreed to that. The DUV, though, is different. DUV equipment is not currently export controlled. This would be a big deal if essentially the Dutch government, the Japanese government, and the U.S. government agreed to control this technology. This is considered sort of mature technology; it’s been around a long time.
This is an effort again, led by the U.S. government to try to sort of re-control or newly control different pieces of the semiconductor manufacturing equipment supply chain. So it’s not clear, in this case, the Dutch government is gonna have to go along with this. Gina Raimondo mentioned the beginning of this trilateral dialogue after the meeting in May of the EU-U.S. Trade and Technology Council.
I’ve talked to many others in the industry, and there’s a lot of skepticism, for example that you rightly point to, about whether the Dutch government — again, the ASML is sort of the crown jewel of their technology sector — whether it’s a good thing for this to happen. I think there’s a lot of detailed discussions about how this would be implemented, who would be affected, what the scope of it would be that the Dutch government’s going to have to understand before it agrees to this. All this came out of the TTC effort to slowly move the export control system into the new era where a lot of these technologies we’re talking about are dual use, arguably, and keep in mind the export control system was set up for WMDs, weapons of mass destruction.
And now it’s being asked to sort of shoulder the burden, the existing system, for things like great power competition between U.S. and China, U.S. technology dominance, human rights, all these other issues. It wasn’t designed for that. And so, this new effort here is part of what will probably be a longer-term effort to come up with a sort of fifth control regime, if you will, that takes into account some of these other geopolitical realities.
Kaiser: It’s clear that we haven’t seen any let-up in technology controls, in export controls in entity listing, or what-have-you, between the Trump and the Biden administration. Instead, it seems plain that we’ve seen a ratcheting up of those restrictions, in fact, yeah? Is that a fair blanket statement?
Paul: It’s a great question, Kaiser. I think there’s been a lot of continuity in terms of sort of overall tech policy approaches. And then, of course, I think this recent effort is definitely an effort to ratchet up, but also, of course, the emphasis is on working with allies to do this and getting away from the sort of unilateral approach that the Trump administration took on all this stuff.
Paul: I think the difference is that the Biden administration at least prefers to have some more coherence to the policy, at least putting policies on a firmer legal foundation for example. We saw some executive orders in the Trump era, for example, like the ban on WeChat and TikTok, rescinded by the Biden team. And they’re going to take a closer look at how the U.S. might approach those complicated issues around TikTok and WeChat, which we should probably do another podcast on, around data and access to data.
The administration is trying to sort of regularize or put on a sound or legal footing some of these areas. The Biden team, for example, also moved management of Chinese military affiliated company’s list to the Treasury Department and put in place a more regular interagency process to choose companies for this list. This is the list that bans U.S. investors from investing in the securities of those companies. And SMIC, of course, was caught up in that process interestingly in the semiconductor space.
But there are several areas where I think there’s ratcheting up. The Biden team has reviewed some Trump era measures, such as adding Chinese companies to the entity list, and sort of piled on there. So it hasn’t really reviewed those entity list actions, but instead it’s… Or hasn’t offered companies ways to get off the entity list as happened in the Obama administration, where ZTE for example, was offered a path to get off the entity list after they were named to it.
So, they sort of continued to add companies to the list, including expanding that into things like biotech companies, as we saw last December added, and quantum computing. Again, the justification here is links to China’s military modernization or military-civil fusion, which is a big deal. Then also I think there’s that issue I mentioned previously, which is looking at the entire export control system. How do you sort of modernize the system which was set up for WMDs to account for these other concerns around things like semiconductors and, even Huawei, around telecommunication gear?
Kaiser: But as you say, trying to put them on a firmer legal footing and trying to do it more multilaterally rather than unilaterally.
Kaiser: That’s a great distinction. We’ve talked quite a bit about the outward dimension of it, what they’re trying to do to try to starve China of certain key inputs. But let’s talk about what they’re doing domestically, and specifically the CHIPS Act, which passed the Senate in May and is now before the House. So CHIPS, by the way, stands for Creating Helpful Incentives to Produce Semiconductors for America Act. I want a job as the guy who comes up with these reverse engineered acronyms.
Paul: Those are all great.
Kaiser: No, they’re fantastic.
Paul: Those acronyms.
Kaiser: First of all, this promises $52 billion to the semiconductor industry. Where does that number come from? I mean, was that pulled out of a hat? I mean, is there a story behind how that number made its way into the bill?
Paul: Well, that number was the result of a process that was done within the industry, the semiconductor industry, to try to find a minimal viable capacity model. It was the result of a realization that this heavy dependence on Taiwan for these advanced semiconductors and advanced nodes was not a good idea for the industry and not a good idea for, arguably, U.S. National Security. So, the attempt to come up with a way forward on that was based on calculations about the capacity and the money needed to close, to some degree, the manufacturing gap between the U.S. and the rest of the world.
And there was a lot of input on that issue from different parts of the industry, and they came up with this number of 52 billion, which includes… There are some stuff in there about research and other areas. And there’s other parts of the bill, the U.S. Innovation Competition Act, which has pieces on R&D to both bolster U.S. R&D, but it was basically a recognition that many other advanced, developed countries subsidized their semiconductor industry — Israel, Germany, Taiwan, Japan, South Korea — and that the U.S. needed to be more active in that game.
The idea though, I think, was to have this be the first down payment on a longer-term investment in onshoring advanced manufacturing to get the U.S. from, say, 10% of global manufacturing now to 20%. The EU also seeks to do the same thing, go from 10% to 20%. But this was an attempt to say, “how do we do that?” And to sort of work back from this, from a 2030 timeframe in terms of what would you need to get this started to attract companies like TSMC, Samsung, Intel, GlobalFoundries, and others to build facilities in the U.S.? And sort of begin to chip away–
Kaiser: As it were.
Paul: … at Taiwan’s dominance, but also other parts of the supply chain.
Kaiser: Yeah. I mean, it sounds like such an enormous sum of money, but given the cost of building fabs and of buying, I mean, just how many EUV machines could it buy from ASML? It’s just not a ton. I know Morris Chang of TSMC has said that costs in America generally are just way, way higher, 25% higher than in Taiwan. I mean, so you talk about it as a down payment, and that makes a little more sense.
Paul: Yeah. Keep in mind, Kaiser, that the subsidies in all these developed countries, there’s some part of the overall CapEx of a major facility like a major new cutting-edge foundry that companies sort of expect to have a little help on. It’s not the whole thing, right? It’s just a percentage, 5% to 10%, 15% of the overall CapEx. If they can get that subsidy one way or another — incentives, grants, benefits for land, all sorts of ways to do that — then that enables them to keep the cost down. And of course, the biggest cost in all this is the equipment, buying all the equipment as you know. One version of ASML’s EUV equipment costs $150 million, and it comes with a huge maintenance tail.
So it’s very, very expensive. I think the newest, the very newest numerical aperture ASML EUV equipment is like double that, right? It’s like 350 million. I mean, it’s really expensive stuff. The idea of the subsidies and the incentives is to just make this a little more palatable and drive down those CapEx costs to the point where the economics of those facilities make sense. Now, it’s still peanuts compared to what a single company like TSMC is gonna expend in CapEx in a single year. 40 billion, I think this year for TSMC, and 44 billion next year. So, if you look out over the time horizon, but that’s a false comparison because this is intended to sort of grease the wheels, to help those companies make a case with their boards and shareholders that they’re investing wisely and that the costs are controllable.
Kaiser: Yeah. And signals support from the government for this, right?
Paul: It happens to be around salaries of engineers and the cost of land, and the cost of water and power, and all those things, which Morris Chang of course at TSMC has said. I think he’s called it a full, hearty effort by the U.S. that doesn’t really make sense economically. Because when he runs the numbers, it doesn’t look good, but they have to do that. TSMC has to do that, for example, because they have had their arm twisted by the U.S. government to do that, to put a plant in the U.S., and to begin allowing the U.S. to have some domestic capacity.
Kaiser: Taiwan obviously is super important in all of this. I think for anyone who’s been paying attention at all and who has any kind of glancing familiarity with national security concerns understands why Taiwan should be really important in our thinking on this. But it seems to me that it’s just surged in national consciousness just very recently. Is that primarily due to the Ukraine invasion, or to the Russian invasion of Ukraine in February? Is that what sparked this renewed interest in Taiwan?
Paul: Yeah, I think that’s part of it. Absolutely. I mean, you can’t swing a dead cat without having somebody make the comparison between Russia-Ukraine, and China and Taiwan. That has definitely been a catalyst here. I think the issue of China and Taiwan was going to come up anyway in terms of the Party Congress in the fall and Xí Jìnpíng 习近平 getting a likely third term. And then there’s also this idea out there put out by a former admiral at PACOM that 2027 is this year that Xi would like to resolve the Taiwan issue and seek reunification.
Kaiser: That’s only based on the 100th anniversary of the founding of the Red Army.
Paul: Right. It’s sort of a-
Kaiser: It’s kind of a silly…
Paul: China and Taiwan analysts will sort of say that they don’t necessarily agree with that as an artificial date. But the point is we’re now in 2022, and now with the chip shortage as we’ve talked about, the idea that TSMC is so critical here to the global value chain is well known. And so it’s like, oh my God, what’s going to happen to Taiwan in five years? Is China going to march in and take over those foundries? The issue is now sort of on the horizon in a way it wasn’t.
Kaiser: For quite some time, Taiwan had always talked about its semiconductor industry as a silicon shield. I mean, Taiwan believed that its very dominance in this just incredibly vital sector would prevent tensions from spilling over because it helped to dissuade Beijing, or to even to deter Beijing, because Beijing presumably would never risk killing the goose that lays the golden eggs or the silicon eggs as it were. Has all this changed? I mean, has the calculus changed?
Paul: Well, it’s a great question. I think it’s a big topic and we probably should do a podcast just on that, but so it’s a big unknown. The big unknown here is to what degree the semiconductor issue is now part of Beijing and Xi Jinping’s calculus on Taiwan? This year, for example, we did see for the first time, at least that I’ve seen, calls from academics for China to prevent Taiwan from sort of being pushed into the blue supply chain and away from the red supply chain. But this is a complex equation. The semiconductors alone, of course, are not gonna affect the China-Taiwan dynamic in ways that are similar to, say, the Russia-Ukraine issue, right? The semiconductor situation definitely adds to the tensions in ways that I think we don’t really fully understand here.
This happens in a couple ways, the policies the U.S. is taking, for example, which could result in a cutoff of all Chinese companies from using TSMC as a manufacturing platform, that’s a real wild card. We saw, of course, Huawei cut off and we’ve seen many other Chinese companies cut off, DJI, Phytium, and then of course all the AI companies who have all been cut off from the U.S. as a result of the entity list action and could also all eventually be cut off from using TSMC as a manufacturing base if the U.S., for example, decided to extend the foreign direct product rule to all those companies, as some in Congress have in fact advocated. So, if we get a Republican Congress, a correctional sweep in the fall for example, we could see calls for that.
The problem is though in this increasing tension that you mentioned over Russia-Ukraine that sort of focused on China-Taiwan, we’ve seen, for example, Yáng Jiéchí 杨洁篪 and Jake Sullivan meeting periodically to sort of de-risk the situation, because, of course, China now firmly believes the U.S. is hollowing out the One China policy on Taiwan. But I doubt if the semiconductor issue is part of those conversations. The Chinese aren’t gonna say, “Hey, you’re messing with Taiwan in terms of semiconductors.” It’s probably not gonna happen.
But the Chinese rhetoric around the issue has gone up quite a bit. So, when this DUV issue that you mentioned, there were potential restrictions on that came up, Chinese spokespersons were calling it things like techno-terrorism and really tough language on that. Chinese officials, of course, are aware of all of this stuff. They’re aware of all these export controls. They’re aware of the Taiwan issue, but how it factors into their calculus, and particularly Xi’s calculus, on Taiwan is complicated. I’ve called it a sort of unknown red line.
Kaiser: Yeah. That was a good phrase. Yeah.
Paul: They’re the known red lines around that we’ve seen pushed quite a bit this year on both sides. And then there’s this sort of semiconductor piece, because one day of disruption of the airspace over Taiwan, for example, would send these huge ripples through the global economy, because you’re talking about airplanes taking off and ships every day that are bringing those semiconductors and assemblies to all parts of the world. So, it’s not an academic question as to how China thinks about this.
To some degree, I think it is a deterrence. It has been a deterrence because Chinese companies are benefiting from all this great manufacturing capacity in Taiwan. But if the U.S. sort of cuts off that capacity, then the calculus changes in Beijing, right? It’s like, “Hey, why are we allowing the U.S. to separate Taiwan in this key sector?”
Kaiser: Yeah. Let’s talk about the U.S. calculus of deterrence. I mean, walk me through this when it comes to Washington, Beijing, Taiwan, do you think that the changes that we’re seeing in just recent years have actually enhanced American and Taiwanese ability to deter China, or do some of these changes actually, in your mind, undermine deterrence?
Paul: I would argue that I think the changes, by and large, are undermining deterrence. Because as I’ve noted, and I’ve written about this quite a bit in other articles, to the extent that you’re cutting off Taiwan from the mainland, from the red supply chain, you’re undermining the silicon shield. You’re preventing Chinese companies from using Taiwan, at the same time as you’re doing things like trying to freeze China at 14 nanometers, right?
If you’re the Chinese government, which views Taiwan as a domestic problem, Taiwan as part of China, and you’re seeing sort of salami tactics, where the U.S. is not only cutting off Chinese companies, but is enticing TSMC to build an advanced fab in the U.S. which they will not build in China, of course, because the Taiwan government will not allow that. TSMC has facilities in Shanghai and Nanjing, for example, but they’re around 28 nanometers, right? They’re nowhere near cutting-edge and that could not happen. So, it’s not just one piece of this. It’s sort of the whole picture for Beijing becomes hard to parse here. It’s like that the U.S. is peeling off a critical piece of the technology equation and preventing China from, both manufacturing at advanced levels and from using Taiwan.
So, the combination of those things, plus the attempt to attract more Taiwan companies to come to the U.S. to build stuff that has to at least be part of the calculus in Beijing about how they might deal with a Taiwan situation if these many other political factors became more problematic, right? Again, nobody thinks that China is gonna go to war or invade Taiwan just over the semiconductor piece, but it adds a really important element that I don’t think gets enough attention, frankly, because I think it’s huge.
Again, I don’t think Yang Jiechi and Jake Sullivan are talking about it. But from the Chinese point of view, the fact that this academic, probably with what permission from the government, is talking about this issue now. I think we’ll see more of that as the months progress here about how the U.S. is seen as sort of peeling off Taiwan in general, and then the semiconductor industry piece of it is pretty critical here.
Kaiser: Yeah. I want to go back to the chip shortage, and I think we know what lessons the Biden administration seems to have learned from it, coming as it did during the COVID pandemic. We’ve seen what their response has been. What are the lessons that the Chinese leadership have learned from the chip shortage? How has that changed thinking on the part of China?
Paul: That’s a great question. I think the short answer is that Beijing has realized that the U.S. indeed has these choke point technologies and is fully capable of weaponizing the supply chains and undermining the business models of Chinese companies. Huawei has gone from a 140 billion company to heading south of a hundred billion pretty fast here, right? They’re trying to reinvent themselves, but it’s a pretty big deal, right? It’s like having a Chinese government action undermine the business model of Apple or Google, and then those companies are in sort of a downward death spiral.
The Chinese government now is in this position of trying to figure out how to deal with that. When I talk to Chinese industry people, they get how hard it would be, for example, for China to recreate the entire semiconductor industry in China. It’s virtually impossible. I don’t know if there’s enough money in the world to do that, more the numbers of people that you would need to do that. But the key players in the industry, for example, the guy who briefs Xi Jinping on the industry, he recognizes this, but he’s under pressure to help make policy that will reduce China’s dependence.
So that means they’re pulling out a lot of stops. They have a multitiered strategy, all sorts of incentives to the semiconductor industry. Using domestic capital markets like the STAR board in Shanghai, companies can list there. There’s virtually an unlimited amount of cash in China, if you’re a semiconductor company, right? The country is awash in cash. The problem is that it’s not about the cash, right?
It’s about the personnel. It’s about the technology and iterating the technology. And it’s a result of…mostly it has been being plugged into these global value chains and developing trust with customers so you can continue to grow and have enough customer base to reinvest the money, and new technology, and new capital equipment, and new factories. That dynamic is tricky. It’s tricky to reinvent that in China in ways that are going to be sustainable. I think right now we’re at a weird point where we don’t really know how far China will be cut off, for example, from that global value chain and that equipment, and from other things, and how quickly they can train engineers and poach, or poach engineers from Taiwan, or whatever, right?
Paul: Taiwan has passed all these new laws to prevent poaching and prevent Chinese from setting up companies in Taiwan to have access to technology. I think it’s a weird experiment in the sense that no country has ever tried to do even big parts of the global value chain on its own. It’s always been this sort of division of labor driven by comparative advantage and driven by market forces that has happened. Now we have governments in China and in the U.S. sort of messing with this global value chain supply chain, and trying to figure out what to do.
Kaiser: One of the other options available to China is to try to dissuade the American government from continuing down this path of trying to dissuade them from messing with the supply chain, as you say. Maybe one argument working in their favor is that, I’ve heard from several people this, the entity listing of Huawei and of other companies actually drove hoarding of chips by a lot of players. And that was part of the reason for, not the entire reason obviously, but it contributed certainly to the chip shortage. Do you buy that argument and do you think that works to China’s advantage?
Paul: Great question, Kaiser. Look, I’ve talked to a lot of people about this. It’s a complicated issue. There are two camps. One is a group that I’ve talked to in industry that thinks that the Huawei entity list action in May 2019 was really the cause of the whole global chip shortage, at least the start of it. Of course, then all these other things came in which contributed like the pandemic fires in various factories and freezes in Texas. But the real fundamental event was putting Huawei on the entity list. It’s a huge company. It bought huge amounts of semiconductors. And then all the other Chinese companies that were afraid of similar action began to ramp up their orders with various suppliers and that sort of disrupted the whole supply chain of the industry.
The other camp believes that, yes, it was important. But it was sort of a contributory factor, which maybe was 15% to 20% to the global shortage and certainly made it worse, but was not really the dominant factor. Now, it’s tricky. I think the truth is probably somewhere in between there, but whatever, it had a huge impact on either making it worse or starting it. And as a perfect example of a sort of government intervention in a complicated global supply chain that definitely had unintended collateral damage of some size, depending on who you believe.
It’s really hard to collect data on this because you’d almost need like a time series of procurement for all the different companies across the world, and then look at sort of May 19th, and then see how it changed, and that data is really hard to come by. And nobody has really run the numbers on that. But people who I trust in industry, who are working very deep in the industry, there’s at least a belief that the whole chip shortage was a result of the Huawei action, or at least, in a minimum, contributed substantially to it.
Kaiser: I wonder if it’s still possible to try to dissuade China from pursuing technological self-reliance. I mean, okay, if I’m sitting in Beijing, do I have any reason to believe that the United States’ unspoken policy is not: “I’m going to kneecap China. I’m going to see China on its belly, technologically. I will not do anything. I won’t stop until China has been eliminated as a serious technological competitor.” I mean, on an even more pragmatic level, it’s like, okay, so even if in 2024 we get a guy or a woman who is going to let off a little bit, what’s the chances that four years later, that won’t completely change again?
I mean, I have to think they’re all in on this. I wonder if, are we doing ourselves any favors by lighting a fire to Beijing’s efforts to wean itself off of the global supply chain and to try to reproduce it domestically? I mean, Christ, it seems just such a colossal waste for China and for the rest of the world.
Paul: No, it absolutely does. I mean, to think of recreating these value chains in different parts of the world…it’s been this sort of virtuous division of labor driven by market forces, driven by really innovative companies like TSMC, that have enabled us to pack all that power into an iPhone. That’s unbelievable. When you start unraveling that you know, the question is who understands it well enough to help unravel it, and what are gonna be the unintended consequences?
Last year, for example, we helped a big semiconductor company submit their report to this request for information from the Commerce Department about their supply chain, how their supply chain worked, and how they had handled the global chip shortage. What became clear from that exercise was, of course, nobody has visibility on all the pieces of this complicated supply chain.
To expect that governments are going to be able to intervene here in an efficient way and make decisions that reflect the reality of the industry, to me, is a bit of a tall order. But in terms of your question, are we lighting a fire under Beijing on this? I think there’s a short answer: absolutely, of course. As we’ve seen just in Beijing’s actions in the last two years to try to ramp up all these different levels of support for self-reliance, the rhetoric, of course, from the top is all about self-reliance.
The long answer, though, is really tricky, because it’s not clear to me that state directed industrial policy, coupled with whole industries being cut off from aspects of the global economy, and R&D, and other commercial value chains is gonna produce a winner here. We’re in sort of uncharted territory here. No country can do everything in all the tech sectors. And semiconductors is probably the poster boy here, right? Maybe China can be the exception because they have this big domestic market, and they have massive government inputs. But you still can’t change sort of the underlying equation where you need some incentives for entrepreneurs to do stuff. You need diversity of the workforce, diversity of workforce education. You need labor mobility, and you need access to capital markets that are very liquid and flexible. So, I’m not sure how you get all those.
Kaiser: Yeah, and you need time. You need time to develop the kind of process knowledge that’s so important to these technologies. What I really worry about is these measures from the U.S. actually forced a stronger reaction from Beijing than what we want. They’re causing China to take an action that it might not otherwise have taken, several actions, including a lot of just the kind of bad behavior. I’m talking about industrial espionage and that sort of thing that we worry about in the first place. I don’t know that we’re doing anything to dissuade them. Anyway…
Paul: No, no, that’s a good thought. Quickly on that, I mean, that has come up, right? Like, we’re gonna force them to do all this, and so they’re gonna step up cyber espionage and steal all the stuff. The problem with that argument is I just don’t buy it. I mean, stealing IP is not a business model. And it really, again, doesn’t allow you necessarily to advance. There’s many, many pieces that are unspoken kinds of capabilities you need that aren’t written down, like engineering or tweaking out a semiconductor manufacturing supply chain.
You can’t steal all elements of that. You still have to actually do it yourself and do it domestically. Some of that might happen. There might be, of course, efforts to steal IP, but it’s still… That’s dangerous then, because then of course, you really are gonna talk about decoupling of a whole set of relationships over supply chain and R&D, and everything that is not really…
Kaiser: Yeah, a vicious circle.
Paul: Again, the question is what’s the benefit, the cost benefit here in terms of gains to U.S. national security for example, and the risks and the more quantifiable risks, you could argue, of breaking up this virtuous industry in ways that are only gonna drive up costs and sort of hurt everybody? My bottom line here is nobody has run the numbers on this, right? Nobody is saying, “Wow, if we restrict all the stuff going to China over all these areas, what’s the benefit to U.S. national security, for example, and what’s the cost?” Nobody has, as far as I can tell, has run those numbers and made that cost benefit analysis.
I think that’s really dangerous in the policy community to understand how this affects your own industry, for example as I mentioned, and how this affects the global industry. I think that’s sort of sorely needed, is actually real data to understand the downsides of some of these policies.
Kaiser: Paul, let me wrap up with one question here. Nancy Pelosi is planning a trip to Taiwan in August. Beijing has already hauled out some of its strongest language in diplomatic remarks, but the U.S. seems pretty determined to push the limit, so what happens now? What’s Beijing’s likely response to this?
Paul: Well, great question, Kaiser. That sort of gets out of the semiconductor space and into the U.S.-Taiwan space. I think it’s important to understand that, if Pelosi goes, that would come… On the one hand, it’s not unprecedented, right, for senior U.S. congressional visitors to Taiwan, but on the other hand, it comes on the heels of a series of events this year. That China, at least, has interpreted, you can argue whether China is overreacting or overinterpreting U.S. action, but it comes after a series of actions: visits in February and March by unofficial delegations, Mike Pompeo, Evan Medeiros, others. And then this misstatement in May that the State Department put up that left out that the U.S. doesn’t support Taiwan independence.
And then of course, President Biden’s off the cuff remarks in Tokyo about coming to the defense of Taiwan. China really has, I think, concluded that the U.S. is intent on hollowing out the One China policy. And again, you can argue over whether they’re overreacting to that, but that’s the perception from all the Chinese leaders that I’ve talked to and academics that I’ve talked to over the last few months. The Pelosi thing would come at arguably one of the most tense moments ever in U.S.-China relations, right? Because the broader relationship is in such tatters. There’s very little communication between leaders except at the sort of Biden C level.
And then this sort of in between, you have these de-risking meetings where Jake Sullivan rushes off to Luxembourg or Geneva to put out a fire. That’s not really a recipe, I think, for long term stability here. Then the military leaders haven’t talked to each other. I was at a conference last week where a former PACOM commander noted that when he was PACOM commander, he had the cell phone numbers of Chinese military counterparts. And so in the event of a Chinese ship running too close to a U.S. ship, he could send them video, and they would respond to that.
Now we have nothing like that. So, the potential for an accidental response to this. For example, I assume if Pelosi lands in Taiwan, the Chinese will launch a series of sorties into the Taiwan ADIZ, maybe even crossing the median as a sign of protest. Now, what if you have an accident as a result of that too? A Taiwan plane gets too close, you have a collision. All this raises the potential for an inadvertent thing. But it also raises the stakes for China, which increasingly sees the U.S. pushing and pushing and pushing and pushing on Taiwan, and could be pushed into a position where they have to make some gesture.
Now I was in China in 1996 when Beijing launched those mobile missiles north of Taiwan. They could do a lot of things that are short of some kind of a full up invasion, right? They could threaten the offshore islands. They could demand demilitarization of some of those offshore islands. They could take some of those islands. And then the U.S. would be in a position where the status quo had changed significantly, and there was a need to respond here. My hope would be that something like that would force a whole renegotiation of the complicated trilateral relationship. Those documents that govern the relationships — the Three Communiqués, the TRA, the Taiwan Relations Act, and the Six Assurances — those are 50 years old, right?
Paul: More than 50 years old, some of them. Things have changed so much since then on all sides that the argument could be made, “Hey, we need a new way to keep the status quo from getting out of hand.” But the problem with that is that requires some trust between Beijing and Washington, which is now, not even in short supply, it’s basically zero. And so, you’d have to sort of rebuild trust before you could even get to the point where you could have a frank discussion about the future of Taiwan and how to prevent anything from going awry on the military front.
Kaiser: Well, on that cheery note, it’s great to have you back on the show as always. Thank you for the really amazingly granular discussion on the semiconductor industry. And Paul, keep up the great work that you’re doing. Let’s move on to recommendations. First, a quick reminder that the Sinica Podcast is powered by SupChina. And if you like what we’re doing with this and other shows in the Sinica Network, by all means, please show your support by subscribing to our China ACCESS newsletter.
While you’re at it, sign up for our business focused ChinaEDGE newsletter, and definitely check out ChinaEDGE Live with the formidable Lizzi Lee. It’s on YouTube, just search up, ChinaEDGE Live, Lizzi Lee, L-I-Z-Z-I L-E-E. You will dig it. Coming soon also to your favorite podcast app on your favorite podcast network. Okay, on to recommendations. Paul, man, what do you have for us this week?
Paul: Okay. I recommend a couple things. One is A Natural History of the Future by Rob Dunn. And it’s looking at sort of the reality of how biology is impacting all of us, whether it’s population or pandemics. And so it’s a really sobering look at the future. And I think it’s a good read about tracing out some of the trends we’re seeing now and out into the next period. On the China front, I would also recommend Ryan Hass’s book which is…
Kaiser: Stronger, it’s called. Yeah. He’s been on the show to discuss it before.
Paul: And he’s probably discussed it. I definitely recommend that. There’s a whole host actually of China-related books, which probably we should have people watching China rate.
Kaiser: Yeah. I’ve got a whole bunch of them that I’m reading for upcoming shows. Fantastic. Save some of those that you really find that you like for the next one. But Rob Dunn, A Natural History of the Future, that sounds right up my alley. I’m definitely gonna get a hold of that. Thank you.
Paul: Yeah. You’d like it.
Kaiser: Yeah, I’m sure I would. All right. My recommendation is going to be way more frivolous. It’s just a TV show. It just concluded its third, I think third season, yeah, on Amazon Prime, it’s called The Boys. And it’s sort of an anti-superhero. I think it’s about a superhero universe where there’s this gigantic corporation that’s been able to sort of manufacture these people with super abilities. And of course, most of them are complete jerks, including one who is a thinly disguised, kind of Trumpian figure named Homelander, who is sort of this blonde and just smug-looking total jerk. I mean, he’s thoroughly evil and ego maniac. He is Trump incarnate, but a fantastic show. It’s actually really, really good.
Paul: That sounds good.
Kaiser: Yeah. It’s not for everybody. It is really violent. I mean, just like it takes violence to gratuitous extremes. They’re very unafraid of a lot of sexual situations and nudity and things like that. So, it’s not for the kids, but definitely check it out. It’s kind of a good antidote to the sort of happy Marvel stuff, but really good. All right. Anyway, The Boys. Last season just concluded.
Paul, man, what a pleasure as always. You are just so full of incredible depths of knowledge on this stuff.
Paul: My pleasure, Kaiser.
Kaiser: Thank you so much, and we’ll have you back on again, not too long.
Paul: Yeah. I could already see when I was preparing, there’s like three other podcasts in there, and sorry for all the long answers. Last thing I’ll say, it is a complicated industry. And one of the things that’s a little frustrating is that everybody is an expert on semiconductors, and so you have a lot of stuff coming out. I see every day that I think, wow, who wrote this? What are their credentials? It really does require at least some time to develop an understanding. I work with clients on a daily basis, trying to understand their problems in the industry across the globe. And I think it requires that kind of attention to it.
And I see a lot of stuff that’s written that I’m like, wow, that doesn’t really reflect the reality of the industry. So, that’s one of the challenges in this space is, because it’s such a hot space, everybody feels they need to write on it. And sometimes the quality of that writing is just a little… It can be agenda driven if there’s a need to make a policy recommendation. And it doesn’t always reflect the sort of reality of the industry, which is quite complicated and requires some long-term exposure to, I would argue at least.
Kaiser: I would encourage everybody to read Paul’s very informed writing on semiconductors and other technology issues on SupChina. He’s a regular contributor to us and we really value the essays that you’ve written for us. So, thanks so much.
Paul: And I’ll probably do one soon on the DUV stuff for you.
Kaiser: Great, great, great. I look forward to it. So, thanks for coming on and we’ll see you again soon.
Paul: Thanks. What a pleasure always. I love the venue.
Kaiser: All right. The Sinica Podcast is powered by SupChina and is a proud part of the Sinica Network. Our show is produced and edited by me, Kaiser Kuo. We would be delighted if you would drop us an email at firstname.lastname@example.org, or just give us a rating and a review on Apple Podcasts as this really does help people discover the show. Meanwhile, follow us on Twitter or on Facebook at @supchinanews, and be sure to check out all the other shows in the Sinica network. We’ve got new shows coming very soon. Thanks for listening and see you next week. Take care.