1 00:00:09,960 --> 00:00:14,080 Speaker 1: Hello, and welcome to another episode of the Odd Lots Podcast. 2 00:00:14,160 --> 00:00:17,439 Speaker 1: I'm Geo Wist and I'm Tracy Alloway. Tracy, do you 3 00:00:17,440 --> 00:00:19,680 Speaker 1: ever want to write a book? One day? Yes? You 4 00:00:19,720 --> 00:00:22,360 Speaker 1: know this, Joe. You know this because I try to 5 00:00:22,360 --> 00:00:24,079 Speaker 1: make you write a book with me all the time. 6 00:00:24,200 --> 00:00:26,000 Speaker 1: I don't want to write a book. It seems like 7 00:00:26,040 --> 00:00:28,120 Speaker 1: a lot of work. No, it would be fun. Well, 8 00:00:28,560 --> 00:00:30,480 Speaker 1: first of all, it is a lot of work. I 9 00:00:30,480 --> 00:00:33,199 Speaker 1: wrote two book proposals last year and it nearly killed me. 10 00:00:33,720 --> 00:00:36,080 Speaker 1: That's a very long story for another time. But I 11 00:00:36,080 --> 00:00:38,800 Speaker 1: think it would be fun. We should do it. All 12 00:00:38,880 --> 00:00:41,479 Speaker 1: thoughts the book come on. Maybe I could see an 13 00:00:41,479 --> 00:00:43,560 Speaker 1: Odd Lots book. But here's what would worry me about 14 00:00:43,600 --> 00:00:46,000 Speaker 1: writing a book, which is you, we could have a 15 00:00:46,040 --> 00:00:49,960 Speaker 1: book idea today, in two maybe we get a deal 16 00:00:50,200 --> 00:00:53,360 Speaker 1: sometime in and then they're like it's going to come 17 00:00:53,360 --> 00:00:56,960 Speaker 1: out and fall six, congrats like that. It's like, I 18 00:00:56,960 --> 00:00:59,880 Speaker 1: don't have time. It's just too long. And then here's 19 00:00:59,920 --> 00:01:01,680 Speaker 1: the thing that I would worry about is that by 20 00:01:01,680 --> 00:01:05,560 Speaker 1: the time like August two comes out, it's like maybe 21 00:01:05,600 --> 00:01:07,520 Speaker 1: no one at all cares about what we want to 22 00:01:07,560 --> 00:01:09,080 Speaker 1: say in our book. All I'm going to say in 23 00:01:09,120 --> 00:01:11,759 Speaker 1: response to that, is how many game stop books came 24 00:01:11,760 --> 00:01:15,760 Speaker 1: out in like, there were at least three, and they 25 00:01:15,800 --> 00:01:18,559 Speaker 1: all came out within a year of it actually happening. Yeah, 26 00:01:18,640 --> 00:01:21,440 Speaker 1: and by then no one cared anymore. There's my point, Okay. 27 00:01:21,560 --> 00:01:24,360 Speaker 1: I take the point that it's very difficult to write 28 00:01:24,480 --> 00:01:27,960 Speaker 1: a well researched book that says something new and interesting 29 00:01:28,520 --> 00:01:31,480 Speaker 1: on a topic at exactly the right time ahead of 30 00:01:31,520 --> 00:01:34,120 Speaker 1: it kind of blowing up in the public consciousness. That 31 00:01:34,280 --> 00:01:37,280 Speaker 1: is difficult. But our guest today has done that. We're 32 00:01:37,280 --> 00:01:39,520 Speaker 1: going to be speaking to a guest who has a 33 00:01:39,520 --> 00:01:42,080 Speaker 1: new book out, and I think the timing could not 34 00:01:42,160 --> 00:01:43,679 Speaker 1: have been better. I don't even know how the timing 35 00:01:43,720 --> 00:01:47,039 Speaker 1: was possible, right, So this is actually a book on 36 00:01:47,080 --> 00:01:50,000 Speaker 1: a topic that we First of all, we should say 37 00:01:50,000 --> 00:01:52,840 Speaker 1: that we have had this guest on before talking about 38 00:01:52,840 --> 00:01:57,120 Speaker 1: a topic that kind of became core all thoughts content 39 00:01:57,240 --> 00:01:59,680 Speaker 1: and also a core issue for the rest of the world. 40 00:02:00,200 --> 00:02:03,360 Speaker 1: It truly exploded into the public's consciousness in a way 41 00:02:03,360 --> 00:02:06,400 Speaker 1: that I don't think any of us weren't necessarily expecting 42 00:02:06,440 --> 00:02:09,040 Speaker 1: when we first started talking about this. I think that 43 00:02:09,120 --> 00:02:13,119 Speaker 1: was back in like late nineteen and then COVID happened, 44 00:02:13,280 --> 00:02:17,679 Speaker 1: and everyone became focused on this particular issue or thing, 45 00:02:18,160 --> 00:02:20,280 Speaker 1: and then fast forward a couple of years and it 46 00:02:20,320 --> 00:02:24,519 Speaker 1: also became an issue of strategic and even military importance. 47 00:02:24,960 --> 00:02:28,720 Speaker 1: That's right. So we've done semiconductor episodes in the past. 48 00:02:28,800 --> 00:02:30,680 Speaker 1: We've done several of them, and we sort of tried 49 00:02:30,680 --> 00:02:33,640 Speaker 1: to trace the semi conductor supply chain, and one of 50 00:02:33,680 --> 00:02:36,160 Speaker 1: the companies at the end or somewhere in the semi 51 00:02:36,200 --> 00:02:39,480 Speaker 1: conductor supply chain that's very important is the Dutch Advanced 52 00:02:39,560 --> 00:02:43,359 Speaker 1: Lithography Maker a SML, which applies equipment, and so we're like, okay, 53 00:02:43,400 --> 00:02:45,239 Speaker 1: who should we talk to about a s m L. 54 00:02:45,639 --> 00:02:47,840 Speaker 1: I never said, you gotta talk to this guy, Chris Miller, 55 00:02:47,880 --> 00:02:49,760 Speaker 1: who knows all about them. And at the time, he 56 00:02:49,880 --> 00:02:52,239 Speaker 1: was writing this book, and the book has just come out. 57 00:02:52,320 --> 00:02:54,480 Speaker 1: It's out, and of course the timing is perfect. So 58 00:02:54,560 --> 00:02:57,760 Speaker 1: Chris Miller, he is the author of Chip War, The 59 00:02:57,840 --> 00:03:01,120 Speaker 1: Fight for the World's Most Critical techno Oology. And of 60 00:03:01,160 --> 00:03:05,520 Speaker 1: course we just had the Biden administration announcing new restrictions 61 00:03:05,639 --> 00:03:09,280 Speaker 1: on semi conductor technology export to China, which is something 62 00:03:09,320 --> 00:03:11,959 Speaker 1: we briefly touched on in a recent episode with Dan Wong, 63 00:03:12,520 --> 00:03:15,280 Speaker 1: But it is truly the perfect guest with the perfect timing, 64 00:03:15,440 --> 00:03:18,280 Speaker 1: the perfect guest with the perfect timing. I'm thrilled that 65 00:03:18,360 --> 00:03:20,640 Speaker 1: we have Chris back on the show to talk about 66 00:03:20,639 --> 00:03:22,760 Speaker 1: this new action and how they fit in the broader 67 00:03:22,800 --> 00:03:25,880 Speaker 1: context and his research. So Chris Miller, thank you so 68 00:03:25,960 --> 00:03:28,760 Speaker 1: much for coming back on the podcast, Thanks for having 69 00:03:28,760 --> 00:03:32,840 Speaker 1: me back, and congratulations. Just like incredible timing, because I mean, 70 00:03:32,840 --> 00:03:35,520 Speaker 1: the book just came out right and then right then 71 00:03:35,640 --> 00:03:39,760 Speaker 1: we have everybody talking about the chip war basically because 72 00:03:39,800 --> 00:03:42,840 Speaker 1: of these actions by the Biden administration. Well, that's right, 73 00:03:42,880 --> 00:03:45,800 Speaker 1: that this the timing of the regulations that were released 74 00:03:45,800 --> 00:03:47,440 Speaker 1: a couple of weeks ago. We're a bit of a 75 00:03:47,480 --> 00:03:50,160 Speaker 1: surprise to me, but I think the trajectory was pretty 76 00:03:50,200 --> 00:03:52,560 Speaker 1: clear from the last time we spoke about two years ago, 77 00:03:52,720 --> 00:03:55,640 Speaker 1: that we're going to see more tightening of controls and 78 00:03:55,680 --> 00:03:59,880 Speaker 1: more bifurcation between the US and China developing separate some 79 00:04:00,000 --> 00:04:05,240 Speaker 1: conector ecosystems. Wait, was the trajectory clear, because you know, 80 00:04:05,360 --> 00:04:07,880 Speaker 1: my impression was there was a lot of noise during 81 00:04:07,880 --> 00:04:12,880 Speaker 1: the Trump administration about technology security and these types of issues, 82 00:04:13,160 --> 00:04:16,640 Speaker 1: but no one was necessarily expecting the Biden administration to 83 00:04:16,680 --> 00:04:20,160 Speaker 1: take this up to the extent that it did. So 84 00:04:20,200 --> 00:04:23,760 Speaker 1: the export controls that were announced recently, well maybe you 85 00:04:23,760 --> 00:04:27,279 Speaker 1: can tell us, like how extreme are they in your view? 86 00:04:27,880 --> 00:04:31,200 Speaker 1: But do you think it was expected? Were you expecting 87 00:04:31,279 --> 00:04:34,320 Speaker 1: something along those lines? Well? I think so. I mean, 88 00:04:34,400 --> 00:04:38,400 Speaker 1: if you look at the Trump administration's actions, I think 89 00:04:38,440 --> 00:04:41,720 Speaker 1: you've got to differentiate a bit between sort of what 90 00:04:41,760 --> 00:04:44,479 Speaker 1: the Twitter feed was showing any individual day and the 91 00:04:44,520 --> 00:04:47,479 Speaker 1: direction of travel of the National security bureaucracy underneath it. 92 00:04:47,560 --> 00:04:50,800 Speaker 1: And I think the bureaucracy was trending in the same 93 00:04:50,839 --> 00:04:54,520 Speaker 1: direction actually, from the late Obama administration in terms of 94 00:04:54,800 --> 00:04:59,279 Speaker 1: greater skepticism of inbound investment into U S SMI connector firms, 95 00:04:59,320 --> 00:05:03,680 Speaker 1: from in terms of greater concern about the difficulty of 96 00:05:03,960 --> 00:05:07,799 Speaker 1: preventing chips from reaching military end users in China, greater 97 00:05:08,040 --> 00:05:10,960 Speaker 1: confidence that computing power and some my conductors would be 98 00:05:11,000 --> 00:05:14,560 Speaker 1: critical to the future of military power, and thus concerned 99 00:05:14,600 --> 00:05:17,240 Speaker 1: that China's access to U S ship technologies would have 100 00:05:17,240 --> 00:05:21,080 Speaker 1: a military benefit. All that has been developing for really 101 00:05:21,080 --> 00:05:24,920 Speaker 1: about a decade in Washington and is less driven by 102 00:05:25,120 --> 00:05:28,200 Speaker 1: the question of of who's president and more the collective 103 00:05:28,200 --> 00:05:30,760 Speaker 1: assessment of the Pentagon and the Commerce Department and then 104 00:05:30,800 --> 00:05:35,120 Speaker 1: our security councils involved, and that's been a pretty steady intensification, 105 00:05:35,160 --> 00:05:38,160 Speaker 1: even though the politicians at the top have have differed. Sorry, 106 00:05:38,160 --> 00:05:40,680 Speaker 1: I'm just looking again at you know, your book was 107 00:05:40,720 --> 00:05:44,440 Speaker 1: released October four, and the restrictions I think we're announced 108 00:05:44,520 --> 00:05:47,880 Speaker 1: October seven or right around there, so basically the same day. 109 00:05:47,920 --> 00:05:51,640 Speaker 1: But you know, before we get again to these specific restrictions, 110 00:05:51,680 --> 00:05:53,360 Speaker 1: you know, I think when we talked about a s 111 00:05:53,440 --> 00:05:56,080 Speaker 1: m L in our episode, you know, even at that time, 112 00:05:56,600 --> 00:06:00,760 Speaker 1: a s mL, I believe was restricted from exporting its 113 00:06:00,839 --> 00:06:05,440 Speaker 1: most advanced technology to Chinese companies. And maybe they were 114 00:06:05,640 --> 00:06:08,480 Speaker 1: still able to export some technologies, but not the most 115 00:06:08,560 --> 00:06:12,000 Speaker 1: advanced lithography. You know, you mentioned some of the factors 116 00:06:12,279 --> 00:06:16,479 Speaker 1: driving concerns about constraining Chinese semiconductors. But where did this 117 00:06:16,560 --> 00:06:21,000 Speaker 1: really start the story of the US deciding that, you know, 118 00:06:21,080 --> 00:06:25,040 Speaker 1: limits on technology transfer would be part of our foreign 119 00:06:25,040 --> 00:06:28,119 Speaker 1: policy with China. Well, I would actually start the story 120 00:06:28,320 --> 00:06:32,320 Speaker 1: before China was the focus, because there have been limitations 121 00:06:32,400 --> 00:06:35,479 Speaker 1: on semi conductor technology transfer since the invention of the 122 00:06:35,480 --> 00:06:38,599 Speaker 1: first chips. And if you think back to where chips 123 00:06:38,600 --> 00:06:42,600 Speaker 1: came from They emerged out of demands to miniat your eyes, 124 00:06:43,080 --> 00:06:46,400 Speaker 1: computing power to fit and guidance computers on missiles during 125 00:06:46,440 --> 00:06:48,960 Speaker 1: the Cold War, and so there's been a deep into 126 00:06:48,960 --> 00:06:53,159 Speaker 1: relationship from the earliest days between defense technology and the 127 00:06:53,240 --> 00:06:56,120 Speaker 1: chip industry, and so controls have been part of the 128 00:06:56,160 --> 00:06:59,559 Speaker 1: US government strategy since that time. So during the Cold 129 00:06:59,560 --> 00:07:03,280 Speaker 1: War So Union, there were really extensive controls placed both 130 00:07:03,320 --> 00:07:05,479 Speaker 1: on the transfer of chips to the Soviet Union and 131 00:07:05,520 --> 00:07:09,239 Speaker 1: also the transfer of chip making equipment and so in 132 00:07:09,240 --> 00:07:11,600 Speaker 1: in some ways, what we're seeing today is just a 133 00:07:11,640 --> 00:07:16,080 Speaker 1: continuation of that process of trying to control technology transfer 134 00:07:16,120 --> 00:07:19,160 Speaker 1: that's militarily relevant. Now it's different because the target is 135 00:07:19,200 --> 00:07:21,840 Speaker 1: different at this time it's China. And it's different because 136 00:07:21,920 --> 00:07:24,600 Speaker 1: unlike in the earliest days of the chip industry and 137 00:07:24,760 --> 00:07:28,559 Speaker 1: the late nineteen fifties and nineteen sixties, at that point, 138 00:07:28,600 --> 00:07:31,080 Speaker 1: most chips were going towards some sort of military use, 139 00:07:31,080 --> 00:07:35,040 Speaker 1: whereas today percent of chips are going towards civilian uses. 140 00:07:35,080 --> 00:07:37,640 Speaker 1: So so that's different. But the basic concept that you 141 00:07:37,680 --> 00:07:40,600 Speaker 1: want to control the ability to produce advanced ships because 142 00:07:40,640 --> 00:07:44,960 Speaker 1: they have military uses is basically something we've always had now, 143 00:07:45,160 --> 00:07:48,119 Speaker 1: certainly there's been a shift in the policy towards China 144 00:07:48,160 --> 00:07:50,760 Speaker 1: over the past couple of years, driven by a couple 145 00:07:50,760 --> 00:07:53,600 Speaker 1: of factors. I think one most important is that the 146 00:07:53,640 --> 00:07:56,160 Speaker 1: military balance in general between the U S and China 147 00:07:56,240 --> 00:07:59,240 Speaker 1: has shifted. Whereas a decade or two ago is clear 148 00:07:59,320 --> 00:08:01,800 Speaker 1: the U S military area was far stronger, today that's 149 00:08:01,880 --> 00:08:05,080 Speaker 1: less clear, especially when you look at potential hotspots like 150 00:08:05,120 --> 00:08:08,240 Speaker 1: the Taiwan Straits, and so it's easy to find analysts 151 00:08:08,280 --> 00:08:11,400 Speaker 1: who say the US would have a hard time defending Taiwan, 152 00:08:11,560 --> 00:08:13,840 Speaker 1: which wouldn't have been the case a decade or two ago. 153 00:08:13,920 --> 00:08:17,040 Speaker 1: So that's that's one. More concerned about the Chinese military capabilities. 154 00:08:17,520 --> 00:08:20,440 Speaker 1: Two is that China plays a bigger role than ever 155 00:08:20,520 --> 00:08:24,360 Speaker 1: before in the global electronics and computing supply chain. Still 156 00:08:24,640 --> 00:08:27,400 Speaker 1: a circumscribe role, and we can talk about the specifics, 157 00:08:27,400 --> 00:08:31,520 Speaker 1: but nevertheless its relative weight is more important than certainly 158 00:08:31,520 --> 00:08:34,480 Speaker 1: a decade or two ago. And then third, China's trying 159 00:08:34,480 --> 00:08:38,040 Speaker 1: to catch up technologically in advanced semiconductors and through a 160 00:08:38,080 --> 00:08:40,800 Speaker 1: number of industrial policy programs, the most famous of one 161 00:08:40,800 --> 00:08:44,319 Speaker 1: which is probably made in China. It's specifically targeting its 162 00:08:44,360 --> 00:08:47,560 Speaker 1: hip industry with really substantial subsidies, and and so this 163 00:08:47,600 --> 00:08:50,880 Speaker 1: has raised concern not only Knew US, but particularly then 164 00:08:51,040 --> 00:08:54,199 Speaker 1: US that China might succeed in reaching the cutting edge 165 00:08:54,200 --> 00:08:56,959 Speaker 1: and a number of key technologies which would have both 166 00:08:57,200 --> 00:09:00,920 Speaker 1: technological and economic but also potentially uh military implications on 167 00:09:01,000 --> 00:09:04,280 Speaker 1: the road. Right. So this is actually something we were 168 00:09:04,320 --> 00:09:09,040 Speaker 1: speaking with Dan Wong about, the the Gavicale tech analyst 169 00:09:09,280 --> 00:09:12,200 Speaker 1: who also wrote a little blurb for your book. So 170 00:09:12,480 --> 00:09:15,840 Speaker 1: it's nice to see multiple odds guests come together in 171 00:09:15,880 --> 00:09:19,480 Speaker 1: that way. But can I ask, like just a very 172 00:09:19,600 --> 00:09:23,360 Speaker 1: general step back question. I think it's become really common 173 00:09:23,400 --> 00:09:27,360 Speaker 1: nowadays to say that semiconductors are the new oil in 174 00:09:27,400 --> 00:09:32,120 Speaker 1: the sense that they're you know, military lee and strategically significant. 175 00:09:32,880 --> 00:09:35,800 Speaker 1: But are they really like oil in the sense that, 176 00:09:36,520 --> 00:09:41,040 Speaker 1: you know, they don't necessarily seem bounded by geography. I mean, 177 00:09:41,080 --> 00:09:43,240 Speaker 1: it's not like we're trying to pull out a finite 178 00:09:43,240 --> 00:09:46,479 Speaker 1: resource from the ground. We're trying to build up expertise 179 00:09:46,880 --> 00:09:52,040 Speaker 1: and technological capability to make additional chips available. And you know, 180 00:09:52,080 --> 00:09:55,400 Speaker 1: there are some fabs in the US, not a lot, 181 00:09:55,480 --> 00:09:57,720 Speaker 1: but the Chips Act is all about building more of 182 00:09:57,760 --> 00:10:01,360 Speaker 1: these chips factories so I guess what are the parallels 183 00:10:01,400 --> 00:10:04,760 Speaker 1: with oil and what are the key differences. Yeah, I 184 00:10:04,760 --> 00:10:08,280 Speaker 1: think you're you're right that there are obviously differences with oil. 185 00:10:08,720 --> 00:10:13,800 Speaker 1: The geographic distribution is one difference though, Actually I think 186 00:10:14,360 --> 00:10:17,880 Speaker 1: oil is more geographically distributed than chip making capabilities, especially 187 00:10:17,960 --> 00:10:19,840 Speaker 1: at the leading edge. And if you look at Saudi 188 00:10:19,880 --> 00:10:22,960 Speaker 1: Arabia produces ten or fifteen percent of the world's oil supply. 189 00:10:23,280 --> 00:10:26,240 Speaker 1: By contrast, Taiwan produces over a third of the world's 190 00:10:26,280 --> 00:10:30,000 Speaker 1: processor chips every year. And if you're talking about the 191 00:10:30,000 --> 00:10:33,480 Speaker 1: most advanced processor chips or SML has got a far 192 00:10:33,520 --> 00:10:37,400 Speaker 1: bigger role in the lithography production process percent of the 193 00:10:38,000 --> 00:10:41,079 Speaker 1: leading edge of UV lithography compared to what everyone else 194 00:10:41,120 --> 00:10:43,520 Speaker 1: has in the oil industry. So in terms of concentration, 195 00:10:43,559 --> 00:10:46,960 Speaker 1: we actually have more concentration at multiple parts of the 196 00:10:47,000 --> 00:10:51,040 Speaker 1: chip supply chain than than Saudi Arabia or even OPEC 197 00:10:51,160 --> 00:10:53,960 Speaker 1: as a whole has. So if you're worried about the 198 00:10:54,040 --> 00:10:57,440 Speaker 1: risks of supply cutoffs, there's more reason to be concerned 199 00:10:57,440 --> 00:10:59,960 Speaker 1: about some my conductors. I think that concerned about oil. 200 00:11:16,720 --> 00:11:18,920 Speaker 1: So this is something that you touch on in your book. 201 00:11:19,120 --> 00:11:23,240 Speaker 1: But when COVID happened. There were plenty of businesses that 202 00:11:23,320 --> 00:11:27,640 Speaker 1: were disrupted by the pandemic, and semiconductors became a big 203 00:11:27,679 --> 00:11:31,760 Speaker 1: issue around that time. And we can argue about whether 204 00:11:31,880 --> 00:11:35,680 Speaker 1: or not there were semiconductor shortages because of the chips 205 00:11:35,720 --> 00:11:40,760 Speaker 1: makers themselves, or because the companies putting in orders massively 206 00:11:40,880 --> 00:11:44,240 Speaker 1: underestimated demand and things like that. But what role did 207 00:11:44,280 --> 00:11:48,760 Speaker 1: the pandemic play in I guess highlighting these shortage issues 208 00:11:48,800 --> 00:11:52,280 Speaker 1: that you just described, or maybe shortage isn't the right word, 209 00:11:52,400 --> 00:11:56,959 Speaker 1: it's it's uneven distribution. I guess yeah, I think that's 210 00:11:57,000 --> 00:11:59,120 Speaker 1: the right way to describe it. And you know, the 211 00:11:59,120 --> 00:12:02,720 Speaker 1: shortage context has to be understood in that supply increased 212 00:12:02,800 --> 00:12:07,120 Speaker 1: dramatically in terms of aggregate chips produced in one It's 213 00:12:07,120 --> 00:12:09,760 Speaker 1: just that demand grew even faster and that created this 214 00:12:09,920 --> 00:12:13,400 Speaker 1: set of shortages in certain subsets. I think if you 215 00:12:13,440 --> 00:12:17,120 Speaker 1: actually look at the track record in both of those 216 00:12:17,160 --> 00:12:21,160 Speaker 1: pandemic years of the global chip industry and producing chips, 217 00:12:21,200 --> 00:12:23,120 Speaker 1: you'd have to say it it's sailed through the crisis 218 00:12:23,160 --> 00:12:27,000 Speaker 1: with comparatively few disruptions. The fact that supply increased dramatically 219 00:12:27,120 --> 00:12:29,920 Speaker 1: across the board, the fact that you know, we had 220 00:12:30,080 --> 00:12:33,720 Speaker 1: major economies that played a big part in the chip 221 00:12:33,800 --> 00:12:37,320 Speaker 1: supply chain being shut down for several months, if not 222 00:12:37,440 --> 00:12:40,760 Speaker 1: longer across COVID, and yet the number of disruptions were 223 00:12:40,760 --> 00:12:43,880 Speaker 1: actually quite limited, suggests the extent that which the supply 224 00:12:43,920 --> 00:12:46,520 Speaker 1: chain actually works pretty well and in producing the chips 225 00:12:46,520 --> 00:12:49,320 Speaker 1: that are needed. And when you look at where we've 226 00:12:49,360 --> 00:12:52,640 Speaker 1: seen the most disruption in terms of shortages, which is 227 00:12:52,640 --> 00:12:54,920 Speaker 1: the auto sector, I think there's a pretty good argument 228 00:12:54,960 --> 00:12:57,200 Speaker 1: that what we're seeing is not a chip shortage but 229 00:12:57,280 --> 00:13:00,280 Speaker 1: a automaker inability to properly plan the number of hips 230 00:13:00,280 --> 00:13:02,680 Speaker 1: that they need, which is the problem that they faced. 231 00:13:02,880 --> 00:13:07,079 Speaker 1: Automakers made an error in early slashing their chips sales 232 00:13:07,400 --> 00:13:11,199 Speaker 1: just as demand for smartphones and PCs, etcetera was spooming 233 00:13:11,360 --> 00:13:14,000 Speaker 1: because of the work from home trend, and as a result, 234 00:13:14,040 --> 00:13:16,800 Speaker 1: when they came back to their chip suppliers a couple 235 00:13:16,800 --> 00:13:18,960 Speaker 1: months later and said, actually, we're not gonna need to 236 00:13:19,000 --> 00:13:21,200 Speaker 1: cut demand as much as we expected, they found that 237 00:13:21,240 --> 00:13:24,319 Speaker 1: capacity was already reallocated. So this is really more a 238 00:13:24,400 --> 00:13:27,320 Speaker 1: question of how automakers managed their procurement and they're just 239 00:13:27,360 --> 00:13:30,440 Speaker 1: in time procurement processes than anything to do with chips 240 00:13:30,520 --> 00:13:33,679 Speaker 1: per se, and if you look at specific disruptions to 241 00:13:33,760 --> 00:13:38,200 Speaker 1: the shipmaking process one you'll find that factory fires or 242 00:13:38,240 --> 00:13:41,440 Speaker 1: ice storms and Texas were probably just as impactful as 243 00:13:41,440 --> 00:13:44,120 Speaker 1: any sort of COVID related disruption. But you know, all 244 00:13:44,120 --> 00:13:47,439 Speaker 1: that is to say that certainly COVID put semi conductor 245 00:13:47,520 --> 00:13:51,120 Speaker 1: shortages in the headlines in a way that catalyzed politicians 246 00:13:51,120 --> 00:13:53,120 Speaker 1: to take the issue more seriously into act. And I 247 00:13:53,120 --> 00:13:55,520 Speaker 1: don't think we would have seen the Chips and Science 248 00:13:55,559 --> 00:13:57,720 Speaker 1: Act passed and the way that it did had it 249 00:13:57,800 --> 00:14:00,240 Speaker 1: not been for the fact that COVID made this and 250 00:14:00,480 --> 00:14:03,000 Speaker 1: visceral and seeming to be relevant to the typical person. 251 00:14:03,280 --> 00:14:05,680 Speaker 1: By the way, speaking of all the people who have 252 00:14:05,760 --> 00:14:10,240 Speaker 1: glowingly blurbed your book, I see Admiral stephrevious. Also is 253 00:14:10,280 --> 00:14:12,760 Speaker 1: one here another odd Lots guests, So I'm glad to 254 00:14:12,760 --> 00:14:16,199 Speaker 1: see the odd Lots crew so represented. All right, let's 255 00:14:16,200 --> 00:14:18,280 Speaker 1: get to the new news. And of course, as we 256 00:14:18,360 --> 00:14:21,480 Speaker 1: discussed there, there were already restrictions on what kind of 257 00:14:21,480 --> 00:14:24,560 Speaker 1: technology could be exported to China. But why don't you 258 00:14:24,600 --> 00:14:28,640 Speaker 1: sort of summarize what was the status quo and say September, 259 00:14:28,720 --> 00:14:32,200 Speaker 1: where were we a month ago, and how dramatically have 260 00:14:32,360 --> 00:14:35,800 Speaker 1: things changed with the new announcements from the Commerce Department 261 00:14:35,880 --> 00:14:39,120 Speaker 1: and the Biden administration on exports, like just summarize, give 262 00:14:39,160 --> 00:14:42,200 Speaker 1: us the big picture differences of them versus now. So 263 00:14:42,480 --> 00:14:45,160 Speaker 1: one month ago, there were i'd say two main categories 264 00:14:45,200 --> 00:14:49,120 Speaker 1: of restrictions that applied to China. First was for equipment 265 00:14:49,200 --> 00:14:52,560 Speaker 1: at the most cutting edge, so UVY lithography machines being 266 00:14:52,680 --> 00:14:55,800 Speaker 1: the best example of this, that were impossible to transfer 267 00:14:55,960 --> 00:14:58,880 Speaker 1: to China, and it's really only referred to a small 268 00:14:58,920 --> 00:15:03,080 Speaker 1: set of tool roles that only restricted Chinese ability to 269 00:15:03,760 --> 00:15:07,320 Speaker 1: move to the most advanced process nodes in producing logic 270 00:15:07,320 --> 00:15:09,480 Speaker 1: ships and certain types of memory ships as well. But 271 00:15:09,520 --> 00:15:12,000 Speaker 1: it was the most cutting edge set of machinery that 272 00:15:12,040 --> 00:15:14,640 Speaker 1: was the first restrictions. And then second there restrictions on 273 00:15:14,680 --> 00:15:19,160 Speaker 1: specific Chinese firms that were restricted either from accessing US 274 00:15:19,200 --> 00:15:22,680 Speaker 1: technology or also in some cases accessing production capacity in 275 00:15:22,680 --> 00:15:25,640 Speaker 1: Taiwan at t SMC that use US technology, and so 276 00:15:25,720 --> 00:15:27,520 Speaker 1: Huawei is the best example of this, but there were 277 00:15:27,560 --> 00:15:29,840 Speaker 1: also a bunch of Chinese firms that were assessed to 278 00:15:29,880 --> 00:15:31,840 Speaker 1: have links to the Chinese military, and so these were 279 00:15:31,920 --> 00:15:34,400 Speaker 1: the two sets of restrictions that were in place today. 280 00:15:34,440 --> 00:15:37,840 Speaker 1: After the new set of regulations, these have been broadened 281 00:15:37,880 --> 00:15:41,240 Speaker 1: pretty substantially. And so, first off, the the set of 282 00:15:41,280 --> 00:15:44,400 Speaker 1: equipment that is no longer possible to ship to China 283 00:15:44,600 --> 00:15:46,840 Speaker 1: is expanded a lot, so it's no longer just the 284 00:15:46,880 --> 00:15:50,520 Speaker 1: most cutting edge, is also tools that are actually kind 285 00:15:50,560 --> 00:15:52,920 Speaker 1: of one or two generations, with three generations behind the 286 00:15:52,920 --> 00:15:55,920 Speaker 1: cutting edge today. So in the logic chip making space, 287 00:15:56,000 --> 00:16:00,000 Speaker 1: anything that's headed towards the facility that's sixteen nanometers or 288 00:16:00,040 --> 00:16:03,040 Speaker 1: blow is no longer possible to send to China. So 289 00:16:03,040 --> 00:16:05,200 Speaker 1: so this is not the most cutting edge ship like 290 00:16:05,240 --> 00:16:08,160 Speaker 1: you find an iPhone processor, for example, But this is 291 00:16:08,200 --> 00:16:10,360 Speaker 1: a much broader swath of chips, and there's a whole 292 00:16:10,360 --> 00:16:12,360 Speaker 1: lot of chips that you still can send equipment forward 293 00:16:12,400 --> 00:16:14,520 Speaker 1: to China, but there's a broader definition of the cutting 294 00:16:14,600 --> 00:16:18,080 Speaker 1: edge which is designed to more tightly restrict China's ability 295 00:16:18,200 --> 00:16:21,080 Speaker 1: to advance. So that's that's one big change. The second 296 00:16:21,160 --> 00:16:23,760 Speaker 1: is that there's a new set of restrictions on the 297 00:16:23,800 --> 00:16:27,240 Speaker 1: ability to send certain chips to China, and this focuses 298 00:16:27,320 --> 00:16:30,600 Speaker 1: on chips called GPUs, which are used in artificial intelligence 299 00:16:30,600 --> 00:16:34,080 Speaker 1: applications and data centers. And alongside this there are restrictions 300 00:16:34,080 --> 00:16:37,440 Speaker 1: on sending chips to Chinese supercomputers. So this set of 301 00:16:37,480 --> 00:16:41,200 Speaker 1: the restrictions focuses on limiting China's ability to develop capabilities 302 00:16:41,200 --> 00:16:43,840 Speaker 1: and artificial intelligence and advanced data centers, so that this 303 00:16:43,880 --> 00:16:47,200 Speaker 1: is a new focus area. And then three, there is 304 00:16:47,240 --> 00:16:51,520 Speaker 1: a set of restrictions on US persons, so citizens, Green 305 00:16:51,560 --> 00:16:54,520 Speaker 1: card holders and US companies on interacting with a bunch 306 00:16:54,560 --> 00:16:56,760 Speaker 1: of different firms in the Chinese ship industry. And this 307 00:16:56,840 --> 00:17:00,320 Speaker 1: is new. Before there weren't any restrictions on individuals in 308 00:17:00,440 --> 00:17:02,800 Speaker 1: such a broad fashion working in the Chinese ship industry. 309 00:17:02,800 --> 00:17:04,399 Speaker 1: And this is going to have major effects because there 310 00:17:04,400 --> 00:17:06,200 Speaker 1: are actually a lot of dual citizens who are working 311 00:17:06,200 --> 00:17:08,560 Speaker 1: in Chinese ship industry, including CEO, so some of the 312 00:17:08,600 --> 00:17:11,040 Speaker 1: biggest firms in China. And so this this is a 313 00:17:11,080 --> 00:17:14,720 Speaker 1: new escalation of the controls designed again to take apart 314 00:17:14,800 --> 00:17:17,240 Speaker 1: to decouple the U s and the Chinese ship industry 315 00:17:17,240 --> 00:17:18,760 Speaker 1: at the cutting edge. But all of this is to 316 00:17:18,800 --> 00:17:21,320 Speaker 1: say that there's still no restrictions on anything that's not 317 00:17:21,359 --> 00:17:23,520 Speaker 1: cutting edge for the most part, and so the goal 318 00:17:23,600 --> 00:17:27,000 Speaker 1: of the regulations is to be fairly targeted in saying 319 00:17:27,280 --> 00:17:29,880 Speaker 1: we're not going to allow any progress in the cutting edge, 320 00:17:29,880 --> 00:17:33,440 Speaker 1: but we're still leading all of the lighting edge, mostly unrestricted, 321 00:17:33,480 --> 00:17:36,000 Speaker 1: so it's it's still targeted, but it's pretty sweeping in 322 00:17:36,000 --> 00:17:39,920 Speaker 1: the sense that it's saying no more progress towards advanced notes. 323 00:17:41,040 --> 00:17:45,280 Speaker 1: So that third point about Americans or Green card holders 324 00:17:45,320 --> 00:17:48,240 Speaker 1: no longer being able to do that kind of work 325 00:17:48,840 --> 00:17:52,080 Speaker 1: in China, That's exactly what I wanted to ask you about, 326 00:17:52,160 --> 00:17:54,840 Speaker 1: because this seems to be so far one of the 327 00:17:54,880 --> 00:17:56,840 Speaker 1: things that is getting a lot of attention, one of 328 00:17:56,840 --> 00:18:00,680 Speaker 1: the most extreme things. And we've already seen at least 329 00:18:00,840 --> 00:18:04,760 Speaker 1: one semiconductor company talk about how it's having to like 330 00:18:05,119 --> 00:18:08,200 Speaker 1: send people home or at least cut off their access 331 00:18:08,280 --> 00:18:12,359 Speaker 1: at work. But first of all, how exactly does this 332 00:18:12,480 --> 00:18:17,080 Speaker 1: restriction work in practice? And then secondly, I have seen 333 00:18:17,160 --> 00:18:20,919 Speaker 1: some people talk about an expectation that maybe some of 334 00:18:20,920 --> 00:18:24,119 Speaker 1: these companies are going to get some type of exemption. 335 00:18:24,840 --> 00:18:28,879 Speaker 1: What's the possibility that that happens, and that some of 336 00:18:28,920 --> 00:18:31,919 Speaker 1: the extreme moves that have been outlined in these export 337 00:18:31,960 --> 00:18:36,840 Speaker 1: controls are actually not necessarily rolled back, but they're sort 338 00:18:36,880 --> 00:18:41,000 Speaker 1: of softened through these types of exemptions. Well, the Commerce 339 00:18:41,040 --> 00:18:44,720 Speaker 1: Department in their announcement did clarify where they were likely 340 00:18:44,760 --> 00:18:47,520 Speaker 1: to give exemptions and not likely to give exemptions, and 341 00:18:47,560 --> 00:18:50,440 Speaker 1: so they were pretty clear that for foreign chip firms 342 00:18:50,480 --> 00:18:53,280 Speaker 1: operating in China, and again there's it depends on the specifics, 343 00:18:53,320 --> 00:18:56,399 Speaker 1: but there's their scope for receiving exemptions. So companies like 344 00:18:56,520 --> 00:18:59,399 Speaker 1: t SMC or some of the big South koreanship makers 345 00:18:59,400 --> 00:19:02,000 Speaker 1: have facilities in China and the goal of these restrictions 346 00:19:02,040 --> 00:19:05,119 Speaker 1: is not to impact their production, so there will be 347 00:19:05,119 --> 00:19:08,120 Speaker 1: exemptions for their ability to acquire certain types of machinery 348 00:19:08,160 --> 00:19:11,000 Speaker 1: things like that. But across most the other restrict restrictions, 349 00:19:11,000 --> 00:19:13,359 Speaker 1: the Commerce Department has what's called a presumption of denial, 350 00:19:13,440 --> 00:19:15,159 Speaker 1: which means that you can apply for license, but the 351 00:19:15,160 --> 00:19:17,240 Speaker 1: presumption is that it will be denied. And again there's 352 00:19:17,520 --> 00:19:19,960 Speaker 1: a lot of nuance in this, but I think broadly 353 00:19:20,000 --> 00:19:22,480 Speaker 1: we should treat this as as something that's not a 354 00:19:22,480 --> 00:19:24,639 Speaker 1: bureautic kick up, and that in fact, most of the 355 00:19:24,640 --> 00:19:27,159 Speaker 1: places where the Commerce Department has said there will be 356 00:19:27,200 --> 00:19:28,880 Speaker 1: a presumption of denial will be a place with there's 357 00:19:28,920 --> 00:19:31,240 Speaker 1: lots of licenses that are denied, and probably people won't 358 00:19:31,280 --> 00:19:34,040 Speaker 1: apply for licenses in the first place. I I think 359 00:19:34,080 --> 00:19:37,119 Speaker 1: we shouldn't. We shouldn't assume that these are regulations that 360 00:19:37,160 --> 00:19:40,320 Speaker 1: are accidentally overly broad. I think the by administration put 361 00:19:40,320 --> 00:19:42,080 Speaker 1: two years of work into this, and I think they 362 00:19:42,119 --> 00:19:44,399 Speaker 1: have a pretty clear sense of what they're trying to accomplish. 363 00:19:44,960 --> 00:19:48,199 Speaker 1: And my guess is that looking at the impact in 364 00:19:48,240 --> 00:19:50,639 Speaker 1: the first couple of days after the regulations were now, 365 00:19:50,680 --> 00:19:54,960 Speaker 1: it's there more or less seeing results that they had anticipated. 366 00:19:55,680 --> 00:19:57,760 Speaker 1: So what's it going to do the sort of in 367 00:19:58,080 --> 00:20:01,000 Speaker 1: the first order effect, I mean, long term, maybe China 368 00:20:01,000 --> 00:20:05,359 Speaker 1: will redouble or accelerated effort at building out its domestic capabilities, 369 00:20:05,560 --> 00:20:08,360 Speaker 1: But what is the effect going to be tomorrow and 370 00:20:08,440 --> 00:20:11,080 Speaker 1: in the short term on Chinese tech? The short term 371 00:20:11,119 --> 00:20:12,439 Speaker 1: effect is that China is going to buy a lot 372 00:20:12,520 --> 00:20:17,240 Speaker 1: less equipment from US suppliers, but also as a result 373 00:20:17,320 --> 00:20:21,320 Speaker 1: from Japanese and Dutch suppliers, just because there's no point 374 00:20:21,359 --> 00:20:24,000 Speaker 1: in buying half a production line. So the overall equipment 375 00:20:24,000 --> 00:20:26,680 Speaker 1: sales will be decreasing. Again, it's not across the board, 376 00:20:26,720 --> 00:20:29,800 Speaker 1: because for lower tech production that's still possible to acquire, 377 00:20:29,880 --> 00:20:32,320 Speaker 1: but but there will be an impact on equipment firms, 378 00:20:32,320 --> 00:20:34,760 Speaker 1: and you've seen that reflected in their public statements. That's 379 00:20:34,800 --> 00:20:37,960 Speaker 1: one to There are a number of Chinese firms that 380 00:20:38,040 --> 00:20:40,720 Speaker 1: were close to the cutting edge that will be directly 381 00:20:40,720 --> 00:20:44,159 Speaker 1: impacted by this. So one example is the Chinese nand 382 00:20:44,280 --> 00:20:47,320 Speaker 1: memory producer called y MTC, which was in the process 383 00:20:47,359 --> 00:20:49,840 Speaker 1: of being of having its chips qualified for use in 384 00:20:49,880 --> 00:20:53,440 Speaker 1: Apple's iPhones. Apple was reportedly going to buy y MTC 385 00:20:53,640 --> 00:20:56,520 Speaker 1: ships for a large chunk of its iPhones going forward, 386 00:20:56,960 --> 00:20:59,720 Speaker 1: and that firm was targeted by these regulations, and so 387 00:20:59,800 --> 00:21:02,159 Speaker 1: now according to media reports, Apple is no longer going 388 00:21:02,200 --> 00:21:04,840 Speaker 1: to be buying y YMTC ships for its iPhones. And 389 00:21:04,880 --> 00:21:06,879 Speaker 1: so for firms that are are directly impacted like that, 390 00:21:06,920 --> 00:21:09,000 Speaker 1: because they're right at the cusp of being at the 391 00:21:09,000 --> 00:21:12,200 Speaker 1: cutting edge and therefore directly impacted by these regulations, we're 392 00:21:12,200 --> 00:21:15,880 Speaker 1: going to see their ability to produce and their ability 393 00:21:15,960 --> 00:21:20,200 Speaker 1: to expand directly impacted. And then third, we're going to 394 00:21:20,320 --> 00:21:25,080 Speaker 1: see disruption of of individuals, jobs and livelihoods because of this. 395 00:21:25,359 --> 00:21:27,960 Speaker 1: Tracy mentioned the restrictions and individuals. That's going to have 396 00:21:28,040 --> 00:21:30,560 Speaker 1: a pretty dramatic effect because there will be people in 397 00:21:30,600 --> 00:21:33,760 Speaker 1: positions of having to choose between continuing their work or 398 00:21:33,840 --> 00:21:37,560 Speaker 1: potentially renouncing cititionship or green cards, and that's a that's 399 00:21:37,560 --> 00:21:40,320 Speaker 1: a dramatic effect. Now, you know, this isn't isn't new, 400 00:21:40,359 --> 00:21:42,640 Speaker 1: because there's lots of companies around the world that Americans 401 00:21:42,640 --> 00:21:44,439 Speaker 1: are barred from doing business with. But I don't think 402 00:21:44,480 --> 00:21:47,280 Speaker 1: many people in who are impacted by this. We're expecting 403 00:21:47,320 --> 00:21:50,000 Speaker 1: this type of regulation, and this is probably the part 404 00:21:50,040 --> 00:22:06,679 Speaker 1: of the rules that have most surprised people in the industry. 405 00:22:09,040 --> 00:22:11,560 Speaker 1: You know, I asked Dan Wong the same question, but 406 00:22:11,600 --> 00:22:14,199 Speaker 1: I would love to hear your response as well. But 407 00:22:14,640 --> 00:22:18,360 Speaker 1: what kind of reaction would you expect from China here? 408 00:22:18,480 --> 00:22:22,600 Speaker 1: So we've already had some statements from various officials talking 409 00:22:22,640 --> 00:22:26,280 Speaker 1: about how it's going to endanger China's economic recovery and 410 00:22:26,320 --> 00:22:30,000 Speaker 1: how it's generally a very bad thing for the economy. 411 00:22:30,160 --> 00:22:35,040 Speaker 1: But would you expect any concrete measures from China after 412 00:22:35,240 --> 00:22:38,399 Speaker 1: these restrictions? Yeah, it's an interesting question because on the 413 00:22:38,440 --> 00:22:41,399 Speaker 1: one hand, you know, I think you'd hypothesize that China 414 00:22:41,440 --> 00:22:43,560 Speaker 1: has to do something in retaliation. On the other hand, 415 00:22:43,600 --> 00:22:46,760 Speaker 1: if you look at the track record of China's responses 416 00:22:46,840 --> 00:22:52,080 Speaker 1: to prior iterations of semiconductor export controls, China hasn't done 417 00:22:52,160 --> 00:22:55,480 Speaker 1: much when the US put Huawei on the entity list 418 00:22:55,520 --> 00:22:58,359 Speaker 1: and thereby restricted the transfer of a lot of US 419 00:22:58,480 --> 00:23:01,240 Speaker 1: chips and chip technologies to wah Way. China issued some 420 00:23:01,320 --> 00:23:04,280 Speaker 1: angry statements, but ended up doing nothing, despite the fact 421 00:23:04,359 --> 00:23:07,320 Speaker 1: that these restrictions hobbled one of China's biggest tech firms, 422 00:23:07,880 --> 00:23:11,760 Speaker 1: which was I think a pretty striking demonstration of of 423 00:23:11,800 --> 00:23:15,640 Speaker 1: where escalation dominance in this sphere. If you will, stands 424 00:23:15,680 --> 00:23:18,320 Speaker 1: that China felt like actually any sort of retaliation would 425 00:23:18,359 --> 00:23:21,359 Speaker 1: leave it worse off than not retaliating. Now, this is 426 00:23:21,400 --> 00:23:25,280 Speaker 1: a an order of magnitude more impactful than the Huawei restrictions, 427 00:23:25,280 --> 00:23:27,680 Speaker 1: although it's in some ways a bit more opaque since 428 00:23:27,680 --> 00:23:30,080 Speaker 1: it doesn't target a single firm and sort of hidden 429 00:23:30,080 --> 00:23:32,600 Speaker 1: in the specifics of Commerce Department regulations with which most 430 00:23:32,600 --> 00:23:34,840 Speaker 1: people don't understand. But I would think China would feel 431 00:23:34,880 --> 00:23:37,840 Speaker 1: more pressure to retaliate this time. But if you ask yourself, 432 00:23:37,880 --> 00:23:40,119 Speaker 1: what's the best way for China to retaliate in a 433 00:23:40,160 --> 00:23:42,480 Speaker 1: way that imposes more costs on the US than it 434 00:23:42,520 --> 00:23:45,280 Speaker 1: does in China, it's not obvious that there's a great 435 00:23:45,520 --> 00:23:49,639 Speaker 1: option there. In China still deeply dependent on importing ships 436 00:23:49,800 --> 00:23:52,479 Speaker 1: across the board, and China's economy, of course, is as 437 00:23:52,520 --> 00:23:55,240 Speaker 1: weak as it's been in many years, so this isn't 438 00:23:55,240 --> 00:23:58,680 Speaker 1: a time when j Ping wants to probably impose further 439 00:23:58,760 --> 00:24:01,320 Speaker 1: cost on the Chinese economy. And in particular, if you're 440 00:24:01,359 --> 00:24:04,159 Speaker 1: a Chinese electronics firm or from in the semi connector space, 441 00:24:04,560 --> 00:24:06,960 Speaker 1: you're desperate right now to keep the connections that you 442 00:24:07,000 --> 00:24:09,240 Speaker 1: do have open, because if there's any way you're gonna 443 00:24:09,240 --> 00:24:11,360 Speaker 1: get around the current regulations, it's going to be by 444 00:24:11,480 --> 00:24:15,760 Speaker 1: whatever collaboration you have still functioning with leading firms in 445 00:24:15,800 --> 00:24:18,879 Speaker 1: other countries, and so keeping those those lines of communication 446 00:24:18,920 --> 00:24:23,080 Speaker 1: and cooperation open with from other countries is really crucial, 447 00:24:23,080 --> 00:24:26,720 Speaker 1: I think for China's hopes of finding some way to 448 00:24:26,760 --> 00:24:29,720 Speaker 1: counter these rules. So I'm not sure actually if China's 449 00:24:29,720 --> 00:24:33,040 Speaker 1: going to take a dramatic retaliatory step, even though I 450 00:24:33,040 --> 00:24:36,439 Speaker 1: think there's obviously a political impetus in China to do 451 00:24:36,560 --> 00:24:38,560 Speaker 1: something and to look like you're not just taking these 452 00:24:39,000 --> 00:24:41,720 Speaker 1: restrictions lying down. So you know, it's easy enough, and 453 00:24:41,760 --> 00:24:44,760 Speaker 1: I kind of did, maybe lazily enough, but it's easy 454 00:24:44,880 --> 00:24:47,120 Speaker 1: enough to say, Okay, well, this is going to hurt 455 00:24:47,359 --> 00:24:49,200 Speaker 1: China in the short term, but in the long term, 456 00:24:49,200 --> 00:24:52,520 Speaker 1: they're only going to you know, accelerate their own work 457 00:24:52,560 --> 00:24:56,600 Speaker 1: on building out an advanced domestic semiconductor industry, which they might, 458 00:24:57,080 --> 00:25:00,439 Speaker 1: but like that seems still really hard, even if you 459 00:25:00,480 --> 00:25:02,960 Speaker 1: have the entire sort of like force and money of 460 00:25:03,000 --> 00:25:05,560 Speaker 1: the government. And I'm just thinking back to when we 461 00:25:05,640 --> 00:25:08,760 Speaker 1: talked before, and the one one of the details that 462 00:25:08,800 --> 00:25:12,840 Speaker 1: I remember from our conversation was like a SML itself 463 00:25:13,320 --> 00:25:15,800 Speaker 1: is a customer of a company that makes the smoothest 464 00:25:15,840 --> 00:25:18,159 Speaker 1: substance on Earth, because that's important, I guess, like for 465 00:25:18,240 --> 00:25:21,160 Speaker 1: the mirrors to like shoot the light or whatever. And 466 00:25:21,280 --> 00:25:25,280 Speaker 1: you know, it does not seem inevitable that China can 467 00:25:25,320 --> 00:25:27,080 Speaker 1: catch up. I mean, we know, like the challenge that 468 00:25:27,160 --> 00:25:30,800 Speaker 1: Intel has with catching up to Taiwan, Semi, etcetera. And 469 00:25:30,880 --> 00:25:34,680 Speaker 1: without this sort of like best expertise of this very 470 00:25:34,760 --> 00:25:38,280 Speaker 1: globally integrated supply chain, Like what is it? How do 471 00:25:38,320 --> 00:25:42,600 Speaker 1: you assess China's chances of becoming a tech leader in 472 00:25:42,640 --> 00:25:45,680 Speaker 1: the space with so many restrictions. Well, I think it's 473 00:25:45,680 --> 00:25:48,800 Speaker 1: going to be really hard. Is catch up possible? I 474 00:25:48,840 --> 00:25:53,160 Speaker 1: think it's certainly possible over some time frame with some 475 00:25:53,280 --> 00:25:55,960 Speaker 1: amount of expenditure, But I think the timeframe is long, 476 00:25:56,000 --> 00:25:58,719 Speaker 1: and the expenditure is going to be large, and so 477 00:25:58,840 --> 00:26:02,240 Speaker 1: China might decide that it's not willing to pay the price. 478 00:26:02,520 --> 00:26:04,959 Speaker 1: If you look just at the this the space if 479 00:26:04,960 --> 00:26:08,480 Speaker 1: some connected manufacturing equipment, which is in some ways the 480 00:26:09,000 --> 00:26:12,880 Speaker 1: space on which all these other restrictions depend, the reality 481 00:26:13,000 --> 00:26:15,359 Speaker 1: is that the number of companies in this space is 482 00:26:15,440 --> 00:26:18,119 Speaker 1: very limited. Most of the key firms have been in 483 00:26:18,160 --> 00:26:20,800 Speaker 1: their market position for several decades and in some cases 484 00:26:20,840 --> 00:26:24,400 Speaker 1: almost half a century, and their supply chains, as you mentioned, 485 00:26:24,680 --> 00:26:28,000 Speaker 1: involved sourcing some of the most complex and heavily engineered 486 00:26:28,080 --> 00:26:30,680 Speaker 1: products ever made, and so it's just going to be 487 00:26:30,880 --> 00:26:33,960 Speaker 1: very difficult to do so. In addition, we're seeing more 488 00:26:34,000 --> 00:26:38,200 Speaker 1: and more the US government targeting China's firms in this space. 489 00:26:38,240 --> 00:26:41,200 Speaker 1: So there's part of the restrictions are that we're rolled 490 00:26:41,200 --> 00:26:45,040 Speaker 1: out recently limit the transfer of component parts to Chinese 491 00:26:45,080 --> 00:26:48,360 Speaker 1: equipment manufacturers, which is again deliberately designed to make their 492 00:26:48,480 --> 00:26:51,520 Speaker 1: process of advancement harder. And if you think about what 493 00:26:51,560 --> 00:26:54,000 Speaker 1: it takes to make a machine tool in this space, 494 00:26:54,040 --> 00:26:55,920 Speaker 1: you know, it's one thing to make a machine tool 495 00:26:56,000 --> 00:26:59,080 Speaker 1: that can do a specific process once or even a 496 00:26:59,080 --> 00:27:02,520 Speaker 1: dozen times or hundred times, but to make a commercially 497 00:27:02,640 --> 00:27:05,119 Speaker 1: viable machine tool in the chip industry, you need a 498 00:27:05,160 --> 00:27:07,520 Speaker 1: machine that not only is able to move materials at 499 00:27:07,520 --> 00:27:09,720 Speaker 1: almost the atomic level, but it also needs to do 500 00:27:09,800 --> 00:27:13,240 Speaker 1: so with you know, basically perfect run time, perfect accuracy, 501 00:27:13,440 --> 00:27:15,840 Speaker 1: and to be cost competitive. And and that's something you 502 00:27:15,840 --> 00:27:20,040 Speaker 1: can only really develop, I think, when you're closely collaborating 503 00:27:20,080 --> 00:27:22,800 Speaker 1: with the firms that are going to be using your equipment. 504 00:27:22,800 --> 00:27:25,119 Speaker 1: And so if you look at how innovation in the 505 00:27:25,640 --> 00:27:27,440 Speaker 1: in the machine part of the machine tool part of 506 00:27:27,480 --> 00:27:29,680 Speaker 1: the chip industry happens right now, it happens in really 507 00:27:29,720 --> 00:27:33,639 Speaker 1: deep collaboration with the customers t smc Intel, Samsung, etcetera. 508 00:27:33,680 --> 00:27:36,320 Speaker 1: Because the tools have to be optimized for those customers. 509 00:27:36,359 --> 00:27:39,040 Speaker 1: And so if you're in China today trying to develop 510 00:27:39,080 --> 00:27:42,040 Speaker 1: your own machine tools, you're working with firms that have 511 00:27:42,400 --> 00:27:44,960 Speaker 1: lower volumes, especially at the cutting edge. They're actually not 512 00:27:45,000 --> 00:27:47,680 Speaker 1: producing the most cutting edge chips. And that's a harder 513 00:27:47,760 --> 00:27:50,199 Speaker 1: environment which to produce machine tools that are going to 514 00:27:50,280 --> 00:27:53,400 Speaker 1: compete at all with the cutting edge you can acquire abroad. 515 00:27:53,480 --> 00:27:55,879 Speaker 1: So if you've got this closed loop innovation system where 516 00:27:55,880 --> 00:27:58,520 Speaker 1: everything is second or third best compared to what you 517 00:27:58,560 --> 00:27:59,879 Speaker 1: can get in Taiwan or what you can get in 518 00:27:59,920 --> 00:28:02,040 Speaker 1: the other lands. It's just a really hard environment to 519 00:28:02,320 --> 00:28:05,800 Speaker 1: catch up in. So this actually leads nicely into something 520 00:28:05,800 --> 00:28:08,240 Speaker 1: I wanted to ask you about, because, of course we've 521 00:28:08,240 --> 00:28:12,159 Speaker 1: been very focused on the US efforts to restrict cutting 522 00:28:12,280 --> 00:28:15,720 Speaker 1: edge chip technology into China. But the other plank of 523 00:28:15,880 --> 00:28:18,840 Speaker 1: U s strategy is to build up its own self 524 00:28:18,880 --> 00:28:22,399 Speaker 1: sufficiency when it comes to semiconductors. And in the book 525 00:28:22,880 --> 00:28:27,160 Speaker 1: you mentioned that there's growing discussion among US officials about 526 00:28:27,160 --> 00:28:32,920 Speaker 1: potentially using export controls on chip design, software and manufacturing 527 00:28:32,920 --> 00:28:36,080 Speaker 1: equipment that comes from the US. So a lot of 528 00:28:36,080 --> 00:28:39,160 Speaker 1: the chips, as we've discussed on this podcast numerous times, 529 00:28:39,200 --> 00:28:42,200 Speaker 1: they're designed in the US and then those designs are 530 00:28:42,200 --> 00:28:45,760 Speaker 1: shipped out to Taiwan to be actually manufactured or whatever. 531 00:28:46,280 --> 00:28:49,520 Speaker 1: And so if you were to impose export controls on 532 00:28:49,600 --> 00:28:54,000 Speaker 1: the designs or the software, that could, in effect, I guess, 533 00:28:54,320 --> 00:28:57,760 Speaker 1: pressure Taiwan to roll out more technology into the US 534 00:28:57,840 --> 00:29:01,000 Speaker 1: and Europe or allow the US and to build up 535 00:29:01,080 --> 00:29:04,560 Speaker 1: their own expertise. Do you think that's the next step 536 00:29:04,680 --> 00:29:07,320 Speaker 1: in these export controls. Is that something that you see 537 00:29:07,360 --> 00:29:09,800 Speaker 1: on the horizon. Well, I'm not aware of it. Having 538 00:29:10,160 --> 00:29:13,440 Speaker 1: been discussed at the government to government level. It is 539 00:29:13,480 --> 00:29:17,160 Speaker 1: something that you can find written about and think tank 540 00:29:17,200 --> 00:29:20,960 Speaker 1: policy papers, and your discussed to some extent in in Washington, 541 00:29:21,120 --> 00:29:23,760 Speaker 1: d C. I don't know if the US government is 542 00:29:23,800 --> 00:29:25,920 Speaker 1: going to go to to that level, and I think 543 00:29:25,960 --> 00:29:28,360 Speaker 1: that would be a pretty dramatic step. You know, It's 544 00:29:28,360 --> 00:29:30,560 Speaker 1: one thing to impose extuport controls on countries that you 545 00:29:30,600 --> 00:29:34,200 Speaker 1: see is primarily competitor. Something else to impose them in 546 00:29:34,320 --> 00:29:37,320 Speaker 1: country countries that you see primarily as a partner. And 547 00:29:37,360 --> 00:29:39,120 Speaker 1: so I think there would be some medicines in the 548 00:29:39,200 --> 00:29:44,320 Speaker 1: US to use that type of coercive strategy visa vi Taiwan. 549 00:29:44,560 --> 00:29:47,440 Speaker 1: But I think it is true that everyone in the 550 00:29:47,480 --> 00:29:50,960 Speaker 1: industry knows that the US does have the ability to 551 00:29:51,000 --> 00:29:54,200 Speaker 1: impose these controls, and indeed, the US already has imposed 552 00:29:54,200 --> 00:29:57,600 Speaker 1: controls on China in a way that forces Taiwan Taiwanese 553 00:29:57,600 --> 00:30:00,120 Speaker 1: firms to take certain steps they wouldn't otherwise taken to 554 00:30:00,200 --> 00:30:03,280 Speaker 1: the Chinese customers. And so the fact that the US 555 00:30:03,400 --> 00:30:07,040 Speaker 1: has this threat I think probably does influence both how 556 00:30:07,120 --> 00:30:10,600 Speaker 1: US government officials and Taiwanese government officials think about the dynamic. 557 00:30:10,720 --> 00:30:13,880 Speaker 1: But if you look right now at at what Taiwan 558 00:30:13,920 --> 00:30:16,680 Speaker 1: these firms, And really that just means t SMC has 559 00:30:16,840 --> 00:30:19,560 Speaker 1: planned in the United States in terms of opening new facilities. 560 00:30:19,600 --> 00:30:22,520 Speaker 1: They're opening one facility in Arizona. It will be relatively 561 00:30:22,560 --> 00:30:25,000 Speaker 1: small in terms of its capacity, and most of t 562 00:30:25,200 --> 00:30:28,160 Speaker 1: s MC's capital expenditure going forward is still going to 563 00:30:28,200 --> 00:30:32,320 Speaker 1: take place in Taiwan. Even if you include the Arizona facility, 564 00:30:32,400 --> 00:30:35,600 Speaker 1: the new facility in Japan, potential facilities and Singapore and 565 00:30:35,640 --> 00:30:37,680 Speaker 1: Europe that are being discussed, we're still talking about a 566 00:30:37,680 --> 00:30:40,840 Speaker 1: company that's primarily focused in Taiwan, especially when it comes 567 00:30:40,840 --> 00:30:43,720 Speaker 1: to its most advanced production nodes. So right now, I 568 00:30:43,760 --> 00:30:48,160 Speaker 1: don't think t SMC is planning to dramatically diverseify away 569 00:30:48,160 --> 00:30:51,200 Speaker 1: from Taiwan in nor do I think it feels really 570 00:30:51,200 --> 00:30:53,320 Speaker 1: substantial pressure to do so. A little bit of pressure, 571 00:30:53,360 --> 00:30:55,360 Speaker 1: but not not substantial un of pressure to make it 572 00:30:55,440 --> 00:30:58,160 Speaker 1: change its business model. By the way, when you were 573 00:30:58,200 --> 00:31:01,120 Speaker 1: talking about, you know, the jail lenge of getting the 574 00:31:01,240 --> 00:31:04,040 Speaker 1: equipment to run perfectly each time, I remember our episode 575 00:31:04,040 --> 00:31:07,120 Speaker 1: we talked to Willy she at HBS and like the 576 00:31:07,160 --> 00:31:12,960 Speaker 1: math is just mind boggling, Like nine point correct execution 577 00:31:13,080 --> 00:31:15,720 Speaker 1: is like not anywhere near sufficient for how like many 578 00:31:15,760 --> 00:31:19,080 Speaker 1: steps and how right everything has to go. But I 579 00:31:19,120 --> 00:31:21,920 Speaker 1: want to pivot to the Chips Act. And you know, 580 00:31:22,040 --> 00:31:25,640 Speaker 1: Tracy mentioned that, yes, part of strategy right now is 581 00:31:25,680 --> 00:31:29,360 Speaker 1: constraining China's ambitions. But also there's been you know, legislation 582 00:31:29,480 --> 00:31:32,040 Speaker 1: that move forward in the US, A laws passed designed 583 00:31:32,040 --> 00:31:35,640 Speaker 1: to bolster US capacity. What do you see, like, how 584 00:31:35,680 --> 00:31:38,960 Speaker 1: would you ascribe just sort of big picture the significance 585 00:31:39,080 --> 00:31:40,880 Speaker 1: of that act and how much do you think that's 586 00:31:40,920 --> 00:31:43,160 Speaker 1: going to move the dial in terms of sort of 587 00:31:43,200 --> 00:31:47,120 Speaker 1: bringing setting t SMC aside bringing on shore to the 588 00:31:47,200 --> 00:31:50,520 Speaker 1: US more more capacity. So I think the Chips Act 589 00:31:50,560 --> 00:31:53,560 Speaker 1: is the biggest move in semi connected policy since the 590 00:31:53,680 --> 00:31:55,760 Speaker 1: nineteen eighties that we've seen from the US government. In 591 00:31:55,800 --> 00:31:58,600 Speaker 1: the there was a trade war with Japan, which was 592 00:31:58,640 --> 00:32:00,480 Speaker 1: at the time a big chip maker, and the U 593 00:32:00,600 --> 00:32:02,720 Speaker 1: S and POST bunch of threaten to impose a bunch 594 00:32:02,760 --> 00:32:05,440 Speaker 1: of tariffs on on Japanese ship imports until the Japanese 595 00:32:05,440 --> 00:32:08,120 Speaker 1: agreed to limit their semi connector exports and that was 596 00:32:08,160 --> 00:32:10,520 Speaker 1: a big focus of policy at the time, and then 597 00:32:10,560 --> 00:32:14,719 Speaker 1: for about four decades, at least three decades, US policymakers 598 00:32:14,760 --> 00:32:17,080 Speaker 1: more or less ignored some my conductors as a as 599 00:32:17,120 --> 00:32:19,800 Speaker 1: a policy issue, in part because U S firms appeared 600 00:32:19,800 --> 00:32:22,840 Speaker 1: to dominate them, and that's obviously changed. The Chip and 601 00:32:22,880 --> 00:32:26,360 Speaker 1: Science Act will do two main things. First, it allocates 602 00:32:26,360 --> 00:32:30,920 Speaker 1: thirty nine billion dollars to subsidizing the manufacture of chips 603 00:32:30,960 --> 00:32:34,840 Speaker 1: in the US, mostly focused on leading edge ships, but 604 00:32:34,960 --> 00:32:37,480 Speaker 1: with some funds for lagging edge and then second to 605 00:32:37,560 --> 00:32:40,960 Speaker 1: votes on twelve billion dollars to R and D efforts 606 00:32:41,040 --> 00:32:44,040 Speaker 1: in in ships and so on. The On the manufacturing side, 607 00:32:44,080 --> 00:32:47,400 Speaker 1: I think there's no doubt that the subsidy funds will 608 00:32:47,480 --> 00:32:50,920 Speaker 1: result in more manufacturing facilities being built in the US, 609 00:32:51,000 --> 00:32:53,960 Speaker 1: and that's sort of a simple economic calculation. And I 610 00:32:54,000 --> 00:32:56,600 Speaker 1: think we've already seen many of the firms that would 611 00:32:56,640 --> 00:33:00,800 Speaker 1: be impacted, Samsung and t SMC, Intel, Mike Run already 612 00:33:00,920 --> 00:33:03,320 Speaker 1: in the process or in the announcement stage of of 613 00:33:03,320 --> 00:33:05,600 Speaker 1: opening new facilities. So so there will certainly be an 614 00:33:05,640 --> 00:33:07,880 Speaker 1: impact on the amount of chip making in the US. 615 00:33:08,000 --> 00:33:12,240 Speaker 1: Is it going to dramatically change the status quo? You know, 616 00:33:12,280 --> 00:33:13,800 Speaker 1: it depends what you mean by that. But right now 617 00:33:13,800 --> 00:33:16,120 Speaker 1: the US produces ten or fifteen percent of chips. We're 618 00:33:16,120 --> 00:33:17,920 Speaker 1: not going to see that go up to fifty or 619 00:33:17,920 --> 00:33:20,080 Speaker 1: anything near that level anytime soon, but we will have 620 00:33:20,120 --> 00:33:22,120 Speaker 1: more leading edge ship making in the U S and 621 00:33:22,200 --> 00:33:24,400 Speaker 1: we otherwise would have. I think that's that's pretty clear. 622 00:33:24,480 --> 00:33:26,160 Speaker 1: I think the more interesting part actually is the R 623 00:33:26,160 --> 00:33:28,360 Speaker 1: and D funds, because my guess is that when we 624 00:33:28,400 --> 00:33:31,360 Speaker 1: look back in ten years time and ask what was 625 00:33:31,400 --> 00:33:33,280 Speaker 1: the biggest impact of the Chips Act, it will actually 626 00:33:33,280 --> 00:33:36,760 Speaker 1: not be the subsidization of manufacturing, but actually the R 627 00:33:36,840 --> 00:33:38,400 Speaker 1: and D because the amount of money that's going into 628 00:33:38,440 --> 00:33:40,680 Speaker 1: chips R and D is is really substantial. There's always 629 00:33:40,720 --> 00:33:44,360 Speaker 1: been some via the National Science Foundation and some via DARPA, 630 00:33:44,480 --> 00:33:47,040 Speaker 1: but twelve million dollars is a big influx over a 631 00:33:47,040 --> 00:33:48,960 Speaker 1: fairly short period of time, and so I think this 632 00:33:49,040 --> 00:33:51,840 Speaker 1: actually has the chance to have a really substantial impact 633 00:33:52,160 --> 00:33:55,440 Speaker 1: overcoming decade. So just listening to this and the amount 634 00:33:55,440 --> 00:33:57,840 Speaker 1: of money and effort that the US is going to 635 00:33:57,880 --> 00:34:02,200 Speaker 1: put into building out its own ships capacity, is the 636 00:34:02,320 --> 00:34:06,560 Speaker 1: US actually aligned with some of its allies such as 637 00:34:06,640 --> 00:34:10,560 Speaker 1: Taiwan potentially Europe when it comes to its plans for 638 00:34:10,680 --> 00:34:14,799 Speaker 1: chips Because it feels like there's a tension here if 639 00:34:14,840 --> 00:34:18,319 Speaker 1: the US has to balance between saying we're going to 640 00:34:18,360 --> 00:34:21,560 Speaker 1: build out our own capability versus we're going to take 641 00:34:21,600 --> 00:34:25,280 Speaker 1: market share away from TSMC, which is a huge company 642 00:34:25,280 --> 00:34:27,800 Speaker 1: in Taiwan, of course, which is one of the US's 643 00:34:28,000 --> 00:34:31,839 Speaker 1: major military allies in the Asia region. That does seem 644 00:34:31,920 --> 00:34:35,080 Speaker 1: like a bit of a balancing act. Yeah, it is, 645 00:34:35,080 --> 00:34:36,920 Speaker 1: and especially when you set that next to the export 646 00:34:36,960 --> 00:34:41,879 Speaker 1: control side as well, which imposes interesting differential effects on 647 00:34:41,880 --> 00:34:45,440 Speaker 1: on different firms um some some negative, some positive, depending 648 00:34:45,440 --> 00:34:47,960 Speaker 1: on which control you're looking at. And so I think 649 00:34:47,960 --> 00:34:49,760 Speaker 1: you do need to look at the Chips Act funding, 650 00:34:49,960 --> 00:34:52,959 Speaker 1: which applies both to US firms and for non US 651 00:34:53,000 --> 00:34:56,080 Speaker 1: firms building facilities in the US, alongside the X four 652 00:34:56,120 --> 00:34:59,759 Speaker 1: controls as part of a unified strategy to coerce and 653 00:34:59,840 --> 00:35:03,480 Speaker 1: a tracked the world's ship firms more into the U 654 00:35:03,480 --> 00:35:07,319 Speaker 1: S ecosystem and less into the Chinese ecosystem. In terms 655 00:35:07,400 --> 00:35:12,719 Speaker 1: of advanced fabrication subsidies, I think there's certainly is a 656 00:35:12,760 --> 00:35:15,719 Speaker 1: bit of a zero sum dynamic between the US and 657 00:35:16,120 --> 00:35:19,200 Speaker 1: Taiwan and the US and Korea in the following sense 658 00:35:19,239 --> 00:35:23,120 Speaker 1: that both the SMC and Samsung prefer to have most 659 00:35:23,160 --> 00:35:25,440 Speaker 1: of the capacity in their home country, and they prefer 660 00:35:25,480 --> 00:35:26,960 Speaker 1: to have most not all of their R and D 661 00:35:27,200 --> 00:35:29,600 Speaker 1: in their home country. And they prefer that both three 662 00:35:29,600 --> 00:35:31,920 Speaker 1: economic reasons that their businesses work quite well and the 663 00:35:31,960 --> 00:35:34,560 Speaker 1: current set up, and also I think for political reasons, 664 00:35:34,640 --> 00:35:37,360 Speaker 1: and so the extent that the US is is pushing 665 00:35:37,400 --> 00:35:39,320 Speaker 1: for more investment in the US, there there is a 666 00:35:39,320 --> 00:35:41,440 Speaker 1: bit of a zerosum dynamic. Now the US is not 667 00:35:41,520 --> 00:35:44,960 Speaker 1: just pushing, it's also subsidizing, so that helps to compensate 668 00:35:45,000 --> 00:35:47,240 Speaker 1: them to some extent. But I think you're right, Tracy 669 00:35:47,320 --> 00:35:50,000 Speaker 1: to suggest that t SMC wouldn't have been pushing for 670 00:35:50,040 --> 00:35:52,120 Speaker 1: any of these changes if it were in charge. It's 671 00:35:52,120 --> 00:35:55,480 Speaker 1: it's responding to shifts that governments are imposing on the 672 00:35:55,640 --> 00:36:00,399 Speaker 1: this my conductor industry more generally in terms of other 673 00:36:00,440 --> 00:36:04,000 Speaker 1: subsidy programs in Japan, Singapore, from Europe, I think it 674 00:36:04,040 --> 00:36:06,960 Speaker 1: depends who you're looking at. The Japanese, for example, have 675 00:36:07,320 --> 00:36:09,960 Speaker 1: subsidized t SMC to build a facility in Japan, but 676 00:36:10,000 --> 00:36:12,240 Speaker 1: it's not focusing on leading edge, is focusing on a 677 00:36:12,280 --> 00:36:15,560 Speaker 1: more of a lacking edge node that is critically relevant 678 00:36:15,600 --> 00:36:17,480 Speaker 1: for Sony, which is a big customer of t s 679 00:36:17,560 --> 00:36:19,719 Speaker 1: mc s and so that doesn't really conflict much at 680 00:36:19,760 --> 00:36:21,400 Speaker 1: all with with what the U S is planning to 681 00:36:21,440 --> 00:36:25,719 Speaker 1: spend money on. With Europe. Europe's own chips bill is 682 00:36:25,760 --> 00:36:28,239 Speaker 1: still a work in progress, and I think we're gonna 683 00:36:28,239 --> 00:36:30,279 Speaker 1: have to wait and see what actually Europe decides to 684 00:36:30,320 --> 00:36:33,160 Speaker 1: spend its money on. I I worry most actually about 685 00:36:33,320 --> 00:36:36,799 Speaker 1: potential market inefficiencies created by europe chip subsidies, since they 686 00:36:36,800 --> 00:36:39,200 Speaker 1: remained quite unclear as to where they're headed. I think 687 00:36:39,239 --> 00:36:42,160 Speaker 1: it's harder to imagine that Europe will succeed leading edge 688 00:36:42,160 --> 00:36:45,080 Speaker 1: manufacturing simply because it has less infrastructure even than the US. 689 00:36:45,200 --> 00:36:48,399 Speaker 1: But if Europe puts its money towards lagging edge manufacturing, 690 00:36:49,040 --> 00:36:51,360 Speaker 1: that's an industry that already is getting plenty of subsidies 691 00:36:51,360 --> 00:36:53,239 Speaker 1: from around the world, including from China. So I think 692 00:36:53,239 --> 00:36:55,480 Speaker 1: there are questions to be asked about how Europe actually 693 00:36:55,480 --> 00:36:58,040 Speaker 1: plans to spend the fifty billion euros that it intends 694 00:36:58,040 --> 00:37:01,520 Speaker 1: to devote to the industry. So mentioned that, Okay, the 695 00:37:01,600 --> 00:37:04,799 Speaker 1: US is probably not going to dramatically jump over the 696 00:37:04,840 --> 00:37:08,680 Speaker 1: next several years in a share of global semiconductor manufacturing, 697 00:37:09,160 --> 00:37:11,640 Speaker 1: but it presumably the goal is for it to increase 698 00:37:11,680 --> 00:37:15,160 Speaker 1: a bit in ten years or however many years. What 699 00:37:15,239 --> 00:37:17,640 Speaker 1: would we have to see to say the Chips Act 700 00:37:17,920 --> 00:37:20,720 Speaker 1: was a success or what would constitute success? I mean again, 701 00:37:20,880 --> 00:37:24,040 Speaker 1: and even R and D can sometimes be ambiguous what 702 00:37:24,120 --> 00:37:26,920 Speaker 1: that turns into in terms of commercialized products. So from 703 00:37:26,960 --> 00:37:31,200 Speaker 1: your perspective, what would constitute okay, this bill actually achieved 704 00:37:31,200 --> 00:37:34,040 Speaker 1: a taim I think part of the way to conceptualize 705 00:37:34,080 --> 00:37:36,160 Speaker 1: this bill is as an insurance policy. And it's an 706 00:37:36,160 --> 00:37:39,400 Speaker 1: insurance policy that is protecting us in case of some 707 00:37:39,440 --> 00:37:43,200 Speaker 1: sort of escalation around Taiwan, which could disrupt hip production 708 00:37:43,280 --> 00:37:45,640 Speaker 1: and if so, it would be dramatically costly for the 709 00:37:45,640 --> 00:37:49,000 Speaker 1: world economy, and so so calculating the efficacy of an 710 00:37:49,000 --> 00:37:51,520 Speaker 1: insurance policy is a hard thing to do, especially because 711 00:37:51,560 --> 00:37:53,920 Speaker 1: ideally you don't have to cash it in. But I 712 00:37:53,960 --> 00:37:56,279 Speaker 1: think that is the right way to look at this. 713 00:37:56,560 --> 00:37:59,399 Speaker 1: And again, it's not that we're going to go from 714 00:38:00,040 --> 00:38:02,760 Speaker 1: scale dependence in Taiwan to zero, but that on the margin, 715 00:38:02,800 --> 00:38:05,080 Speaker 1: we're going to become less dependent on the supply of 716 00:38:05,160 --> 00:38:08,200 Speaker 1: chips crossing through the Taiwan straight and that that reduces 717 00:38:08,239 --> 00:38:10,400 Speaker 1: the risk in case of some sort of war, and 718 00:38:10,440 --> 00:38:13,480 Speaker 1: it also reduces China's ability to use our reliance on 719 00:38:13,480 --> 00:38:16,080 Speaker 1: Taiwan as a as a mechanism of pressure. So I 720 00:38:16,160 --> 00:38:17,800 Speaker 1: think that's that's the first thing is that there is 721 00:38:17,840 --> 00:38:20,600 Speaker 1: sort of a national security insurance mechanism here that is 722 00:38:20,800 --> 00:38:23,640 Speaker 1: is hard to price in economic terms. The second thing 723 00:38:23,640 --> 00:38:27,000 Speaker 1: on R and D, I think actually we should be 724 00:38:27,040 --> 00:38:30,520 Speaker 1: able to pretty clearly know in a decade's time whether 725 00:38:31,120 --> 00:38:33,160 Speaker 1: the R and D funding worked or didn't work, and 726 00:38:33,320 --> 00:38:36,520 Speaker 1: if if we can't find clear evidence of pretty direct 727 00:38:36,560 --> 00:38:39,480 Speaker 1: lines between spending and commercial successes, I think we should 728 00:38:39,480 --> 00:38:41,600 Speaker 1: tell ourselves that it it failed. And this is where 729 00:38:41,640 --> 00:38:44,000 Speaker 1: I think there's the most work actually to be done 730 00:38:44,040 --> 00:38:45,799 Speaker 1: at the Commerce Department and else where in the US 731 00:38:45,880 --> 00:38:48,480 Speaker 1: government is that it's it's easy to dole out funds 732 00:38:48,560 --> 00:38:52,680 Speaker 1: to subsidize evance manufacturing, especially when the number of plausible 733 00:38:52,680 --> 00:38:55,279 Speaker 1: firms that could be involved is really limited. And if 734 00:38:55,280 --> 00:38:57,520 Speaker 1: you're going to build an advanced logic facility, there's only 735 00:38:57,560 --> 00:38:58,880 Speaker 1: three firms in the world they're going to do it, 736 00:38:59,280 --> 00:39:01,440 Speaker 1: and the US government will probably give money to all 737 00:39:01,440 --> 00:39:03,440 Speaker 1: three of them, so that that job is relatively easy. 738 00:39:03,520 --> 00:39:05,200 Speaker 1: The harder job is how do you set up and 739 00:39:05,480 --> 00:39:08,359 Speaker 1: build out a R and D infrastructure that will fill 740 00:39:08,440 --> 00:39:10,359 Speaker 1: some of the gaps that we currently have, and that's 741 00:39:10,360 --> 00:39:12,320 Speaker 1: a much harder task, but I think in some ways 742 00:39:12,320 --> 00:39:15,400 Speaker 1: an even more important one. So since you mentioned the 743 00:39:15,440 --> 00:39:19,520 Speaker 1: idea of the US kind of hedging disruption in Taiwan, 744 00:39:19,960 --> 00:39:22,799 Speaker 1: I have to ask a question that's getting a lot 745 00:39:22,840 --> 00:39:25,680 Speaker 1: of attention at the moment. You know, loads of people 746 00:39:25,800 --> 00:39:30,560 Speaker 1: on Twitter talking about theoretical situations and various possibilities here. 747 00:39:30,960 --> 00:39:35,320 Speaker 1: But if there was some sort of invasion of Taiwan 748 00:39:35,560 --> 00:39:39,239 Speaker 1: or some sort of military action against the country, how 749 00:39:39,239 --> 00:39:43,080 Speaker 1: would you expect that to impact t s MC. And 750 00:39:43,200 --> 00:39:45,440 Speaker 1: you know, one of the things that comes up in 751 00:39:45,480 --> 00:39:48,719 Speaker 1: these discussions is the idea that if China did a 752 00:39:48,840 --> 00:39:52,440 Speaker 1: full on invasion and they took over all the fabs 753 00:39:52,440 --> 00:39:55,840 Speaker 1: in Taiwan, would they be able to just keep manufacturing 754 00:39:55,880 --> 00:40:00,080 Speaker 1: semiconductors like they were before the invasion or is there 755 00:40:00,160 --> 00:40:03,279 Speaker 1: something particular about the chip making business that makes that 756 00:40:03,400 --> 00:40:07,160 Speaker 1: scenario more difficult. Well, I think if if we have 757 00:40:07,280 --> 00:40:09,600 Speaker 1: a sort of D Day style invasion, if you will, 758 00:40:09,800 --> 00:40:13,719 Speaker 1: the likelihood that any semi chectural facilities in Taiwan remain 759 00:40:13,760 --> 00:40:17,160 Speaker 1: intact after the fact is pretty low. We're talking about 760 00:40:17,160 --> 00:40:20,439 Speaker 1: the most precise machinery ever made. Chip fabs are full 761 00:40:20,440 --> 00:40:24,920 Speaker 1: of explosive gases. It seems perfectly designed to not survive 762 00:40:24,960 --> 00:40:28,160 Speaker 1: a war. And even if by some miracle, t SMCS 763 00:40:28,160 --> 00:40:30,840 Speaker 1: facilities did survive a war, you'd still need all the 764 00:40:30,880 --> 00:40:33,680 Speaker 1: personnel in place, and in place in a way that 765 00:40:33,719 --> 00:40:37,440 Speaker 1: they wanted to keep facilitating the production process rather than 766 00:40:37,480 --> 00:40:41,160 Speaker 1: sabotaging it. And you still need supplies from abroad because 767 00:40:41,200 --> 00:40:44,680 Speaker 1: t SMC relies on chemicals, for example, sourced from from 768 00:40:44,760 --> 00:40:48,640 Speaker 1: Japan and elsewhere. So I think it's it's implausible, close 769 00:40:48,680 --> 00:40:52,520 Speaker 1: to a zero present probability that China could invade Taiwan, 770 00:40:52,600 --> 00:40:55,000 Speaker 1: would have a massive war, and at the end, China 771 00:40:55,000 --> 00:40:59,240 Speaker 1: would have perfectly functional chipmaking facilities. Now, whether Chinese leaders 772 00:40:59,320 --> 00:41:02,680 Speaker 1: understand that, you know, I don't know. I don't know 773 00:41:02,760 --> 00:41:05,960 Speaker 1: to what extent they've dug into the chip making production process. 774 00:41:05,960 --> 00:41:07,719 Speaker 1: And I do worry a bit that there is a 775 00:41:07,719 --> 00:41:10,320 Speaker 1: fair amount of discussion in Chinese social media, for example, 776 00:41:10,360 --> 00:41:14,560 Speaker 1: and even among Chinese government aligned analysts about taking Taiwan 777 00:41:14,640 --> 00:41:16,400 Speaker 1: but to seize the chip fabs. But I think that 778 00:41:16,440 --> 00:41:20,399 Speaker 1: idea is a fantasy. But I also think that when 779 00:41:20,440 --> 00:41:23,600 Speaker 1: you envision military escalation scenarios there's a lot that could 780 00:41:23,640 --> 00:41:26,160 Speaker 1: happen between peace where we are right now and a 781 00:41:26,239 --> 00:41:29,680 Speaker 1: D Day style invasion. And in those types of coercion, 782 00:41:29,920 --> 00:41:32,600 Speaker 1: coursive measures or gray zone measures, if you will, in 783 00:41:32,680 --> 00:41:36,879 Speaker 1: military lingo, that's where I think you could envision down 784 00:41:36,920 --> 00:41:40,239 Speaker 1: the road, China applying pressure making demands on Taiwan, and 785 00:41:40,280 --> 00:41:42,000 Speaker 1: if Taiwan doesn't feel like it has the backing of 786 00:41:42,000 --> 00:41:43,680 Speaker 1: the United States or doesn't feel like the U S 787 00:41:43,680 --> 00:41:47,520 Speaker 1: could actually successfully come to its aid militarily, Taiwan might 788 00:41:47,560 --> 00:41:50,880 Speaker 1: be pressured too, for example, change its regulations when it 789 00:41:50,920 --> 00:41:53,920 Speaker 1: came to access to t SMC s FABS. And so 790 00:41:53,960 --> 00:41:56,640 Speaker 1: that that's a more plausible scenario to me, which which 791 00:41:56,640 --> 00:41:59,680 Speaker 1: could still have dramatic implications for the United States. And 792 00:41:59,719 --> 00:42:03,200 Speaker 1: you can terms also in military terms. Chris Miller, it 793 00:42:03,280 --> 00:42:05,879 Speaker 1: was such a treat to have you back on and 794 00:42:05,920 --> 00:42:10,160 Speaker 1: sort of again extraordinary confluence of expertise, news and new 795 00:42:10,200 --> 00:42:13,439 Speaker 1: book out. Appreciate you coming back on odd lots. Well, 796 00:42:13,440 --> 00:42:15,360 Speaker 1: thanks for having me back. Thanks so much, Chris. That 797 00:42:15,440 --> 00:42:31,839 Speaker 1: was great. It was fantastic, Trazy, that was great. I mean, 798 00:42:31,920 --> 00:42:34,040 Speaker 1: Chris really was the perfect guest. But you know, just 799 00:42:34,080 --> 00:42:36,840 Speaker 1: sort of like working backwards through some of the things 800 00:42:36,920 --> 00:42:42,320 Speaker 1: that he discussed, you know, the implausibility of Taiwan semiconductor 801 00:42:42,320 --> 00:42:45,120 Speaker 1: capacity like holding up in like a sort of like 802 00:42:45,560 --> 00:42:49,200 Speaker 1: true like hot war style invasion. I think, like, just right, 803 00:42:49,239 --> 00:42:51,520 Speaker 1: that's an interesting point to start because again, you know, 804 00:42:51,560 --> 00:42:56,240 Speaker 1: ship manufacturing is so complex, relies on such a global network, 805 00:42:56,400 --> 00:42:58,799 Speaker 1: relies on so much sort of embedded expertise of the 806 00:42:58,800 --> 00:43:00,560 Speaker 1: people who work there. It really is sort of hard 807 00:43:00,600 --> 00:43:03,520 Speaker 1: to imagine that you could have that and the chip 808 00:43:03,560 --> 00:43:06,160 Speaker 1: fabs just keep running, right. This was sort of why 809 00:43:06,200 --> 00:43:09,520 Speaker 1: I asked that question about the oil analogy at the beginning, 810 00:43:09,600 --> 00:43:12,400 Speaker 1: because I think, you know, lots of people look at 811 00:43:12,440 --> 00:43:15,280 Speaker 1: it and they go, oh, it's a strategic resource. People 812 00:43:15,440 --> 00:43:19,960 Speaker 1: use chips for military applications. There's actually some great anecdotes 813 00:43:20,000 --> 00:43:22,759 Speaker 1: in Chris's book about how Russia's lack of chips is 814 00:43:22,920 --> 00:43:26,680 Speaker 1: basically hindering its invasion of Ukraine. Right now, the temptation 815 00:43:27,000 --> 00:43:30,680 Speaker 1: is to draw a direct parallel between oil and chips, 816 00:43:30,719 --> 00:43:34,480 Speaker 1: But on the other hand, there is a key difference. 817 00:43:34,560 --> 00:43:38,759 Speaker 1: You know, you can build up your own semiconductor capacity 818 00:43:38,880 --> 00:43:41,640 Speaker 1: in many respects, it's far more difficult than you know, 819 00:43:41,840 --> 00:43:45,640 Speaker 1: just installing an oil rig to pull crude oil out 820 00:43:45,640 --> 00:43:48,200 Speaker 1: of the ground. It does seem to take years and 821 00:43:48,280 --> 00:43:51,560 Speaker 1: vast amounts of money, but there is that possibility. It's 822 00:43:51,640 --> 00:43:55,719 Speaker 1: not bounded by geography in the same way that oil is. 823 00:43:56,000 --> 00:43:58,439 Speaker 1: You know, another thing that's really helpful trying to create 824 00:43:58,719 --> 00:44:01,680 Speaker 1: is the sort of distinction. And both are important, but 825 00:44:01,760 --> 00:44:06,239 Speaker 1: the distinction between leading edge and lagging tech and both 826 00:44:06,280 --> 00:44:09,239 Speaker 1: are preticulational. I mean, we started talking about chips, you know, 827 00:44:09,320 --> 00:44:11,120 Speaker 1: one of the big themes was like, well, what's going 828 00:44:11,120 --> 00:44:14,120 Speaker 1: on with the automakers that's not advanced tech? Right? A 829 00:44:14,120 --> 00:44:16,880 Speaker 1: lot of that stuff was like the least advanced chips. 830 00:44:17,200 --> 00:44:19,960 Speaker 1: It was, you know, stuff to like that displayed an 831 00:44:20,000 --> 00:44:23,040 Speaker 1: image on a dashboard or something like that, about as 832 00:44:23,040 --> 00:44:25,520 Speaker 1: basic as it gets. But if you don't have that, 833 00:44:25,640 --> 00:44:27,560 Speaker 1: then you know, it doesn't matter that is leading edge 834 00:44:27,760 --> 00:44:29,640 Speaker 1: or sort of lagging edge or whatever. If you don't 835 00:44:29,680 --> 00:44:32,200 Speaker 1: have it, you can't make the car. And so, you know, 836 00:44:32,280 --> 00:44:34,520 Speaker 1: thinking about the Chips Act, and as he put it, 837 00:44:34,560 --> 00:44:38,480 Speaker 1: you know, this sort of insurance against this disruption, like, Okay, 838 00:44:38,520 --> 00:44:43,040 Speaker 1: maybe the subsidies won't necessarily bring the leading edge manufacturing 839 00:44:43,160 --> 00:44:45,960 Speaker 1: or expand that, but expanding access to the lagging edge 840 00:44:46,000 --> 00:44:49,080 Speaker 1: is also pretty crucial, you know, the ubiquity of chips 841 00:44:49,120 --> 00:44:52,480 Speaker 1: and everything from cars to refrigerators and washing machines or 842 00:44:52,480 --> 00:44:57,160 Speaker 1: whatever appliance. I do find that kind of concerning the 843 00:44:57,239 --> 00:45:01,560 Speaker 1: idea that you could use chips to hold the global 844 00:45:01,600 --> 00:45:05,960 Speaker 1: economy hostage for a certain strategic or military goal in 845 00:45:06,000 --> 00:45:08,760 Speaker 1: the same way that you know, to some extent, Russia 846 00:45:08,800 --> 00:45:11,160 Speaker 1: has used oil and gas to do the same thing 847 00:45:11,760 --> 00:45:14,239 Speaker 1: that worries me, it does. You know. The other thing 848 00:45:14,320 --> 00:45:17,400 Speaker 1: about that, though, is like again, because there's so much interconnection, 849 00:45:17,480 --> 00:45:19,600 Speaker 1: and you brought up software. You know, I'm glad you 850 00:45:19,680 --> 00:45:21,480 Speaker 1: brought that out because actually, at some point we should 851 00:45:21,480 --> 00:45:24,520 Speaker 1: do like a design software thing, go back to the 852 00:45:24,560 --> 00:45:28,080 Speaker 1: supply chain. But like everybody depends on anyone everyone else, right, 853 00:45:28,400 --> 00:45:31,480 Speaker 1: So in a way, it's hard to imagine like this 854 00:45:31,600 --> 00:45:34,880 Speaker 1: sort of like standoff like without like really hurting yourself 855 00:45:34,920 --> 00:45:38,360 Speaker 1: because everyone, you know, it's so cross border. You mentioned 856 00:45:38,400 --> 00:45:42,319 Speaker 1: like Japanese chemicals and software that might be designed in 857 00:45:42,360 --> 00:45:44,319 Speaker 1: the United States, which I think the US is still 858 00:45:44,320 --> 00:45:47,319 Speaker 1: really good at designing a design software, And so it's 859 00:45:47,400 --> 00:45:52,960 Speaker 1: that sort of makes the chips chips chips diplomacy tricky 860 00:45:53,000 --> 00:45:55,600 Speaker 1: and unique in the sense that it seems really hard 861 00:45:55,640 --> 00:45:59,320 Speaker 1: to like hurt your enemy or hurt another entity without 862 00:45:59,360 --> 00:46:02,160 Speaker 1: harming yourself. Chips Diplomacy is a good way of putting it. 863 00:46:02,360 --> 00:46:06,239 Speaker 1: That should be the name of our book, Joe. The 864 00:46:06,320 --> 00:46:09,480 Speaker 1: market for good semiconductor books, I think has just been 865 00:46:09,520 --> 00:46:16,400 Speaker 1: filled knockoff version of chip wark Tray chip Diplomacy. All right, 866 00:46:16,719 --> 00:46:19,239 Speaker 1: shall we leave it? Let's leave it there. Okay, this 867 00:46:19,320 --> 00:46:22,319 Speaker 1: has been another episode of the Odd Lots podcast. I'm 868 00:46:22,320 --> 00:46:25,080 Speaker 1: Tracy Alloway. You can follow me on Twitter at Tracy 869 00:46:25,120 --> 00:46:27,600 Speaker 1: Alloway and I'm Joe wi Isn't Thalt. You can follow 870 00:46:27,600 --> 00:46:31,000 Speaker 1: me on Twitter at the Stalwart. Definitely follow our guest 871 00:46:31,080 --> 00:46:34,239 Speaker 1: Chris Miller. He's at c R Miller One. Follow our 872 00:46:34,280 --> 00:46:38,040 Speaker 1: producer Kerman Rodriguez at Kerman Arman, and check out all 873 00:46:38,080 --> 00:46:41,640 Speaker 1: of our podcasts at Bloomberg under the handle at podcasts. 874 00:46:41,800 --> 00:47:04,640 Speaker 1: Thanks for listening year to