1 00:00:10,840 --> 00:00:15,120 Speaker 1: Hello, and welcome to another episode of the Odd Lots Podcast. 2 00:00:15,160 --> 00:00:20,080 Speaker 1: I'm Joe Wisn't Thal and I'm Tracy Alloway. So obviously, Tracy, 3 00:00:20,239 --> 00:00:23,759 Speaker 1: we've been talking a lot about chips lately, but for 4 00:00:23,840 --> 00:00:26,880 Speaker 1: all the episodes we've done, we haven't hit like what 5 00:00:27,080 --> 00:00:28,800 Speaker 1: is sort of like UM. I guess I would say 6 00:00:29,040 --> 00:00:31,639 Speaker 1: the elephant in the room or the guerrilla in the 7 00:00:31,720 --> 00:00:36,520 Speaker 1: room that keeps coming up over and over again. Yeah, 8 00:00:36,680 --> 00:00:39,239 Speaker 1: we've been going at it, I would say from a 9 00:00:39,360 --> 00:00:43,600 Speaker 1: US perspective, very focused on the troubles at Intel, but 10 00:00:43,960 --> 00:00:47,320 Speaker 1: we haven't really talked about the success story that is 11 00:00:47,560 --> 00:00:54,120 Speaker 1: Taiwan Semiconductor Manufacturing Company better known as t s m C. Yeah, 12 00:00:54,160 --> 00:00:56,640 Speaker 1: exactly right. So every episode that we do, it sort 13 00:00:56,640 --> 00:00:59,840 Speaker 1: of comes back to them what a juggernaut they've become. 14 00:01:00,000 --> 00:01:02,880 Speaker 1: And of course, you know, the sort of basic story 15 00:01:03,360 --> 00:01:09,280 Speaker 1: is that manufacturing chips is extremely hard, it's extremely expensive, 16 00:01:10,160 --> 00:01:13,160 Speaker 1: and it's a very difficult to do its scale until 17 00:01:13,480 --> 00:01:17,200 Speaker 1: UM is one of the rare companies that designs and 18 00:01:17,360 --> 00:01:20,120 Speaker 1: manufactures its own chips. But that for a lot of 19 00:01:20,160 --> 00:01:25,280 Speaker 1: these companies that are sort of exploding riot gaining market share, 20 00:01:25,400 --> 00:01:29,880 Speaker 1: they're having Taiwan Semi manufacture them, and Taiwan Semi is 21 00:01:29,920 --> 00:01:34,479 Speaker 1: getting extremely good or is extremely good at manufacturing and 22 00:01:34,560 --> 00:01:37,920 Speaker 1: maybe pulling away from Intel to some extent in terms 23 00:01:37,920 --> 00:01:42,560 Speaker 1: of its manufacturing capabilities, right, And I think they actually 24 00:01:42,600 --> 00:01:45,840 Speaker 1: invented the foundry model, which you know, this idea of 25 00:01:45,880 --> 00:01:49,200 Speaker 1: just manufacturing chips, which keeps coming up over and over 26 00:01:49,240 --> 00:01:52,120 Speaker 1: and over again in all of our conversations, as you know, 27 00:01:52,280 --> 00:01:55,720 Speaker 1: one reason the entire semiconductor industry has changed and one 28 00:01:55,760 --> 00:01:59,680 Speaker 1: reason why Intel is struggling. But the thing I find 29 00:01:59,760 --> 00:02:03,720 Speaker 1: kay amazing, Like in one we all take it for 30 00:02:03,760 --> 00:02:06,600 Speaker 1: granted that t SMC is this massive player in the 31 00:02:06,680 --> 00:02:11,480 Speaker 1: semiconductor industry in the world, really, but I find it 32 00:02:11,639 --> 00:02:15,679 Speaker 1: really really noteworthy, and you know, somewhat surprising in retrospect 33 00:02:15,720 --> 00:02:19,959 Speaker 1: that what is a single company on an island has 34 00:02:20,520 --> 00:02:24,760 Speaker 1: emerged to really dominate chips that are now vital to 35 00:02:25,200 --> 00:02:29,600 Speaker 1: all sorts of things, so computers, smartphones, cars, everything. I 36 00:02:29,639 --> 00:02:34,680 Speaker 1: think TSMC manufactures a little over half of the world's chips, 37 00:02:34,960 --> 00:02:38,720 Speaker 1: so it's not quite a monopoly. But again, like they 38 00:02:38,760 --> 00:02:42,680 Speaker 1: are the juggernaut in the room as you put it. Yeah, absolutely, 39 00:02:42,680 --> 00:02:44,680 Speaker 1: and it's such a I mean, it's such a pivotal 40 00:02:44,680 --> 00:02:48,000 Speaker 1: company for a lot of reasons. I mean you mentioned cars. 41 00:02:48,080 --> 00:02:50,960 Speaker 1: There was just a story this week recording this January twelve, 42 00:02:51,040 --> 00:02:52,760 Speaker 1: by the way, there was just a story this week 43 00:02:53,280 --> 00:02:57,120 Speaker 1: about how car manufacturers around the world are actually running 44 00:02:57,120 --> 00:03:00,320 Speaker 1: into supply constraints because they can't get chips because of 45 00:03:00,320 --> 00:03:04,720 Speaker 1: the chip manufacturers decided to reduce their production of automotive 46 00:03:04,800 --> 00:03:08,160 Speaker 1: chips during the crisis in the spring on the expectation 47 00:03:08,240 --> 00:03:10,120 Speaker 1: that demand wasn't going to be there, and now they 48 00:03:10,120 --> 00:03:13,320 Speaker 1: haven't ramped up. Taiwan Semi, uh, they're going to They're 49 00:03:13,360 --> 00:03:16,480 Speaker 1: actually reporting earnings this week expected to be very strong. 50 00:03:16,720 --> 00:03:19,800 Speaker 1: Earnings are soaring, the stock is soaring. And then of 51 00:03:19,840 --> 00:03:23,919 Speaker 1: course because it's in this strategic position, and because Taiwan 52 00:03:24,000 --> 00:03:29,040 Speaker 1: itself is in this strategic position obviously between the US 53 00:03:29,120 --> 00:03:31,880 Speaker 1: and China and the rollover its status, it's just an 54 00:03:31,919 --> 00:03:37,800 Speaker 1: incredibly central player and sort of necessary to understand the story, right, 55 00:03:37,920 --> 00:03:40,960 Speaker 1: A big player in tech, a big player across multiple 56 00:03:41,000 --> 00:03:43,920 Speaker 1: supply chains in a bunch of different industries, and certainly 57 00:03:43,960 --> 00:03:47,480 Speaker 1: a player in geopolitics as well. So we really got 58 00:03:47,480 --> 00:03:50,360 Speaker 1: to talk more Taiwan Semi, and we have the perfect 59 00:03:50,720 --> 00:03:53,000 Speaker 1: guests to do that, someone who knows all about the company, 60 00:03:53,040 --> 00:03:55,440 Speaker 1: who has been covering them for a long time, since 61 00:03:55,920 --> 00:04:00,320 Speaker 1: long before they were the dominant player, uh they are today. 62 00:04:00,360 --> 00:04:02,480 Speaker 1: We're going to be speaking to Tim Colpen. He is 63 00:04:02,520 --> 00:04:06,120 Speaker 1: a tech columnist at Bloomberg Opinion. It's been with us 64 00:04:06,120 --> 00:04:09,720 Speaker 1: a Bloomberg for fifteen years and he's actually been covering 65 00:04:10,040 --> 00:04:14,800 Speaker 1: the tech industry from Taipei specifically for the last twenty 66 00:04:14,840 --> 00:04:19,760 Speaker 1: one years. So there's everything about Taiwan Semi Conductor, where 67 00:04:19,800 --> 00:04:22,920 Speaker 1: it came from, how it got to be this important player. 68 00:04:23,520 --> 00:04:26,520 Speaker 1: Um So, Tim, thank you very much for joining us. Hey, 69 00:04:26,560 --> 00:04:29,240 Speaker 1: thanks for your time, guys, good to talk to you. Tim. 70 00:04:29,279 --> 00:04:33,800 Speaker 1: How would you describe the importance of Taiwan Semi Conductor 71 00:04:33,880 --> 00:04:38,080 Speaker 1: in the world right now? Well as your assess just 72 00:04:38,160 --> 00:04:41,359 Speaker 1: saying they are so huge Tracy pointed out, they create, 73 00:04:41,800 --> 00:04:44,400 Speaker 1: you know, such a large share of the world's chips. 74 00:04:45,320 --> 00:04:48,320 Speaker 1: There is probably not a single device in the world 75 00:04:48,720 --> 00:04:52,200 Speaker 1: that doesn't have t SMC in there somewhere, whether it's 76 00:04:52,200 --> 00:04:56,440 Speaker 1: a car, the iPhone of course famously, but even some 77 00:04:56,520 --> 00:04:59,760 Speaker 1: really unimportant boring things like a temperature sense that he 78 00:05:00,760 --> 00:05:05,320 Speaker 1: or some led lighter rate there. They do so much 79 00:05:05,360 --> 00:05:08,560 Speaker 1: of the walls, products they're in, they're basically everywhere, and 80 00:05:08,680 --> 00:05:12,440 Speaker 1: so if they were to stop production tomorrow, I think 81 00:05:12,440 --> 00:05:15,080 Speaker 1: the global technology industry would grind to a halt very 82 00:05:15,160 --> 00:05:18,599 Speaker 1: very quickly. It's so, I mean, I mentioned this in 83 00:05:18,600 --> 00:05:21,600 Speaker 1: the intro, but we take it as a given that 84 00:05:21,680 --> 00:05:26,200 Speaker 1: this one country has such a big position in global manufacturing, 85 00:05:26,279 --> 00:05:31,400 Speaker 1: in the global supply chain. How did this actually come about? 86 00:05:31,520 --> 00:05:34,520 Speaker 1: Or let me rephrase that. Let's start at the beginning. 87 00:05:35,279 --> 00:05:40,400 Speaker 1: How did TSMC actually start? Well, the story of t 88 00:05:40,600 --> 00:05:43,400 Speaker 1: s MC is in many ways the story of Taiwan's 89 00:05:43,600 --> 00:05:49,520 Speaker 1: electronics industry. It was founded by Marris Young himself back 90 00:05:49,560 --> 00:05:53,760 Speaker 1: in seven, but the seeds of Taiwan's seven conductor industry 91 00:05:54,360 --> 00:05:58,680 Speaker 1: predated that by about fourteen or fifteen years, when an 92 00:05:58,800 --> 00:06:01,560 Speaker 1: organization was set up by the Taiwan government back then 93 00:06:01,600 --> 00:06:04,600 Speaker 1: in nineteen seventy three called it TRY the Industrial Technology 94 00:06:04,640 --> 00:06:10,240 Speaker 1: Research Institute, and at the time the government realized that technology, 95 00:06:11,120 --> 00:06:14,880 Speaker 1: and not specifically semi conductors because semi conductors were relatively new, 96 00:06:14,960 --> 00:06:18,360 Speaker 1: but integrated circuits were going to be big. The other 97 00:06:18,880 --> 00:06:22,040 Speaker 1: big area for Taiwan at the time was plastics, and 98 00:06:22,040 --> 00:06:25,080 Speaker 1: for most of plastics is one of the world's in 99 00:06:25,080 --> 00:06:28,240 Speaker 1: the plastics industry. And of course you may remember back 100 00:06:28,279 --> 00:06:30,640 Speaker 1: here and your kid, you have these cheap plastic toys 101 00:06:30,680 --> 00:06:33,360 Speaker 1: that were made in Taiwan. Barbie dolls, for example, were 102 00:06:33,360 --> 00:06:35,920 Speaker 1: made in Taiwan for many, many years. So there were 103 00:06:35,920 --> 00:06:39,240 Speaker 1: the two kind of twin engines of Taiwan's future, and 104 00:06:39,279 --> 00:06:42,520 Speaker 1: the government knew that, and so they started getting into 105 00:06:42,560 --> 00:06:46,600 Speaker 1: it through a government institute, spending money on it, and 106 00:06:46,640 --> 00:06:49,359 Speaker 1: then working with the Americans and to a lesser extent 107 00:06:49,480 --> 00:06:54,719 Speaker 1: Japanese and Europeans to transfer that technology to Taiwan to 108 00:06:54,800 --> 00:06:57,440 Speaker 1: be made in Taiwan. So the first chips made out 109 00:06:57,440 --> 00:07:01,039 Speaker 1: of Taiwan. Really we're rolling off the presses or the 110 00:07:01,120 --> 00:07:05,960 Speaker 1: fabs in the mid seventies, and t SMC came along 111 00:07:06,000 --> 00:07:08,360 Speaker 1: in the mid to late eighties, so there was it 112 00:07:08,440 --> 00:07:12,680 Speaker 1: was already underway. But as Joe pointed out, it was 113 00:07:13,120 --> 00:07:17,160 Speaker 1: the foundry of foundries. It was the first one to 114 00:07:17,240 --> 00:07:22,840 Speaker 1: decide that offering a pure manufacturing solution so that clients 115 00:07:22,880 --> 00:07:25,680 Speaker 1: don't need to have factories, and tier SMC would do 116 00:07:25,720 --> 00:07:29,120 Speaker 1: it for them without designing their own chips, only purely 117 00:07:29,120 --> 00:07:32,320 Speaker 1: manufactured for somebody else. That was very, very new it was. 118 00:07:32,440 --> 00:07:35,960 Speaker 1: It was a ridiculous concept at the time, because you know, 119 00:07:36,000 --> 00:07:38,720 Speaker 1: at the time, Jerry Sanders of a m D later 120 00:07:38,800 --> 00:07:41,240 Speaker 1: said in that famous line that real men have fabs. 121 00:07:41,680 --> 00:07:44,920 Speaker 1: At the time you designed and new manufactured, you did 122 00:07:44,920 --> 00:07:48,800 Speaker 1: them both, Cyprus and Faraday and uh sort of free 123 00:07:48,800 --> 00:07:51,920 Speaker 1: scale and all those other companies, of course Intel being 124 00:07:51,960 --> 00:07:54,440 Speaker 1: one of them, they did both. A m D did both. 125 00:07:54,520 --> 00:07:58,160 Speaker 1: And it was only after t SMC came along and said, hey, 126 00:07:58,360 --> 00:08:00,480 Speaker 1: actually you don't need to we can do for you. 127 00:08:00,960 --> 00:08:04,000 Speaker 1: That changed a lot of people's minds and it really 128 00:08:04,040 --> 00:08:08,200 Speaker 1: turned silicon value around. So where did that idea come from? 129 00:08:08,240 --> 00:08:12,840 Speaker 1: I mean, what was the sort of realization that this 130 00:08:12,880 --> 00:08:16,080 Speaker 1: model could be powerful? Making to order has always been 131 00:08:16,160 --> 00:08:20,440 Speaker 1: part of Taiwan's industrial DNA, going back many many years. 132 00:08:20,760 --> 00:08:24,200 Speaker 1: You probably also know the fox Con which makes which 133 00:08:24,240 --> 00:08:26,840 Speaker 1: makes your iPhone is also essentially an o EM they 134 00:08:26,880 --> 00:08:29,840 Speaker 1: manufacture for others, and fox Con, which was getting going 135 00:08:29,840 --> 00:08:33,400 Speaker 1: at a round about the same time, started off making 136 00:08:33,679 --> 00:08:37,240 Speaker 1: basic plastic TV knobs. Terry Goa, who founded that, was 137 00:08:37,320 --> 00:08:39,760 Speaker 1: really a plastics guy. He wasn't really an electronics gay, 138 00:08:39,880 --> 00:08:42,800 Speaker 1: was a plastics guy. And so that was pretty much 139 00:08:42,840 --> 00:08:46,320 Speaker 1: Taiwan's DNA. Don't do the design work, don't do the branding, 140 00:08:46,360 --> 00:08:49,280 Speaker 1: don't do any of that. Just hey America, Hey Japan, 141 00:08:49,400 --> 00:08:52,679 Speaker 1: Hey you're bring us your stuff. Your designs will make 142 00:08:52,720 --> 00:08:55,920 Speaker 1: it for you cheaper, quicker, and all of that, and 143 00:08:55,920 --> 00:08:59,600 Speaker 1: and so semiconductors at the time were the next thing 144 00:08:59,679 --> 00:09:02,280 Speaker 1: and area, and you know it was. It was also 145 00:09:02,320 --> 00:09:05,120 Speaker 1: in a similar period of time that what is now 146 00:09:05,160 --> 00:09:09,280 Speaker 1: known as ASA started doing I B MR PC clones. 147 00:09:09,679 --> 00:09:12,440 Speaker 1: Uh and we'll actually manufacturing for IBM. So it's the 148 00:09:12,480 --> 00:09:14,920 Speaker 1: same kind of period as it was a little bit later, 149 00:09:14,920 --> 00:09:18,559 Speaker 1: but it's the same general generation of Taiwan. But really 150 00:09:18,559 --> 00:09:22,560 Speaker 1: the founding idea came from Mauris Young himself, who who 151 00:09:22,640 --> 00:09:25,240 Speaker 1: went to Taiwan at the age of the ripe old 152 00:09:25,240 --> 00:09:27,560 Speaker 1: age of fifty four to take up a retirement job. 153 00:09:28,520 --> 00:09:30,920 Speaker 1: It's not well known, but in fact Marris Young is 154 00:09:30,960 --> 00:09:33,840 Speaker 1: not Taiwanese. He was born in China and left China 155 00:09:34,240 --> 00:09:36,800 Speaker 1: at the height of the Civil War in night. He 156 00:09:36,800 --> 00:09:38,600 Speaker 1: was in Hong Kong for a brief stint and then 157 00:09:38,640 --> 00:09:41,440 Speaker 1: found himself in the US and went on to study 158 00:09:41,600 --> 00:09:43,200 Speaker 1: M I. T. And he spent a quarter of a 159 00:09:43,240 --> 00:09:46,360 Speaker 1: century at t I Texas Instruments, and it was only 160 00:09:46,400 --> 00:09:50,160 Speaker 1: after leaving Texas Instruments that the Taiwan government invited him 161 00:09:50,160 --> 00:09:53,319 Speaker 1: to Taiwan to lead it TRY. He wasn't the founding 162 00:09:53,400 --> 00:09:55,599 Speaker 1: chairman of it TRY, but he was I think that 163 00:09:55,800 --> 00:09:59,720 Speaker 1: the third president of Vitry and UH and he came 164 00:09:59,760 --> 00:10:03,079 Speaker 1: in and was brought to Taiwan not as a Taiwanese 165 00:10:03,120 --> 00:10:07,160 Speaker 1: but as actually an American citizen, and Uh and Morris 166 00:10:07,240 --> 00:10:11,520 Speaker 1: Jung got the idea of of doing a foundry model 167 00:10:11,640 --> 00:10:15,120 Speaker 1: and that was really the start of the whole process. 168 00:10:15,160 --> 00:10:18,720 Speaker 1: So how difficult was it to get the foundry model 169 00:10:18,880 --> 00:10:21,160 Speaker 1: off the ground at the very beginning, because as you 170 00:10:21,240 --> 00:10:27,880 Speaker 1: point out, that wasn't how people had been making semiconductors historically. 171 00:10:27,920 --> 00:10:30,959 Speaker 1: They've been doing the designing and manufacturing in house. How 172 00:10:31,040 --> 00:10:35,560 Speaker 1: much of a adaptation or change was it for the industry, 173 00:10:35,600 --> 00:10:37,760 Speaker 1: And how difficult was it to get clients on board 174 00:10:37,800 --> 00:10:39,480 Speaker 1: to actually go to them and say, hey, we're going 175 00:10:39,520 --> 00:10:42,480 Speaker 1: to make your chips for you. It was. It was 176 00:10:42,640 --> 00:10:45,199 Speaker 1: very difficult. And Morris Young recognized that, you know, he'd 177 00:10:45,200 --> 00:10:47,640 Speaker 1: worked in the American semi conductor industry for twenty or 178 00:10:47,679 --> 00:10:51,400 Speaker 1: thirty years. He knew that they were his people. Uh. 179 00:10:51,720 --> 00:10:54,800 Speaker 1: And so when he started tiss him see, one thing 180 00:10:54,840 --> 00:10:57,000 Speaker 1: that he knew early early on is a lot he 181 00:10:57,120 --> 00:11:00,679 Speaker 1: had Taiwan government backing, and there was you know, very pockets. 182 00:11:01,120 --> 00:11:04,320 Speaker 1: He knew that he needed to get industry on board, 183 00:11:04,440 --> 00:11:07,680 Speaker 1: so he went hunting for an industry player that would 184 00:11:07,679 --> 00:11:11,200 Speaker 1: be a cornerstone investor. He wanted to get someone on 185 00:11:11,280 --> 00:11:15,080 Speaker 1: board that would take early steak in the company so 186 00:11:15,120 --> 00:11:18,079 Speaker 1: that he could really make sure that they they were 187 00:11:18,160 --> 00:11:20,280 Speaker 1: vested in the country. It was in the company. It 188 00:11:20,360 --> 00:11:23,199 Speaker 1: was a vested interest to have TSMC succeeds. So he 189 00:11:23,240 --> 00:11:27,080 Speaker 1: shopped it around to the Americans and they really wouldn't 190 00:11:27,400 --> 00:11:30,840 Speaker 1: they wouldn't buy it. And eventually he stumbled across wild 191 00:11:30,920 --> 00:11:32,920 Speaker 1: and stumble across, but he came across an investor that 192 00:11:33,080 --> 00:11:35,000 Speaker 1: was willing, that really did see the vision it was 193 00:11:35,000 --> 00:11:37,600 Speaker 1: willing to take a punt, and that was Phillips, the 194 00:11:37,720 --> 00:11:41,880 Speaker 1: Dutch company which was for a while in the semiconductor industry. 195 00:11:42,040 --> 00:11:45,280 Speaker 1: And so Phillips was actually a founding investor and had 196 00:11:45,320 --> 00:11:47,839 Speaker 1: I think it was about thirty percent steak and they 197 00:11:47,880 --> 00:11:51,520 Speaker 1: had that thirty percent steak for for I think almost 198 00:11:51,520 --> 00:11:54,199 Speaker 1: twenty years until they started selling it down when they 199 00:11:54,200 --> 00:11:57,920 Speaker 1: did their own kind of revamp in the early two thousand's, 200 00:11:57,960 --> 00:12:00,640 Speaker 1: So that really really helped it show that you had 201 00:12:00,679 --> 00:12:03,439 Speaker 1: someone in the industry on board and he had a 202 00:12:03,480 --> 00:12:06,080 Speaker 1: certain amount of gravitas, and I think that was the 203 00:12:06,160 --> 00:12:08,240 Speaker 1: key thing for for Lawrence Jung. And the other thing 204 00:12:08,280 --> 00:12:10,719 Speaker 1: was that he was American. You know, he grabbed it 205 00:12:10,880 --> 00:12:14,120 Speaker 1: through the ranks of Texas Instruments. He was a general 206 00:12:14,160 --> 00:12:16,200 Speaker 1: instruments for a brief time before he went to Taiwan. 207 00:12:16,360 --> 00:12:18,960 Speaker 1: So the industry in America knew who he was. They 208 00:12:19,040 --> 00:12:22,240 Speaker 1: knew he was the real deal. There wasn't some unknown 209 00:12:22,400 --> 00:12:25,360 Speaker 1: coming out from a far off land pitching a crazy idea. 210 00:12:26,000 --> 00:12:28,800 Speaker 1: And so I think those two things together helped him 211 00:12:28,840 --> 00:12:31,560 Speaker 1: get traction with the early clients who were willing to 212 00:12:31,559 --> 00:12:34,800 Speaker 1: take a punt, and they did so and they were 213 00:12:34,840 --> 00:12:54,360 Speaker 1: rewarded with it. So I want to just it'll go 214 00:12:54,400 --> 00:12:56,640 Speaker 1: back to that founding a little bit more. I mean, 215 00:12:56,679 --> 00:13:00,760 Speaker 1: you mentioned that prior to t s MC that the 216 00:13:00,760 --> 00:13:03,920 Speaker 1: Taiwan government for already several years had been sort of 217 00:13:03,960 --> 00:13:08,240 Speaker 1: like researching and building up this capability. Were there other 218 00:13:09,120 --> 00:13:12,040 Speaker 1: players who could have been t SMC where there are 219 00:13:12,080 --> 00:13:14,920 Speaker 1: other entities trying to viy to be sort of like 220 00:13:15,000 --> 00:13:19,120 Speaker 1: a domestic national champion. And also, you know, like I've 221 00:13:19,120 --> 00:13:23,120 Speaker 1: read a little bit about Asian development models, like particularly 222 00:13:23,160 --> 00:13:28,040 Speaker 1: in Korea, and this idea of exports subsidies for companies 223 00:13:28,080 --> 00:13:32,439 Speaker 1: that were um finding traction with sales outside, and I'm curious, 224 00:13:32,440 --> 00:13:37,319 Speaker 1: like how much that effort to attract that foreign investor 225 00:13:37,600 --> 00:13:42,040 Speaker 1: Phillips in that case was also about proving themselves domestically 226 00:13:42,200 --> 00:13:46,520 Speaker 1: that t SMC could be the international player that the 227 00:13:46,640 --> 00:13:50,559 Speaker 1: country should get behind. Yeah, I think that's a valid point. 228 00:13:50,640 --> 00:13:54,000 Speaker 1: I think that they needed to well, they needed to 229 00:13:54,040 --> 00:13:55,959 Speaker 1: sell up the masses. But you know, Taiwan was not 230 00:13:56,120 --> 00:13:58,440 Speaker 1: democratic at the time. It was under martial law, so 231 00:13:58,440 --> 00:14:00,880 Speaker 1: it's not like they had to the government of the 232 00:14:00,920 --> 00:14:03,160 Speaker 1: Guaming Dung at the time really needed to worry about 233 00:14:03,160 --> 00:14:06,520 Speaker 1: elections or or anything like that. But they did very 234 00:14:06,600 --> 00:14:10,000 Speaker 1: much have concerns about the Communist Party on the other 235 00:14:10,000 --> 00:14:13,880 Speaker 1: side of the strait that that tension continues to this day. 236 00:14:13,960 --> 00:14:17,520 Speaker 1: And so there was definitely a feeling that Taiwana had 237 00:14:17,520 --> 00:14:20,680 Speaker 1: to be big, strong place. You know, it's called Republic 238 00:14:20,680 --> 00:14:22,800 Speaker 1: of China, and it needed to be strong and powerful 239 00:14:22,800 --> 00:14:25,840 Speaker 1: and make money so that it could fend off the 240 00:14:25,880 --> 00:14:31,720 Speaker 1: communist insurgency. And so these industrial policies were in a 241 00:14:31,840 --> 00:14:35,440 Speaker 1: kind of war mentality, not unlike what Career went through. 242 00:14:35,480 --> 00:14:37,160 Speaker 1: You know, they went through their own civil war and 243 00:14:37,200 --> 00:14:40,440 Speaker 1: had a kind of war mentality after the Korean War ended, 244 00:14:40,480 --> 00:14:43,760 Speaker 1: and had the tensions with between North and South, and 245 00:14:43,800 --> 00:14:48,520 Speaker 1: of course Japain was rebuilding after its own war with 246 00:14:48,640 --> 00:14:51,480 Speaker 1: America and much of Asia. So there was a war 247 00:14:51,600 --> 00:14:55,240 Speaker 1: mentality for a lot of these countries. And that really 248 00:14:55,280 --> 00:15:00,000 Speaker 1: I think directed industrial policy and without having to worry 249 00:15:00,000 --> 00:15:02,720 Speaker 1: about democracy, to put it bluntly, they could go in 250 00:15:02,720 --> 00:15:07,160 Speaker 1: and basically make decisions, make decisions for the long term 251 00:15:07,280 --> 00:15:10,120 Speaker 1: and and really laid down what they were going to 252 00:15:10,160 --> 00:15:13,000 Speaker 1: do for the next ten or twenty years. And so 253 00:15:13,040 --> 00:15:16,160 Speaker 1: I think that getting the early investor on board in 254 00:15:16,240 --> 00:15:19,880 Speaker 1: Phillips was very, very important. But to your earlier point, 255 00:15:20,800 --> 00:15:24,920 Speaker 1: t SMC wasn't the earliest one. In fact, born out 256 00:15:24,960 --> 00:15:29,320 Speaker 1: of it try Itchery was started, you know, in the seventies, 257 00:15:29,680 --> 00:15:34,040 Speaker 1: but born out of Itchry in around eighty was another 258 00:15:34,080 --> 00:15:38,400 Speaker 1: company called u MC United micro Electronics, and that company 259 00:15:38,480 --> 00:15:40,160 Speaker 1: was spun out of Itchory, and that was in the 260 00:15:40,200 --> 00:15:43,560 Speaker 1: semiconductor industry, but it was also designing its own chips, 261 00:15:44,360 --> 00:15:47,440 Speaker 1: and in fact that then spun out of UMC in 262 00:15:47,480 --> 00:15:50,320 Speaker 1: the late nineties was a company you called Media Tech, 263 00:15:50,400 --> 00:15:52,400 Speaker 1: which some people might have heard of because they're actually 264 00:15:52,480 --> 00:15:56,160 Speaker 1: quite a competitor to Intel and Core Common others today, 265 00:15:56,400 --> 00:15:59,880 Speaker 1: they did have other competitors. It wasn't given that tear 266 00:16:00,120 --> 00:16:04,000 Speaker 1: MC would win. They had competitor, a very strong competitor 267 00:16:04,040 --> 00:16:06,600 Speaker 1: in UMC, and at the time there was also competitors 268 00:16:06,600 --> 00:16:09,720 Speaker 1: overseas in Singapore there was a company called Charted there 269 00:16:09,800 --> 00:16:13,200 Speaker 1: was also looking quite strong for a while until until 270 00:16:13,240 --> 00:16:16,000 Speaker 1: they just couldn't keep up with t SMC. So t 271 00:16:16,160 --> 00:16:19,960 Speaker 1: SMC had competition in the early days in foundries, but 272 00:16:20,040 --> 00:16:21,880 Speaker 1: they were the ones who really stuck to date guns 273 00:16:21,880 --> 00:16:26,640 Speaker 1: and managed to win out. So just on that winning point, 274 00:16:26,760 --> 00:16:30,200 Speaker 1: there's one other thing from what I understand, that Chung 275 00:16:30,240 --> 00:16:34,480 Speaker 1: actually did that helped t s mc um get itself 276 00:16:34,560 --> 00:16:37,280 Speaker 1: into first place in the long run, and that was 277 00:16:37,720 --> 00:16:43,280 Speaker 1: that they sort of sacrifice profits early on in order 278 00:16:43,360 --> 00:16:47,160 Speaker 1: to grow their market share very quickly, which is something 279 00:16:47,200 --> 00:16:50,280 Speaker 1: that we've seen in other big tech companies, I mean, 280 00:16:50,440 --> 00:16:54,440 Speaker 1: notably Amazon and others like that. How was that decision 281 00:16:54,440 --> 00:16:57,520 Speaker 1: made and how important was it to t s m 282 00:16:57,600 --> 00:17:01,280 Speaker 1: C s a long term success. Well, the key really 283 00:17:01,280 --> 00:17:03,760 Speaker 1: for for t SMC cost base it really comes down 284 00:17:03,760 --> 00:17:05,960 Speaker 1: to what is the cost right And so of course, 285 00:17:06,119 --> 00:17:09,639 Speaker 1: if you if you're trying to win over customers, but 286 00:17:09,680 --> 00:17:12,040 Speaker 1: you're offering a price that is not acceptable or not 287 00:17:12,080 --> 00:17:14,439 Speaker 1: competitive to them doing it themselves, then you're not going 288 00:17:14,520 --> 00:17:17,399 Speaker 1: to win. And the cost based for for the t 289 00:17:17,560 --> 00:17:22,359 Speaker 1: SMC model is it's mostly depreciation. You spend buckets of money, 290 00:17:22,440 --> 00:17:25,120 Speaker 1: you know, a ten or twenty billion dollars a year 291 00:17:25,520 --> 00:17:29,119 Speaker 1: at at the current rate, and that depreciates pretty quickly, 292 00:17:29,160 --> 00:17:31,000 Speaker 1: to be honest, in only a few years. So your 293 00:17:31,040 --> 00:17:33,520 Speaker 1: depreciation is very very high. And once you've built that 294 00:17:33,600 --> 00:17:36,440 Speaker 1: in the border the equipment, if it sits idle, then 295 00:17:36,600 --> 00:17:40,040 Speaker 1: you're not making money, right, So it's it's immediate cost 296 00:17:40,040 --> 00:17:41,480 Speaker 1: off the bat. The other one is R and D, 297 00:17:41,680 --> 00:17:45,080 Speaker 1: and that's where they really haven't skipped. They've been willing 298 00:17:45,080 --> 00:17:47,000 Speaker 1: to spend buckets of money on R and D, which 299 00:17:47,000 --> 00:17:48,720 Speaker 1: is engineers. At the end of the day, it's it's 300 00:17:48,760 --> 00:17:52,160 Speaker 1: the best engineers money can buy. And so those two 301 00:17:52,200 --> 00:17:54,560 Speaker 1: things they basically had to take it they reallyn't have 302 00:17:54,560 --> 00:17:57,840 Speaker 1: a choice. They had to spend money to build the 303 00:17:57,920 --> 00:18:01,040 Speaker 1: capacity and get the technology going, and they needed to 304 00:18:01,080 --> 00:18:04,359 Speaker 1: spend money on the staff the engineers to to make 305 00:18:04,440 --> 00:18:08,040 Speaker 1: the best best they could of this equipment. And then 306 00:18:08,119 --> 00:18:10,040 Speaker 1: on the flip side, when they're going to their customers 307 00:18:10,080 --> 00:18:12,159 Speaker 1: and saying, what we want to charge you this much, well, 308 00:18:12,800 --> 00:18:15,359 Speaker 1: you have to be price competitive. So in a way 309 00:18:15,359 --> 00:18:18,119 Speaker 1: it was it was a deliberate tactic, but also it 310 00:18:18,200 --> 00:18:21,720 Speaker 1: was also an issue of the circumstances of the time. 311 00:18:23,400 --> 00:18:28,000 Speaker 1: So when we think about like contract manufacturing in any industry, 312 00:18:28,000 --> 00:18:32,000 Speaker 1: and of course the basic model as you identified, isn't 313 00:18:32,000 --> 00:18:34,320 Speaker 1: it's not new. It's not new to chip people. They 314 00:18:34,320 --> 00:18:37,600 Speaker 1: have always been set up to where one company designed something, 315 00:18:37,720 --> 00:18:41,879 Speaker 1: another country does the building. I think for a long time, 316 00:18:42,280 --> 00:18:45,480 Speaker 1: the business of and you can correct me if I'm wrong, 317 00:18:45,480 --> 00:18:48,600 Speaker 1: but the business of sort of being a fab I mean, 318 00:18:48,600 --> 00:18:51,880 Speaker 1: I don't think it was seen as particularly particularly sexy. 319 00:18:52,040 --> 00:18:54,960 Speaker 1: It may have been a profitable business model, an important one, 320 00:18:55,600 --> 00:18:58,800 Speaker 1: but it seems like in the discussions we're having lately, 321 00:18:58,880 --> 00:19:02,600 Speaker 1: like something like it went from being like, okay, here's 322 00:19:02,640 --> 00:19:08,000 Speaker 1: the here's the company that takes in your designs and 323 00:19:08,040 --> 00:19:11,160 Speaker 1: then spits them back out to here's a company that's 324 00:19:11,160 --> 00:19:14,639 Speaker 1: like a true tech leader, like they really have something 325 00:19:14,680 --> 00:19:18,480 Speaker 1: in their manufacturing, know how that is increasingly difficult to 326 00:19:18,520 --> 00:19:23,200 Speaker 1: be replicated. Was there some inflection point, some flip where 327 00:19:23,240 --> 00:19:25,560 Speaker 1: like when did that happen? When did that sort of 328 00:19:25,640 --> 00:19:29,560 Speaker 1: change of perception start to occur? I would argue that 329 00:19:29,600 --> 00:19:33,120 Speaker 1: it happened during the period of the crossover from PCs 330 00:19:33,160 --> 00:19:37,440 Speaker 1: to smartphones, because, as we know, PCs were Intel's domain 331 00:19:37,560 --> 00:19:39,960 Speaker 1: and you know, am D's domain and turned up to 332 00:19:39,960 --> 00:19:42,960 Speaker 1: the party and they made their own stuff. Am D 333 00:19:43,119 --> 00:19:45,800 Speaker 1: doesn't anymore, but for the longest time they did make 334 00:19:45,840 --> 00:19:47,720 Speaker 1: their own their own stuff. They designed it and made 335 00:19:47,720 --> 00:19:50,359 Speaker 1: it themselves. They call them I, D M, right. And 336 00:19:50,400 --> 00:19:53,639 Speaker 1: then we went into the smartphone business. There was a 337 00:19:53,640 --> 00:19:55,280 Speaker 1: short period of time I you know, the two G 338 00:19:55,520 --> 00:19:58,480 Speaker 1: phones that you know, the Nokias and so forth. But 339 00:19:58,480 --> 00:20:03,199 Speaker 1: then we moved into smartphones and those were powered basically 340 00:20:03,240 --> 00:20:06,119 Speaker 1: by by companies like Qualcom and Media Tech and and 341 00:20:06,119 --> 00:20:08,159 Speaker 1: actually t I if you might remember, we're in that 342 00:20:08,200 --> 00:20:11,000 Speaker 1: business for a little while. And these were companies that 343 00:20:11,040 --> 00:20:13,440 Speaker 1: didn't have their own factories. Well t I did, but 344 00:20:13,720 --> 00:20:17,400 Speaker 1: Qualcom certainly didn't. And media tech in and a lot 345 00:20:17,400 --> 00:20:20,840 Speaker 1: of companies didn't have their own their own factories. They 346 00:20:20,880 --> 00:20:22,800 Speaker 1: because they didn't need to have their own factories, a 347 00:20:22,840 --> 00:20:26,240 Speaker 1: company like Caulcom could start up. Anybody could get together 348 00:20:26,280 --> 00:20:29,359 Speaker 1: a few friends fresh out of grad school and try 349 00:20:29,400 --> 00:20:33,080 Speaker 1: and start a chip design company, and Calcom was able 350 00:20:33,119 --> 00:20:35,960 Speaker 1: to get away with doing that and and was very successful. 351 00:20:36,080 --> 00:20:39,320 Speaker 1: So you have a company like Qualcom becoming very very 352 00:20:39,320 --> 00:20:41,360 Speaker 1: successful by saying, you know what, we designed the best 353 00:20:41,400 --> 00:20:45,359 Speaker 1: chips for for mobile phones, and we don't need to 354 00:20:45,400 --> 00:20:47,880 Speaker 1: manufacture them. And I think the rest of the industry 355 00:20:47,920 --> 00:20:51,960 Speaker 1: cottoned on. And as we're moving away from the PC 356 00:20:52,200 --> 00:20:55,359 Speaker 1: era into the smartphone era and then electronic devices and 357 00:20:55,440 --> 00:20:59,840 Speaker 1: we've got PlayStations and handhelds like the switch, and also 358 00:21:00,040 --> 00:21:02,800 Speaker 1: to other devices, a lot of these chips were no 359 00:21:02,880 --> 00:21:07,840 Speaker 1: longer the big hunky uh you know, Intel processes that 360 00:21:07,920 --> 00:21:10,960 Speaker 1: were smaller and lighter. And then of course Apple came 361 00:21:10,960 --> 00:21:14,840 Speaker 1: along and decided to do their own ships as well. 362 00:21:14,880 --> 00:21:16,879 Speaker 1: And so I think that's really where we saw the 363 00:21:16,960 --> 00:21:20,640 Speaker 1: flip from having to have your own factory to being 364 00:21:20,680 --> 00:21:22,639 Speaker 1: able to do it without. And at the same period 365 00:21:22,680 --> 00:21:25,160 Speaker 1: of time, the economics just wasn't working out for companies 366 00:21:25,200 --> 00:21:27,280 Speaker 1: like a m D anymore, and a lot of companies 367 00:21:27,320 --> 00:21:30,720 Speaker 1: who were making their own were being burdened and saddled 368 00:21:30,720 --> 00:21:33,800 Speaker 1: by these massive factories and the depreciation costs they had 369 00:21:33,840 --> 00:21:36,119 Speaker 1: to bear. No matter whether they sold one ship or 370 00:21:36,119 --> 00:21:38,720 Speaker 1: a million chips, that appreciation was going to be the same. 371 00:21:39,440 --> 00:21:42,439 Speaker 1: And so if you have a bad year or two, 372 00:21:42,520 --> 00:21:44,840 Speaker 1: or there is a shift away from your business such 373 00:21:44,880 --> 00:21:47,320 Speaker 1: as PCs to something else, and you don't have the 374 00:21:47,400 --> 00:21:49,480 Speaker 1: chips to keep up, then you're going to be in 375 00:21:49,520 --> 00:21:51,399 Speaker 1: debt very very quickly. And so we saw a lot 376 00:21:51,400 --> 00:21:54,480 Speaker 1: of the companies that were designing and manufacturing for themselves 377 00:21:54,840 --> 00:21:58,200 Speaker 1: just couldn't keep up economically or financially, and some went 378 00:21:58,240 --> 00:22:02,160 Speaker 1: fabulous like a m D, and some just disappeared. Can 379 00:22:02,200 --> 00:22:05,119 Speaker 1: we talk a little bit about the role of the 380 00:22:05,240 --> 00:22:07,600 Speaker 1: US and China in all of it? So we mentioned 381 00:22:07,600 --> 00:22:12,520 Speaker 1: that Taiwan itself is sort of perennially caught between these 382 00:22:12,560 --> 00:22:17,000 Speaker 1: two forces, the US and China, which doesn't actually recognize 383 00:22:17,000 --> 00:22:22,360 Speaker 1: it as an independent country. What influence has that relationship 384 00:22:22,520 --> 00:22:28,000 Speaker 1: had on TSMC and how does TSMC feedback into that 385 00:22:28,119 --> 00:22:32,119 Speaker 1: geopolitical relationship. T S m c s view of the 386 00:22:32,160 --> 00:22:36,919 Speaker 1: geopolitical relationship is quite smart and diplomatic in almost the 387 00:22:36,960 --> 00:22:39,800 Speaker 1: same way that they view their customers, where they want 388 00:22:39,840 --> 00:22:42,800 Speaker 1: to be the neutral supplier to everybody. They want to 389 00:22:42,800 --> 00:22:45,560 Speaker 1: be everybody's foundry. They don't want to be everybody's partner. 390 00:22:45,560 --> 00:22:48,840 Speaker 1: And they take that view to the broader the global 391 00:22:49,000 --> 00:22:53,840 Speaker 1: geopolitical situation by saying, well, you know, we're not in 392 00:22:53,880 --> 00:22:56,200 Speaker 1: the American camp, we're not in the China camp. We're 393 00:22:56,240 --> 00:22:59,520 Speaker 1: just in the camp of whichever client's needs. And that's 394 00:22:59,640 --> 00:23:01,600 Speaker 1: the lie and that they have been taking for the 395 00:23:01,640 --> 00:23:06,240 Speaker 1: longest time. Now, what's interesting is I think people misunderstand 396 00:23:06,280 --> 00:23:10,040 Speaker 1: t SMC because Taiwan and t SMC is so close 397 00:23:10,119 --> 00:23:13,399 Speaker 1: to China, and of course many devices are actually assembled 398 00:23:13,400 --> 00:23:16,639 Speaker 1: in China. Most most electronic devices are assembled in China. 399 00:23:16,960 --> 00:23:19,000 Speaker 1: People see it as being, oh, must be kind of 400 00:23:19,040 --> 00:23:23,119 Speaker 1: close to China, But in fact the history in the 401 00:23:23,200 --> 00:23:26,119 Speaker 1: genesis of t SMC is far more American than it 402 00:23:26,240 --> 00:23:28,919 Speaker 1: is Chinese. Morris the two people who have held the 403 00:23:29,000 --> 00:23:33,960 Speaker 1: chairmanship of TSMC both American citizens, Morris Young, the first 404 00:23:34,000 --> 00:23:37,960 Speaker 1: one was not even Taiwanese, and the current one, Mark 405 00:23:38,080 --> 00:23:40,720 Speaker 1: Leo is if you look at the annual report of 406 00:23:40,760 --> 00:23:43,640 Speaker 1: t SMC, it say cities American citizen. I suspect he's 407 00:23:43,640 --> 00:23:46,359 Speaker 1: a dual national aspect, he's got taiwan citizenship as well, 408 00:23:46,920 --> 00:23:50,240 Speaker 1: But in actual fact they're both Americans. The style and 409 00:23:50,280 --> 00:23:53,760 Speaker 1: the culture and the processes of the company are very 410 00:23:53,880 --> 00:23:57,800 Speaker 1: very American. The corporate governance approach is very very American. 411 00:23:57,840 --> 00:24:00,080 Speaker 1: It's very very different to the way most Taiwanese or 412 00:24:00,080 --> 00:24:03,680 Speaker 1: Asian companies are run. And so that culture, of course 413 00:24:03,760 --> 00:24:07,000 Speaker 1: is very important. And the clients for the longest time 414 00:24:07,080 --> 00:24:10,240 Speaker 1: have been American. Yes, the devices get assembled in China, 415 00:24:10,720 --> 00:24:14,080 Speaker 1: but all those big name semi conducted clients, the ones 416 00:24:14,119 --> 00:24:18,000 Speaker 1: that pay the bills, they're all American. And so the 417 00:24:18,080 --> 00:24:20,760 Speaker 1: relationship is much much closer to America than it is 418 00:24:20,800 --> 00:24:23,639 Speaker 1: to China. Even though China of course is a large 419 00:24:23,720 --> 00:24:25,960 Speaker 1: buyer of chips now and of course it's the place 420 00:24:26,000 --> 00:24:31,000 Speaker 1: where so many semi conductors are then installed inside devices. 421 00:24:31,040 --> 00:24:34,160 Speaker 1: The relationship is much much closer to America than China. 422 00:24:34,440 --> 00:24:38,080 Speaker 1: Of course, it may change in the future. Uh, And 423 00:24:38,200 --> 00:24:40,760 Speaker 1: both sides want to kind of drag t SMC to 424 00:24:40,800 --> 00:24:43,400 Speaker 1: their side. But you know what, as of right now, 425 00:24:43,440 --> 00:24:50,160 Speaker 1: you know, January, like cent or probably of the production 426 00:24:50,160 --> 00:24:54,480 Speaker 1: capacity is in Taiwan. They've got pretty small production facilities 427 00:24:55,240 --> 00:24:57,280 Speaker 1: or capacity in China, and of course they're going to 428 00:24:57,320 --> 00:24:59,960 Speaker 1: do something in Arizona which would be at a similar 429 00:25:00,200 --> 00:25:03,359 Speaker 1: small scale, but Taiwan still is the hard and soul 430 00:25:03,400 --> 00:25:07,280 Speaker 1: of t SMC. So just on that note, on the 431 00:25:07,320 --> 00:25:11,159 Speaker 1: idea of t SMC being quite smart in maintaining a 432 00:25:11,280 --> 00:25:14,399 Speaker 1: neutral stance or trying to maintain a neutral stance between 433 00:25:14,440 --> 00:25:18,480 Speaker 1: the US and China, how long do you think they 434 00:25:18,480 --> 00:25:24,240 Speaker 1: can keep doing that given that semiconductors seem to be becoming, 435 00:25:24,400 --> 00:25:29,399 Speaker 1: you know, rapidly politicized, and tech in general is becoming politicized. 436 00:25:29,440 --> 00:25:32,240 Speaker 1: So we just saw, well over the past couple of years, 437 00:25:32,280 --> 00:25:37,120 Speaker 1: we've seen the US announced all these restrictions on semiconductor technology. 438 00:25:37,440 --> 00:25:40,120 Speaker 1: In the past week or so, again we're recording this 439 00:25:40,280 --> 00:25:44,320 Speaker 1: on January twelve, we've seen the US announced fresh sanctions 440 00:25:44,320 --> 00:25:48,480 Speaker 1: on Chinese military firms, and meanwhile China is saying that 441 00:25:48,560 --> 00:25:53,080 Speaker 1: it is going to allow people to sue international companies 442 00:25:53,119 --> 00:25:57,359 Speaker 1: if they comply with U S sanctions. It feels like 443 00:25:58,320 --> 00:26:01,639 Speaker 1: remaining neutral is beca mean more and more difficult in 444 00:26:01,680 --> 00:26:04,359 Speaker 1: the current environment. Of course, everything could change when the 445 00:26:04,400 --> 00:26:07,520 Speaker 1: new US administration comes in, but in the meantime, it 446 00:26:07,840 --> 00:26:10,440 Speaker 1: does feel like companies are going to be forced to 447 00:26:10,520 --> 00:26:13,320 Speaker 1: choose sites. Yeah, Tracy, I think I think that's the 448 00:26:13,400 --> 00:26:15,280 Speaker 1: key issue, and I think that's one of the most 449 00:26:15,320 --> 00:26:18,280 Speaker 1: important things that we need to consider in this this 450 00:26:18,359 --> 00:26:21,320 Speaker 1: whole situation. In fact, I've been arguing for for more 451 00:26:21,359 --> 00:26:24,480 Speaker 1: than five years that there's this Berlin Wall of tech 452 00:26:24,840 --> 00:26:29,600 Speaker 1: brewing between America and China, or American and Chinese fheres 453 00:26:29,640 --> 00:26:33,320 Speaker 1: of influence. Even before Donald Trump came to power, that 454 00:26:33,440 --> 00:26:36,320 Speaker 1: was already going to happen. Now, Trump clearly accelerated a 455 00:26:36,320 --> 00:26:39,240 Speaker 1: lot of those moves, but it was already underway. T 456 00:26:39,480 --> 00:26:41,719 Speaker 1: s MC wants to sit on that fence. They've been 457 00:26:41,840 --> 00:26:43,840 Speaker 1: very good at sitting on that fence. But I've argued 458 00:26:43,840 --> 00:26:46,119 Speaker 1: in columns of the last couple of years that they 459 00:26:46,119 --> 00:26:48,280 Speaker 1: can't do it forever. They are going to have to 460 00:26:48,400 --> 00:26:51,200 Speaker 1: choose sites. Now. I've got a lot of pushback from 461 00:26:51,440 --> 00:26:54,120 Speaker 1: people at t SMC and others in the industry saying no, no, 462 00:26:54,119 --> 00:26:56,879 Speaker 1: no where neutral, Always be neutral. We can manage to 463 00:26:56,880 --> 00:26:59,680 Speaker 1: be neutral forever, and I just don't agree. I think 464 00:26:59,720 --> 00:27:02,159 Speaker 1: that we saw that with the Huaowei situation over the 465 00:27:02,200 --> 00:27:04,840 Speaker 1: last couple of years when at one point the US 466 00:27:04,880 --> 00:27:08,679 Speaker 1: basically said almost directly to t SEMC without naming, like, 467 00:27:08,800 --> 00:27:11,360 Speaker 1: you shall not make chips for Hawaiwei anymore, and they 468 00:27:11,400 --> 00:27:13,960 Speaker 1: had to stop write a ground to a halt. Now 469 00:27:14,000 --> 00:27:17,240 Speaker 1: that may open up again under a new administration in 470 00:27:17,680 --> 00:27:21,399 Speaker 1: d C, but that was really should have been a 471 00:27:21,440 --> 00:27:23,679 Speaker 1: wake up call to management at t SMC that you 472 00:27:23,720 --> 00:27:27,119 Speaker 1: can't sit on the fence any longer. And I think 473 00:27:27,160 --> 00:27:30,400 Speaker 1: that they are. They're realizing now they've they've got they've 474 00:27:30,480 --> 00:27:34,240 Speaker 1: hired some new lobbyists a former Intel lobbyist amongst others 475 00:27:35,000 --> 00:27:37,640 Speaker 1: in d C to help their cause. I think they 476 00:27:37,680 --> 00:27:39,920 Speaker 1: now realize they can't sit on that fence forever, but 477 00:27:39,960 --> 00:27:42,119 Speaker 1: they're going to try to hold on as long as 478 00:27:42,200 --> 00:27:44,440 Speaker 1: they can before they have to choose sides. And when 479 00:27:44,440 --> 00:27:47,400 Speaker 1: they do, they're going to choose the American side. That's 480 00:27:47,400 --> 00:27:49,520 Speaker 1: where their bread is buttered. It really seems like kind 481 00:27:49,560 --> 00:27:52,600 Speaker 1: of stressful though for the world, Like if literally you 482 00:27:52,720 --> 00:27:56,080 Speaker 1: can't come up with a piece of electronics that in 483 00:27:56,119 --> 00:27:59,760 Speaker 1: some way hasn't been touched by t SMC and they 484 00:28:00,000 --> 00:28:02,439 Speaker 1: were men can just say don't work for that company anymore. 485 00:28:02,480 --> 00:28:05,639 Speaker 1: And there's no alternative to t SMC anywhere near its scale. 486 00:28:05,680 --> 00:28:08,080 Speaker 1: Like it's like just a pretty it seems like a 487 00:28:08,119 --> 00:28:13,200 Speaker 1: pretty big sort of choke point for everything. Yeah, I mean, 488 00:28:13,359 --> 00:28:14,600 Speaker 1: if you if you were to go back to the 489 00:28:14,680 --> 00:28:17,600 Speaker 1: year two thousand when everyone is using Windows PCs, can 490 00:28:17,640 --> 00:28:20,040 Speaker 1: you imagine a government saying you can't use Windows anymore 491 00:28:20,680 --> 00:28:23,720 Speaker 1: and Android Sorry not Andrew, but Lennox wasn't really coming along. 492 00:28:23,760 --> 00:28:27,320 Speaker 1: You can you imagine that the problems and the trouble 493 00:28:27,400 --> 00:28:29,960 Speaker 1: that would happen, that would be the kind of equivalent 494 00:28:30,000 --> 00:28:32,440 Speaker 1: today if you would be told you can't use t SMC. 495 00:28:32,560 --> 00:28:34,880 Speaker 1: In Huawei struggled. They oh my god, the problems they 496 00:28:34,920 --> 00:28:37,960 Speaker 1: faced in not being able to use They're putting in 497 00:28:38,040 --> 00:28:40,720 Speaker 1: rush orders, rush orders to the last minute, so the 498 00:28:40,760 --> 00:28:43,360 Speaker 1: t SMC would churn stuff out and get it out 499 00:28:43,400 --> 00:28:47,040 Speaker 1: the door before the barriers came down. And that's really 500 00:28:47,080 --> 00:28:49,240 Speaker 1: caused a lot of problems for Huawei. And they don't 501 00:28:49,240 --> 00:29:08,800 Speaker 1: really have a backup source. So you know, one of 502 00:29:08,840 --> 00:29:11,480 Speaker 1: the one of the themes, one of the big things 503 00:29:11,520 --> 00:29:14,160 Speaker 1: we're talking about is this sort of like tech advantage 504 00:29:14,360 --> 00:29:19,040 Speaker 1: that they're building up such scale, such manufacturing know how, 505 00:29:19,240 --> 00:29:23,120 Speaker 1: ability to do it at an excellent quality at improving price. 506 00:29:23,640 --> 00:29:26,880 Speaker 1: Um Intel, as we know, stumbling with their own manufacturers. 507 00:29:26,920 --> 00:29:30,520 Speaker 1: In fact, their reports that Bloomberg ourselves have reported about 508 00:29:30,520 --> 00:29:35,440 Speaker 1: perhaps Intel perhaps outsourcing some of its manufacturing to t SMC. 509 00:29:36,360 --> 00:29:40,000 Speaker 1: Talk to us about the tech edge that the company 510 00:29:40,040 --> 00:29:43,320 Speaker 1: is building and how difficult it is for even one 511 00:29:43,360 --> 00:29:47,000 Speaker 1: of the largest competitors, let alone some of the smaller competitors, 512 00:29:47,240 --> 00:29:51,440 Speaker 1: to match them in terms of price and quality. Semiconductors 513 00:29:51,480 --> 00:29:54,000 Speaker 1: are really tough. They've always been tough, but they're getting 514 00:29:54,040 --> 00:29:57,080 Speaker 1: tougher and tougher because we're now at this You've heard 515 00:29:57,080 --> 00:29:59,480 Speaker 1: about More's law and the idea that we're getting to 516 00:29:59,520 --> 00:30:03,040 Speaker 1: the end of Moore's law or beyond Moore's law, but 517 00:30:03,080 --> 00:30:06,960 Speaker 1: we're getting some really really tough ends of semi conductors 518 00:30:07,040 --> 00:30:11,440 Speaker 1: are essentially a mix of chemical engineering with physics and 519 00:30:11,600 --> 00:30:15,480 Speaker 1: electronic engine electrical engineering, and they're getting to the point 520 00:30:15,560 --> 00:30:18,680 Speaker 1: now with this new EUV technology that there's very very 521 00:30:18,680 --> 00:30:22,400 Speaker 1: few companies that can afford to buy this EUV equipment 522 00:30:22,560 --> 00:30:25,760 Speaker 1: and be have the know how, including the staff and 523 00:30:25,800 --> 00:30:28,720 Speaker 1: the process is to run it. So even if someone 524 00:30:28,720 --> 00:30:31,600 Speaker 1: else was managing to get their hands on one or 525 00:30:31,600 --> 00:30:35,120 Speaker 1: two hundred million dollar piece of equipment that can do 526 00:30:35,240 --> 00:30:38,000 Speaker 1: the later stuff. There's not many people who could run it, 527 00:30:38,040 --> 00:30:43,520 Speaker 1: and so it becomes this this self fulfilling prophecy that 528 00:30:43,600 --> 00:30:46,600 Speaker 1: t SMC is ahead of everybody else, and because they've 529 00:30:46,640 --> 00:30:49,160 Speaker 1: heard of everyone else, they get all the best orders 530 00:30:49,240 --> 00:30:52,360 Speaker 1: and all the money, and they've got really nice, fat margins. 531 00:30:52,360 --> 00:30:55,239 Speaker 1: They've got one of the largest operating margins in the 532 00:30:55,280 --> 00:30:58,640 Speaker 1: tech sector, and nobody else can compete because they don't 533 00:30:58,640 --> 00:31:00,480 Speaker 1: have the money to keep up, which they don't have 534 00:31:00,480 --> 00:31:02,320 Speaker 1: the money to buy all the equipment, which is they 535 00:31:02,320 --> 00:31:04,400 Speaker 1: don't have the money to hire the best engineers and 536 00:31:04,400 --> 00:31:07,000 Speaker 1: so on and so forth. And we really are at 537 00:31:07,000 --> 00:31:11,720 Speaker 1: that point where t SMC a lot of people see 538 00:31:11,760 --> 00:31:14,560 Speaker 1: that they're unassailable. I mean, of course they can mess up, 539 00:31:14,560 --> 00:31:16,680 Speaker 1: and they probably will mess up, but we're now at 540 00:31:16,720 --> 00:31:19,560 Speaker 1: this point where money talks, and it's not just money, 541 00:31:19,560 --> 00:31:21,880 Speaker 1: it's that know how. It's the store of knowledge that 542 00:31:21,880 --> 00:31:24,880 Speaker 1: they've built up over thirty years that no one else 543 00:31:24,920 --> 00:31:27,760 Speaker 1: can get near, not even Intel. And when Intel stumbles, 544 00:31:27,760 --> 00:31:29,680 Speaker 1: you know that things are getting really difficult in the 545 00:31:29,760 --> 00:31:32,760 Speaker 1: chip industry. So what would a mess up actually look 546 00:31:32,840 --> 00:31:36,160 Speaker 1: like for t SMC or what would it take for 547 00:31:36,200 --> 00:31:39,440 Speaker 1: them to lose this top spot that they currently have 548 00:31:39,520 --> 00:31:43,000 Speaker 1: in the industry. Well, I think there's one thing they 549 00:31:43,040 --> 00:31:45,800 Speaker 1: could be losing the top spot, I guess to Samsung 550 00:31:46,040 --> 00:31:48,720 Speaker 1: is probably the only other company that might take over 551 00:31:48,800 --> 00:31:51,600 Speaker 1: from them, And Samsung is a very very strong competitor, 552 00:31:51,640 --> 00:31:53,200 Speaker 1: and they were neck and neck for a while, and 553 00:31:53,440 --> 00:31:55,920 Speaker 1: I think Samsung will come back again. They've got a 554 00:31:56,000 --> 00:31:58,040 Speaker 1: lot of money. They've got a lot of their own 555 00:31:58,200 --> 00:32:01,160 Speaker 1: orders in the nand and d RAM business, so that 556 00:32:01,240 --> 00:32:04,960 Speaker 1: really helps them push their own semiconductort manufacturing. But in 557 00:32:05,080 --> 00:32:07,520 Speaker 1: terms of t SMC, you know, having a bit of 558 00:32:07,520 --> 00:32:10,240 Speaker 1: a nose dive or any any four. I think there's 559 00:32:10,240 --> 00:32:14,200 Speaker 1: actually three areas that they could they could face problems. 560 00:32:15,080 --> 00:32:17,880 Speaker 1: The first one I think is that they could be 561 00:32:18,320 --> 00:32:22,640 Speaker 1: a victim of their own success in that they always 562 00:32:22,680 --> 00:32:28,240 Speaker 1: at the cutting edge with the most incredibly good foundry technology, 563 00:32:28,280 --> 00:32:31,680 Speaker 1: to the point where they're technology is so good that 564 00:32:32,120 --> 00:32:34,920 Speaker 1: nobody really needs it. Like if you think of the 565 00:32:35,280 --> 00:32:38,040 Speaker 1: CPU wars between Intel and a m D of twenty 566 00:32:38,120 --> 00:32:40,920 Speaker 1: years ago, it was the coolest thing to be the 567 00:32:40,920 --> 00:32:44,280 Speaker 1: one with the fastest clock speed, but today nobody cares 568 00:32:44,280 --> 00:32:46,840 Speaker 1: about clock speed, right, nobody really pays attention they maybe 569 00:32:46,880 --> 00:32:49,680 Speaker 1: they'd care about how many cares and a CPU. But 570 00:32:49,800 --> 00:32:53,440 Speaker 1: all that kind of chest thumping is gone in the 571 00:32:53,520 --> 00:32:57,720 Speaker 1: CPU industry. And I think we'll see a point in 572 00:32:57,760 --> 00:33:01,400 Speaker 1: the future where maybe it won't be the leading companies 573 00:33:01,440 --> 00:33:04,600 Speaker 1: like Apple or Hong Kom or Huawei, who say we 574 00:33:04,640 --> 00:33:07,560 Speaker 1: don't need the latest technology, but maybe one of their 575 00:33:07,600 --> 00:33:10,400 Speaker 1: top ten customers, maybe the tenth largest customers, is you 576 00:33:10,400 --> 00:33:12,520 Speaker 1: know what, we're going to sit out this this round 577 00:33:12,520 --> 00:33:16,120 Speaker 1: of new technology. The previous round is good enough for us, 578 00:33:16,200 --> 00:33:19,640 Speaker 1: you know the the N minus too, which is you know, 579 00:33:19,800 --> 00:33:22,920 Speaker 1: two generations old. That's good enough for now. We'll we'll 580 00:33:22,920 --> 00:33:25,320 Speaker 1: stick with that. But t SM see, of course, has 581 00:33:25,320 --> 00:33:28,320 Speaker 1: spent all this money to upgrade. And then of course 582 00:33:28,360 --> 00:33:30,280 Speaker 1: another client might drop off and say, you know what, 583 00:33:30,400 --> 00:33:32,760 Speaker 1: it's getting a bit expensive. What's what we've got now 584 00:33:32,840 --> 00:33:35,480 Speaker 1: is good enough, We'll stick with that. And so I 585 00:33:35,520 --> 00:33:38,840 Speaker 1: think we could see t s MC being too good 586 00:33:39,080 --> 00:33:41,640 Speaker 1: and being a victim of their own success in that regard. 587 00:33:42,400 --> 00:33:45,440 Speaker 1: The other area where I think that they could really 588 00:33:45,640 --> 00:33:50,200 Speaker 1: kind of come down is is if there is a 589 00:33:50,320 --> 00:33:54,160 Speaker 1: slowdown in the tech industry. Is so vast that it 590 00:33:54,240 --> 00:33:57,520 Speaker 1: could really cut into their margins and their costs and 591 00:33:57,600 --> 00:34:00,920 Speaker 1: so forth, because even though they're the best in the world, 592 00:34:01,800 --> 00:34:04,480 Speaker 1: you know, people just don't have the money to buy things. Now. 593 00:34:04,480 --> 00:34:06,520 Speaker 1: Of course, we all thought this pandemic might be yet 594 00:34:06,520 --> 00:34:08,440 Speaker 1: a year ago when it was starting to unfold, we 595 00:34:08,480 --> 00:34:10,440 Speaker 1: thought that the tech industry was going to come to 596 00:34:10,880 --> 00:34:13,480 Speaker 1: its knees. It didn't pan out that way. And then 597 00:34:13,520 --> 00:34:16,520 Speaker 1: the other thing actually that I think t SMC has 598 00:34:16,560 --> 00:34:21,280 Speaker 1: to worry about is power, like energy electricity. The new 599 00:34:21,320 --> 00:34:27,560 Speaker 1: technologies are the new EUV machines guzzle a lot of energy. 600 00:34:27,680 --> 00:34:31,160 Speaker 1: In fact, and there's three science parks in Taiwan, the North, 601 00:34:31,200 --> 00:34:34,439 Speaker 1: Center and South. The South Science Park, which is where 602 00:34:34,480 --> 00:34:37,960 Speaker 1: the apple chips are made, actually had to rush in 603 00:34:38,000 --> 00:34:40,800 Speaker 1: the power company had to rush in new transformers to 604 00:34:40,920 --> 00:34:44,080 Speaker 1: feed enough electricity into the science park into t SMCS 605 00:34:44,120 --> 00:34:49,640 Speaker 1: factories because they are such power hugs, and Taiwan actually 606 00:34:49,640 --> 00:34:53,240 Speaker 1: has a power problem. Taiwan, for political reasons, is starting 607 00:34:53,239 --> 00:34:56,360 Speaker 1: to shut down its nuclear power stations. Nobody wants coal, 608 00:34:56,440 --> 00:34:59,759 Speaker 1: and the new neewable energy ideas that are coming on 609 00:35:00,120 --> 00:35:03,479 Speaker 1: not coming on quickly enough, and so Taiwan, in fact, 610 00:35:03,640 --> 00:35:07,160 Speaker 1: just just a week ago, UMC arrival face to power 611 00:35:07,160 --> 00:35:10,480 Speaker 1: outage and Taiwan's face power outages, and this is a 612 00:35:10,640 --> 00:35:14,280 Speaker 1: pending crisis. This is a looming crisis for the entirety 613 00:35:14,320 --> 00:35:17,080 Speaker 1: of Taiwan, and if they don't get their act together, 614 00:35:17,200 --> 00:35:21,479 Speaker 1: this could really hurt t SMC big time. So there 615 00:35:21,520 --> 00:35:24,560 Speaker 1: are a few land mines out there that t SMC 616 00:35:24,719 --> 00:35:26,680 Speaker 1: could step on, and not all of them will be 617 00:35:26,840 --> 00:35:29,160 Speaker 1: will be up to itself to avoid it, will need 618 00:35:29,200 --> 00:35:32,680 Speaker 1: help from the government. Another question that I have is, 619 00:35:32,719 --> 00:35:35,280 Speaker 1: so we've been talking about how like some of these 620 00:35:35,320 --> 00:35:39,759 Speaker 1: incumbent tech giants are. You know, they're going further up 621 00:35:40,239 --> 00:35:44,440 Speaker 1: the tech stack themselves, some of these huge cloud players 622 00:35:44,680 --> 00:35:48,640 Speaker 1: designing their own servers, designing their own silicon um. We 623 00:35:48,840 --> 00:35:52,160 Speaker 1: had an episode we talked about Apple. Could some of 624 00:35:52,200 --> 00:35:57,360 Speaker 1: these companies, these tech giants that have tremendous amounts of money, 625 00:35:57,360 --> 00:36:01,160 Speaker 1: ever have a competitive reason to get in to manufacturing 626 00:36:01,280 --> 00:36:06,080 Speaker 1: chips themselves. I don't think so. I don't see why 627 00:36:06,160 --> 00:36:09,200 Speaker 1: a Google or a Facebook, or a Microsoft or anyone 628 00:36:09,280 --> 00:36:12,200 Speaker 1: like that would want to get into the business. It 629 00:36:12,239 --> 00:36:16,560 Speaker 1: would take probably in the order of a willingness on 630 00:36:16,600 --> 00:36:18,920 Speaker 1: the part of a CFO or a CEO to burn 631 00:36:19,840 --> 00:36:23,080 Speaker 1: maybe twenty or thirty billion dollars, Like I'm not joking 632 00:36:23,120 --> 00:36:25,680 Speaker 1: because t SMC is going to be spending ten to 633 00:36:25,760 --> 00:36:28,399 Speaker 1: twenty next year. It's spent ten to twenty last year. 634 00:36:29,200 --> 00:36:31,600 Speaker 1: That's just on the equipment. And then you've got to 635 00:36:31,600 --> 00:36:36,320 Speaker 1: get the engineers. And sad to say, the best semi 636 00:36:36,360 --> 00:36:39,520 Speaker 1: conductor manufacturing engineers, not the design engineers. That the best 637 00:36:39,520 --> 00:36:43,279 Speaker 1: semi conductor manterfacturing engineers are really all in Asia now 638 00:36:43,640 --> 00:36:46,360 Speaker 1: distribute amongst the Koreans, the Japanese and the Taiwanese and 639 00:36:46,400 --> 00:36:49,680 Speaker 1: more and more in China. Now you do still have Intel, 640 00:36:49,760 --> 00:36:52,319 Speaker 1: and we shouldn't count our Intel now. Intel it's still 641 00:36:52,320 --> 00:36:55,319 Speaker 1: a very very solid company. They're definitely struggling right now, 642 00:36:55,360 --> 00:36:58,600 Speaker 1: but they have very very good engineers. So if you're 643 00:36:58,600 --> 00:37:00,719 Speaker 1: an American company thinking I want to get into the 644 00:37:00,800 --> 00:37:04,239 Speaker 1: chip game, I don't think any board of directors would 645 00:37:04,239 --> 00:37:05,839 Speaker 1: sign off on that, to be honest, I think you'd 646 00:37:05,880 --> 00:37:07,440 Speaker 1: be You'd be burning a lot of money for not 647 00:37:07,480 --> 00:37:09,120 Speaker 1: a lot of upside. Why do they need it? They've 648 00:37:09,160 --> 00:37:12,160 Speaker 1: got t SMC to do that for them. So here's 649 00:37:12,160 --> 00:37:14,799 Speaker 1: something that I'm still struggling with a little bit and 650 00:37:14,840 --> 00:37:18,319 Speaker 1: we've talked about on other episodes. Is this idea of 651 00:37:18,800 --> 00:37:23,960 Speaker 1: learning by doing the link between semiconductor design and semi 652 00:37:24,040 --> 00:37:29,600 Speaker 1: conductor manufacturing. Some of these red hot chip companies like 653 00:37:29,760 --> 00:37:33,799 Speaker 1: say a Nvidia, do they are there ever going to 654 00:37:33,840 --> 00:37:36,439 Speaker 1: be risks to their business model that they aren't more 655 00:37:36,680 --> 00:37:41,040 Speaker 1: intimately involved with the manufacturing aspect of the job and 656 00:37:41,080 --> 00:37:44,600 Speaker 1: are sort of strictly design or is that? Is that sustainable? 657 00:37:45,680 --> 00:37:47,439 Speaker 1: And now I think that is a risk. I don't 658 00:37:47,440 --> 00:37:50,399 Speaker 1: think it's a risk for the near term. I think 659 00:37:50,400 --> 00:37:53,160 Speaker 1: it's a risk for maybe ten or twenty years out. 660 00:37:53,400 --> 00:37:56,120 Speaker 1: I mean, quantum computing is kind of around the corner. 661 00:37:56,160 --> 00:37:58,439 Speaker 1: I don't think we're going to see quantum computers next year, 662 00:37:58,520 --> 00:38:01,160 Speaker 1: but you know, that's the thinking is in in the 663 00:38:01,239 --> 00:38:05,279 Speaker 1: next generation, So say twenty years, quantum computers will be 664 00:38:05,280 --> 00:38:09,320 Speaker 1: out there, and and quantum computing is just it turns 665 00:38:09,400 --> 00:38:12,560 Speaker 1: on its head the semi conductor industry because instead of 666 00:38:12,560 --> 00:38:15,799 Speaker 1: having ones and zeros, you've got cubits and and some 667 00:38:15,880 --> 00:38:20,919 Speaker 1: kind of mid state between off and on that is, 668 00:38:21,719 --> 00:38:24,719 Speaker 1: you know, different to the base of how semiconductors work now, 669 00:38:24,719 --> 00:38:27,120 Speaker 1: which is very binary. Is literally a transistor is on 670 00:38:27,239 --> 00:38:29,200 Speaker 1: or a transistor is off, and you write all of 671 00:38:29,239 --> 00:38:35,200 Speaker 1: your code and you design semiconductor circuits around this basic 672 00:38:35,320 --> 00:38:37,840 Speaker 1: logic gates this idea of on and off and quantum 673 00:38:37,880 --> 00:38:42,040 Speaker 1: computing is very very different, and so if that comes through, 674 00:38:42,800 --> 00:38:45,799 Speaker 1: it may be in the best interests of someone like 675 00:38:45,840 --> 00:38:47,960 Speaker 1: an video to say, you know what, we want to 676 00:38:48,000 --> 00:38:50,920 Speaker 1: dabble in making it ourselves because we we think we 677 00:38:51,000 --> 00:38:53,880 Speaker 1: know the best way to do it. And maybe someone 678 00:38:53,920 --> 00:38:55,640 Speaker 1: like in the video would say, you know, we're going 679 00:38:55,680 --> 00:38:57,879 Speaker 1: to start a little test fab just you know, here 680 00:38:57,880 --> 00:39:00,560 Speaker 1: in Silicon Valley and and play around with them, see 681 00:39:00,600 --> 00:39:03,319 Speaker 1: how we do, and maybe they'll discover the works well 682 00:39:03,400 --> 00:39:05,000 Speaker 1: for them, and they get a lot of orders and 683 00:39:05,000 --> 00:39:07,359 Speaker 1: they decide that this is something they can scale up. 684 00:39:07,920 --> 00:39:10,880 Speaker 1: But they would have to be designing and kind of 685 00:39:10,920 --> 00:39:13,920 Speaker 1: discovering the technology themselves to do so. I don't think 686 00:39:13,920 --> 00:39:18,279 Speaker 1: you can set up shop tomorrow and get that technology 687 00:39:18,320 --> 00:39:21,880 Speaker 1: off the shelf. So much of what any semiconductor maker 688 00:39:22,000 --> 00:39:27,000 Speaker 1: is doing now is very very proprietary, to the point 689 00:39:27,080 --> 00:39:30,399 Speaker 1: where nobody else could copy it, even if they were 690 00:39:30,400 --> 00:39:33,760 Speaker 1: to take over the factory, like if if if for example, 691 00:39:33,840 --> 00:39:36,799 Speaker 1: China invaded and got access to t SMC factories and 692 00:39:36,800 --> 00:39:39,640 Speaker 1: t SMC engineers disappeared. I don't think they would know 693 00:39:39,680 --> 00:39:42,239 Speaker 1: what to do with it. So you'd have to have 694 00:39:42,360 --> 00:39:46,680 Speaker 1: that fundamental knowledge and really be building that knowledge from 695 00:39:46,680 --> 00:39:50,120 Speaker 1: the ground up. And I think there's a possibility that's 696 00:39:50,160 --> 00:39:54,279 Speaker 1: someone like an Nvidio, maybe a m D or maybe 697 00:39:54,320 --> 00:39:56,520 Speaker 1: a chip company we haven't come across yet decides that 698 00:39:56,600 --> 00:40:00,400 Speaker 1: actually we want to have a closer relationship with man factoring. 699 00:40:00,440 --> 00:40:02,480 Speaker 1: And in fact, I think the company that is very 700 00:40:02,600 --> 00:40:06,600 Speaker 1: very closely connected to manufacturing is probably Apple. As Apple 701 00:40:06,640 --> 00:40:09,919 Speaker 1: goes down the Apples silicon path, I think we'll see 702 00:40:10,440 --> 00:40:13,160 Speaker 1: them get a bit more involved in the manufacturing process. 703 00:40:14,360 --> 00:40:16,719 Speaker 1: What scope is there for government policy to play a 704 00:40:16,840 --> 00:40:20,440 Speaker 1: role when it comes to encouraging domestic industries that are 705 00:40:20,480 --> 00:40:24,880 Speaker 1: able to compete with t SMC. So, for instance, in China, 706 00:40:25,000 --> 00:40:28,000 Speaker 1: we know that the government there has already announced its 707 00:40:28,040 --> 00:40:33,120 Speaker 1: intentions to build up its own chip manufacturing capacity. We've 708 00:40:33,160 --> 00:40:36,399 Speaker 1: talked a lot in this podcast about the idea that 709 00:40:36,440 --> 00:40:38,960 Speaker 1: you know, more than half of the world semiconductors are 710 00:40:38,960 --> 00:40:44,160 Speaker 1: coming from this one company that's on a rather vulnerable island, 711 00:40:44,239 --> 00:40:48,800 Speaker 1: caught between two very very big geopolitical forces. If countries 712 00:40:49,000 --> 00:40:54,560 Speaker 1: recognize chips as strategically important. Will they be trying to 713 00:40:54,880 --> 00:40:58,560 Speaker 1: encourage their own domestic production and how successful do you 714 00:40:58,600 --> 00:41:02,120 Speaker 1: think they might be in that? Yeah, I think that's 715 00:41:02,239 --> 00:41:07,440 Speaker 1: very very true. I think both both Beijing and Washington 716 00:41:07,520 --> 00:41:11,600 Speaker 1: of recognized semi conductors as being of national security and 717 00:41:11,680 --> 00:41:15,800 Speaker 1: geopolitical interest to them. China is recognized that for twenty years. 718 00:41:15,840 --> 00:41:17,920 Speaker 1: But you know what, they're still to be blunt with you, 719 00:41:18,040 --> 00:41:20,640 Speaker 1: they're not really doing that well. The people say our 720 00:41:20,719 --> 00:41:22,960 Speaker 1: China is catching up. China's catching up, and that's true, 721 00:41:23,000 --> 00:41:25,400 Speaker 1: but it's not like t SMC and everyone else is 722 00:41:25,400 --> 00:41:28,480 Speaker 1: standing still. They're still moving forward and leaps and bounds 723 00:41:28,480 --> 00:41:34,040 Speaker 1: and and Smick Semi Collacter Matufacturing International, which is kind 724 00:41:34,040 --> 00:41:36,600 Speaker 1: of I guess the most famous of the flag bearers 725 00:41:36,600 --> 00:41:40,120 Speaker 1: in China is really still struggling to get close to 726 00:41:40,239 --> 00:41:43,320 Speaker 1: company like t SMC, and they really can't replace t 727 00:41:43,520 --> 00:41:46,880 Speaker 1: SMC for example, if in Huawe needs to outsource to 728 00:41:46,920 --> 00:41:49,560 Speaker 1: somebody else. So I think China is a really good 729 00:41:49,600 --> 00:41:52,239 Speaker 1: example of how difficult it is that with even with 730 00:41:52,800 --> 00:41:57,080 Speaker 1: endless amounts of money and very very smart engine You know, 731 00:41:57,160 --> 00:41:59,840 Speaker 1: China has really good engineers, right, But they're just not 732 00:42:00,040 --> 00:42:02,160 Speaker 1: as good as the Taiwanese or even the Americans or 733 00:42:02,160 --> 00:42:05,000 Speaker 1: the Japanese or Koreans. Right, and then the other side, 734 00:42:05,040 --> 00:42:07,200 Speaker 1: of course, the United States, they would like to get 735 00:42:07,360 --> 00:42:10,279 Speaker 1: their chip glory back. But you know, one of the 736 00:42:10,280 --> 00:42:13,160 Speaker 1: big problems is if you look at Silicon Valley, like 737 00:42:13,239 --> 00:42:17,000 Speaker 1: literally that stretch of land we call Silicon Valley, almost 738 00:42:17,000 --> 00:42:20,080 Speaker 1: no silicon is made there anymore. Right. In fact, there's 739 00:42:20,080 --> 00:42:23,640 Speaker 1: more silicon made in Arizona, where Intel has a factory 740 00:42:23,680 --> 00:42:27,759 Speaker 1: that is made in California. That's that's data I checked out. Yeah, 741 00:42:27,880 --> 00:42:30,879 Speaker 1: it's it's it's amazing. Arizona is the new Silicon Valley. 742 00:42:31,239 --> 00:42:35,360 Speaker 1: So that's part of the problem that any government will face. Obviously, 743 00:42:35,680 --> 00:42:39,920 Speaker 1: Donald Trump wanted to get things going, using State and 744 00:42:39,960 --> 00:42:43,160 Speaker 1: Commerce and various other departments to push that forward. It 745 00:42:43,200 --> 00:42:45,759 Speaker 1: remains to be seen where the Biden will will take 746 00:42:45,800 --> 00:42:49,720 Speaker 1: that forward. But America does have a chance to bring 747 00:42:49,760 --> 00:42:52,120 Speaker 1: back and revitalize its chip industry, and I think the 748 00:42:52,160 --> 00:42:55,400 Speaker 1: way they will do it will be semi national interest 749 00:42:55,480 --> 00:43:00,080 Speaker 1: protectionist policies such as you know, any new equipment it 750 00:43:00,160 --> 00:43:04,480 Speaker 1: goes into Department of Defense server farm must have chips 751 00:43:04,480 --> 00:43:06,200 Speaker 1: made in America that kind of thing. You know, you 752 00:43:06,280 --> 00:43:11,080 Speaker 1: put up national security policies and that basically means it all, 753 00:43:11,120 --> 00:43:12,719 Speaker 1: if you want to sell to these companies, you have 754 00:43:12,840 --> 00:43:15,120 Speaker 1: to make it on our terms, which probably means on 755 00:43:15,160 --> 00:43:18,240 Speaker 1: American soil. So that's what they're doing with t SMC. 756 00:43:18,400 --> 00:43:20,160 Speaker 1: They're trying to get t SMC to set up and 757 00:43:20,200 --> 00:43:23,319 Speaker 1: tier SMC will set up in Arizona. And I think 758 00:43:23,360 --> 00:43:26,520 Speaker 1: if you were to take that to its logical conclusion, 759 00:43:27,360 --> 00:43:29,759 Speaker 1: I can imagine where there would be some kind of 760 00:43:30,120 --> 00:43:35,719 Speaker 1: technology transfer where companies outside of the United States, Taiwanese 761 00:43:35,800 --> 00:43:40,000 Speaker 1: or Japanese or Korean companies will be incentivized and and 762 00:43:40,280 --> 00:43:43,799 Speaker 1: pushed to transfer some of the technology or do joint 763 00:43:43,880 --> 00:43:47,120 Speaker 1: ventures with American companies to help transfer that technology back, 764 00:43:47,400 --> 00:43:51,120 Speaker 1: which would be a beautiful irony because the Asian countries 765 00:43:51,280 --> 00:43:54,600 Speaker 1: got their technology from America fifty years ago, so it 766 00:43:54,600 --> 00:43:57,279 Speaker 1: would really be a nice kind of closing of the 767 00:43:57,320 --> 00:44:00,200 Speaker 1: loop if America then started to have the technolo lets 768 00:44:00,200 --> 00:44:04,000 Speaker 1: you trying to come back the other way, tim that 769 00:44:04,120 --> 00:44:07,200 Speaker 1: was that was fantastic. It's it's it answered a bunch 770 00:44:07,200 --> 00:44:09,120 Speaker 1: of questions that we have. Is sort of the perfect 771 00:44:09,560 --> 00:44:11,279 Speaker 1: way to move the story forward for us. So I 772 00:44:11,320 --> 00:44:14,239 Speaker 1: really appreciate you join us, no problem. It's fun to 773 00:44:14,239 --> 00:44:19,000 Speaker 1: touch you guys. Thanks to Tim Crash course in TSMC. Yeah, 774 00:44:19,000 --> 00:44:40,040 Speaker 1: that was fantastic. I really I really enjoyed that, you know. 775 00:44:40,080 --> 00:44:43,560 Speaker 1: I think that last question that you asked and his 776 00:44:43,680 --> 00:44:48,120 Speaker 1: answer really helped put some pieces together for me about say, 777 00:44:48,200 --> 00:44:52,920 Speaker 1: like the challenge that China has been sort of replicating 778 00:44:52,960 --> 00:44:55,600 Speaker 1: a world class chip system, and it really sort of 779 00:44:55,920 --> 00:44:58,040 Speaker 1: the whole story of t SMC and really all of 780 00:44:58,040 --> 00:45:00,920 Speaker 1: this and how much of this knowledge is just embedded 781 00:45:00,960 --> 00:45:03,480 Speaker 1: in the engineers and the processes that have been going 782 00:45:03,480 --> 00:45:07,120 Speaker 1: on forever. This idea, like you cannot replicate it just 783 00:45:07,160 --> 00:45:10,040 Speaker 1: with money. You could throw an infinite literally infinite amount 784 00:45:10,040 --> 00:45:12,960 Speaker 1: of money at something or virtually infinite, but there's no 785 00:45:13,000 --> 00:45:16,640 Speaker 1: way to sort of recreate that, know how overnight. I 786 00:45:16,680 --> 00:45:21,000 Speaker 1: think it's just a really really important thing. Yeah. I 787 00:45:21,040 --> 00:45:23,719 Speaker 1: think one of the things that's standing out in all 788 00:45:23,760 --> 00:45:27,120 Speaker 1: of these episodes is just how deep the moat is 789 00:45:27,360 --> 00:45:32,120 Speaker 1: around semi conductor expertise. The more I learned about it, 790 00:45:32,320 --> 00:45:35,000 Speaker 1: the more difficult it seems to me to you know, 791 00:45:35,120 --> 00:45:38,880 Speaker 1: catch up to these players, and in particular t SMC, 792 00:45:39,120 --> 00:45:42,960 Speaker 1: which as we just discussed, has this big leading position 793 00:45:43,239 --> 00:45:45,839 Speaker 1: in the industry. The other thing I loved was that 794 00:45:46,000 --> 00:45:49,960 Speaker 1: last bit of irony from Tim, this idea that the 795 00:45:50,040 --> 00:45:55,120 Speaker 1: US might end up doing tech transfers to import semiconductor 796 00:45:55,440 --> 00:46:00,240 Speaker 1: expertise after having exported it many many years ago. I 797 00:46:00,239 --> 00:46:02,480 Speaker 1: I love how we come full circle in that way. 798 00:46:02,560 --> 00:46:04,279 Speaker 1: But I guess, just going back to the other point 799 00:46:04,360 --> 00:46:08,919 Speaker 1: about the idea that there's this deep moat around foundries 800 00:46:09,000 --> 00:46:13,719 Speaker 1: and semiconductor manufacturing. I mean twenty or thirty years ago, 801 00:46:14,400 --> 00:46:18,400 Speaker 1: a lot of people wouldn't have expected Taiwan to have 802 00:46:18,600 --> 00:46:21,680 Speaker 1: built up an industry like this, So I guess things 803 00:46:21,760 --> 00:46:26,960 Speaker 1: can change over the long term. Yeah, things can change 804 00:46:26,960 --> 00:46:29,600 Speaker 1: over the over the long term, and Tim sort of 805 00:46:29,680 --> 00:46:34,480 Speaker 1: identified some things that over time could perhaps erode t 806 00:46:34,680 --> 00:46:39,759 Speaker 1: S MCS, MOTE or maybe some different models, whereby, as 807 00:46:39,760 --> 00:46:42,160 Speaker 1: you mentioned, maybe companies don't feel they need to pay 808 00:46:42,239 --> 00:46:45,880 Speaker 1: for always the state of the art as a tech changes. 809 00:46:46,360 --> 00:46:49,000 Speaker 1: But I really do think that drove home this idea. 810 00:46:49,080 --> 00:46:51,560 Speaker 1: It's like there's sort of the whole thing, just the 811 00:46:51,600 --> 00:46:55,279 Speaker 1: compounding advantage that t SMC is building up. Now you 812 00:46:55,280 --> 00:46:57,080 Speaker 1: have to do it. Look at the company's stock price 813 00:46:57,800 --> 00:47:00,239 Speaker 1: and I think it's now six or seven of your 814 00:47:00,280 --> 00:47:03,480 Speaker 1: billion dollar company almost which puts in lead with some 815 00:47:03,600 --> 00:47:06,719 Speaker 1: of the real, you know, huge tech winners. You can 816 00:47:06,800 --> 00:47:11,800 Speaker 1: see this is increasingly this mode, this advantage is increasingly 817 00:47:11,840 --> 00:47:17,719 Speaker 1: reflected in the investor perceptions of the company. Absolutely all right, 818 00:47:17,760 --> 00:47:21,400 Speaker 1: So now my question is how many more semiconductor episodes 819 00:47:21,480 --> 00:47:23,200 Speaker 1: are we going to be doing. I think we have 820 00:47:23,239 --> 00:47:26,520 Speaker 1: one more. I think we maybe whether that we could 821 00:47:26,560 --> 00:47:28,319 Speaker 1: maybe pause. I think we're going to be talking to 822 00:47:28,360 --> 00:47:31,680 Speaker 1: one of our old favorites, Dan Waging, pretty soon to 823 00:47:31,800 --> 00:47:35,600 Speaker 1: really sort of figure out, Okay, basically a year post COVID, 824 00:47:35,840 --> 00:47:39,160 Speaker 1: how is the state of China's tech industry and where 825 00:47:39,160 --> 00:47:41,719 Speaker 1: they really stand with all of this. But I think 826 00:47:41,719 --> 00:47:44,279 Speaker 1: that will be the last one for now. For now 827 00:47:44,480 --> 00:47:47,839 Speaker 1: is the key here, because I mean, I'm joking a 828 00:47:47,840 --> 00:47:50,440 Speaker 1: little bit because you wanted to turn all thoughts into 829 00:47:50,520 --> 00:47:54,040 Speaker 1: a semiconductor podcast. But I do think another one of 830 00:47:54,080 --> 00:47:57,880 Speaker 1: the standouts from all these conversations is just how important 831 00:47:57,880 --> 00:48:01,200 Speaker 1: this technology is going to be in the future and 832 00:48:01,239 --> 00:48:04,560 Speaker 1: this idea that it really is becoming a strategic asset. 833 00:48:04,640 --> 00:48:08,040 Speaker 1: I've seen some people compare it to oil in the 834 00:48:08,120 --> 00:48:12,360 Speaker 1: sense that it underpins all these different things in the 835 00:48:12,400 --> 00:48:15,279 Speaker 1: global economy. So I do think this is going to 836 00:48:15,719 --> 00:48:19,120 Speaker 1: come up again and again and again. Yeah, no, totally, 837 00:48:19,160 --> 00:48:21,640 Speaker 1: Like and I hadn't really even thought about, you know, 838 00:48:21,719 --> 00:48:24,640 Speaker 1: that point about the US putting pressure on t SMC 839 00:48:25,440 --> 00:48:28,719 Speaker 1: not to sell chips to Lahwei. That's super interesting in 840 00:48:28,760 --> 00:48:31,359 Speaker 1: its own right, Like, here is this company that's in 841 00:48:32,040 --> 00:48:38,080 Speaker 1: Taiwan that China itself, would you know, Taiwan from China's 842 00:48:38,120 --> 00:48:40,440 Speaker 1: view as part of China, and yet still from the 843 00:48:40,480 --> 00:48:42,560 Speaker 1: business perspective and sort of the way things are that 844 00:48:42,600 --> 00:48:46,759 Speaker 1: it's the US that can pressure t SMC not to 845 00:48:46,800 --> 00:48:51,799 Speaker 1: build for Huawei, then the other way around. Right, So 846 00:48:51,960 --> 00:48:54,759 Speaker 1: we've spent you know, all this time talking about resiliency 847 00:48:54,880 --> 00:48:58,560 Speaker 1: of supply chains and worrying about what happens if you know, 848 00:48:58,640 --> 00:49:02,880 Speaker 1: there's a pandemic in the country where you manufacture something 849 00:49:02,920 --> 00:49:07,480 Speaker 1: that's vital for your final product, or if sanctions come 850 00:49:07,480 --> 00:49:10,719 Speaker 1: in that suddenly cut off supplier or something like that. 851 00:49:11,480 --> 00:49:16,160 Speaker 1: And here we have an area where the impact could 852 00:49:16,200 --> 00:49:19,279 Speaker 1: be really, really strong. But also it seems likely that 853 00:49:19,320 --> 00:49:23,600 Speaker 1: because of the strategic importance of semiconductors, they will actually 854 00:49:23,680 --> 00:49:27,839 Speaker 1: be targeted by countries that are trying to gain a 855 00:49:27,920 --> 00:49:32,560 Speaker 1: tech advantage. Absolutely, it's super, super interesting dynamics that I 856 00:49:32,560 --> 00:49:35,880 Speaker 1: hadn't really thought at all before prior to this conversation. 857 00:49:37,160 --> 00:49:40,839 Speaker 1: All right, so we're all a semiconductor enthusiasts and nerds. Now, 858 00:49:41,160 --> 00:49:43,920 Speaker 1: shall we leave it there? Yes? Leave it there? All right? 859 00:49:44,160 --> 00:49:47,120 Speaker 1: This has been another episode of the All Thoughts Podcast. 860 00:49:47,200 --> 00:49:49,960 Speaker 1: I'm Tracy Alloway. You can follow me on Twitter at 861 00:49:49,960 --> 00:49:52,680 Speaker 1: Tracy Alloway and I'm Joe Why Isn't Thought? You can 862 00:49:52,719 --> 00:49:56,000 Speaker 1: follow me on Twitter at The Stalwart. Follow our guest 863 00:49:56,080 --> 00:49:59,399 Speaker 1: Tim Colpin. He's on Twitter at t Colpen and check 864 00:49:59,400 --> 00:50:02,520 Speaker 1: out all of his columns at Bloomberg Opinion. Follow our 865 00:50:02,560 --> 00:50:06,400 Speaker 1: producer Laura Carlson. She's at Laura M. Carlson. Followed the 866 00:50:06,400 --> 00:50:10,200 Speaker 1: Bloomberg head of podcast, Francesco Levi at Francesca Today, and 867 00:50:10,360 --> 00:50:13,879 Speaker 1: check out all of our podcasts under the handle at podcasts. 868 00:50:14,000 --> 00:50:14,760 Speaker 1: Thanks for listening.