WEBVTT - The Massive Economic Impact If China Invades Taiwan 0:00:10.200 --> 0:00:14.160 Hello, and welcome to another episode of the Odd Lots Podcast. 0:00:14.320 --> 0:00:16.760 I'm Joe Wisenthal and I'm Tracy Alloway. 0:00:17.079 --> 0:00:21.200 Tracy, you know, for years, I've always thought that if 0:00:21.200 --> 0:00:23.680 you want to be like a smart sounding pundit at 0:00:23.680 --> 0:00:25.959 the beginning of the year and someone ask you, like, 0:00:26.200 --> 0:00:28.960 what risks are you thinking about right now, you like 0:00:29.040 --> 0:00:31.160 stroke your chin and look off and you say, I'm 0:00:31.160 --> 0:00:34.280 really worried about like geopolitical risk. And that's always like 0:00:34.320 --> 0:00:37.320 a safe answer anytime any year. Like that always makes 0:00:37.320 --> 0:00:38.320 you sound sophisticated. 0:00:38.479 --> 0:00:43.800 Yes, unexpected geopolitical risks are an evergreen category of punditry. 0:00:43.840 --> 0:00:44.400 This is true. 0:00:44.680 --> 0:00:48.559 There is one specific risk within that category that I 0:00:48.720 --> 0:00:51.879 feel like has come up every year for almost as 0:00:51.960 --> 0:00:54.760 long as I can remember, and that is the risk 0:00:55.040 --> 0:00:58.400 of something happening between China and Taiwan. And I have 0:00:58.440 --> 0:01:01.880 a really embarrassing confession to make, Okay. I wrote my 0:01:02.040 --> 0:01:05.840 dissertation on I wrote it in Beijing, and it must 0:01:05.880 --> 0:01:07.360 have been in like two thousand and five or two 0:01:07.400 --> 0:01:11.959 thousand and six, predicting that China would invade Taiwan before 0:01:12.040 --> 0:01:15.119 the Beijing Olympics. I think that was like two thousand 0:01:15.160 --> 0:01:16.040 and eight or something. 0:01:16.120 --> 0:01:19.520 So when people talk about the risk of China invading 0:01:19.640 --> 0:01:24.960 Taiwan or attempting to formally reunify, this is like twenty 0:01:25.000 --> 0:01:26.759 years people have been talking about this and it keeps not. 0:01:26.680 --> 0:01:29.399 Happening, Yes, and getting it wrong consistently for twenty years, 0:01:29.440 --> 0:01:30.080 or at least I have. 0:01:30.600 --> 0:01:33.680 But it does seem right like for whatever reason. And 0:01:33.800 --> 0:01:36.480 we'll get into it that we are once again in 0:01:36.520 --> 0:01:40.720 a wave of heightened concern. And you know, there are 0:01:40.720 --> 0:01:45.440 two very active wars happening right now currently between Russia 0:01:45.480 --> 0:01:48.680 and Ukraine, the Israel and Hamas war, and so the 0:01:48.760 --> 0:01:53.000 idea of war, unexpected war is on people's minds. And 0:01:53.000 --> 0:01:56.680 then we know that tension has been building in various 0:01:56.720 --> 0:02:00.320 ways Taiwan. There was all of the angst over the 0:02:00.600 --> 0:02:03.560 Nancy Pelosi trip a couple of years ago, for example, 0:02:03.720 --> 0:02:07.440 and the heightened awareness that the world has on semiconductors 0:02:07.440 --> 0:02:10.919 that are manufactured in Taiwan, and sort of a more 0:02:10.960 --> 0:02:15.960 aggressive posture I believe from Xi Jinping and the military 0:02:16.080 --> 0:02:20.480 of more aggressive war games and exercises. So many people 0:02:21.000 --> 0:02:24.600 are concerned about some prospect of a hot conflict in 0:02:24.600 --> 0:02:25.440 the next few years. 0:02:25.560 --> 0:02:28.320 Yeah, you mentioned semiconductors and I feel like this is 0:02:28.360 --> 0:02:31.119 really the key thing, which is in the past two 0:02:31.400 --> 0:02:34.440 or well, I guess four years now, since the outbreak 0:02:34.440 --> 0:02:36.840 of the global pandemic, we have had a sort of 0:02:37.000 --> 0:02:40.920 crash course in the importance of the Taiwanese economy to 0:02:41.400 --> 0:02:44.320 the broader world. And I feel like that's sort of 0:02:44.560 --> 0:02:46.760 done a couple things. So one, it's made everyone a 0:02:46.760 --> 0:02:49.560 lot more nervous about what could happen if that supply 0:02:49.639 --> 0:02:53.400 chain gets disrupted again, although in a very different way. 0:02:53.600 --> 0:02:56.840 And then secondly, maybe it's given a little bit of 0:02:56.960 --> 0:03:01.839 ammunition to China to saber rattle on this issue even more. Right, 0:03:01.919 --> 0:03:05.440 if it threatens something against Taiwan, the whole world now 0:03:05.520 --> 0:03:06.720 knows what's at stake. 0:03:07.240 --> 0:03:09.920 Yeah, that's absolutely right. So this is one of these 0:03:10.000 --> 0:03:12.840 things that is you could feel it building more and 0:03:12.840 --> 0:03:14.960 more people are talking about it. My understanding and what 0:03:15.000 --> 0:03:17.160 I've seen is people say, oh, in DC, everyone is 0:03:17.200 --> 0:03:19.560 like obsessed with this question. I don't feel like that's 0:03:19.600 --> 0:03:21.720 as much the case in New York, but I do 0:03:21.760 --> 0:03:25.040 get the impression that in DC talk of a war 0:03:25.440 --> 0:03:27.280 is very high. And so I think, like, you know, 0:03:27.400 --> 0:03:29.320 we have to do more on this, And I think 0:03:29.560 --> 0:03:32.600 geopolitics in general, thinking back to some of our conversations 0:03:32.600 --> 0:03:35.200 with Zultan et cetera about you know, the sort of 0:03:35.240 --> 0:03:37.840 rearranging world, like, we need to dive into these questions 0:03:37.880 --> 0:03:38.440 a little bit more. 0:03:38.600 --> 0:03:38.800 Yeah. 0:03:38.840 --> 0:03:41.520 Absolutely, And the other reason we need to do it 0:03:41.560 --> 0:03:45.280 is Taiwan is holding an election. In fact, we are 0:03:45.320 --> 0:03:49.000 recording this on January ninth. By the time this episode 0:03:49.040 --> 0:03:51.440 comes out, the election will have been held and so 0:03:51.520 --> 0:03:55.440 we'll know who the winning candidate is. But in the meantime, like, 0:03:55.560 --> 0:03:58.840 there is clearly a big reason to discuss this. 0:03:59.120 --> 0:03:59.320 Yeah. 0:03:59.360 --> 0:04:01.960 Absolutely, Well, I'm very excited. We have two of our 0:04:02.040 --> 0:04:05.040 colleagues here at Bloomberg that we've never had on the 0:04:05.040 --> 0:04:07.840 podcast before, but they are the perfect guests to discuss 0:04:07.880 --> 0:04:10.760 this because they've been doing a lot of work in 0:04:10.800 --> 0:04:13.600 this area. We're going to be speaking with Jennifer Welch, 0:04:13.880 --> 0:04:18.160 chief geoeconomics analyst at Bloomberg Economics, and Gerard De Pippo, 0:04:18.400 --> 0:04:22.880 Senior geoeconomics analyst at Bloomberg Economics. They've both published a 0:04:22.920 --> 0:04:25.960 lot of work on some of these questions related questions 0:04:26.000 --> 0:04:29.919 wargaming the risk the economic stakes involved here, and so 0:04:29.960 --> 0:04:32.240 we're going to start to go down our path of 0:04:32.360 --> 0:04:34.840 understanding this tension better. So, Jennifer and Gerard, thank you 0:04:34.960 --> 0:04:36.800 so much for coming on odd lots. 0:04:36.839 --> 0:04:38.240 Thank you so much for having us. 0:04:38.360 --> 0:04:39.039 Glad to be here. 0:04:39.560 --> 0:04:42.680 I have a sort of very basic, I don't know, 0:04:42.760 --> 0:04:44.599 maybe it's not basic zoom od question. You know, we 0:04:44.640 --> 0:04:47.719 talked to like normal talk to economists who you know, 0:04:47.760 --> 0:04:51.320 they have their recession forecast or forecast of soft landing, whatever, 0:04:51.480 --> 0:04:54.640 and I sort of have some idea of how they 0:04:54.680 --> 0:04:58.880 go about, you know, modeling different scenarios when it comes 0:04:58.920 --> 0:05:03.839 to wargam and modeling the odds of a war or 0:05:03.920 --> 0:05:07.080 modeling how a war might go if one were pursued. 0:05:07.680 --> 0:05:11.839 What do you economics analysts, like, what is the toolkit 0:05:12.400 --> 0:05:16.480 that you have to actually go about modeling the risks 0:05:16.520 --> 0:05:17.640 and the outcomes of a war. 0:05:18.080 --> 0:05:20.800 That's a great question, and let me dry it. I'll 0:05:21.000 --> 0:05:22.440 start off, and then if you want to jump in 0:05:22.520 --> 0:05:24.760 as well, so on the first part of your question 0:05:24.880 --> 0:05:28.320 of how do we estimate the probability of war? To 0:05:28.400 --> 0:05:31.920 be perfectly candid, this is definitely more art than science. 0:05:32.279 --> 0:05:35.159 What we did in the case of our recent project 0:05:35.279 --> 0:05:37.719 on Taiwan and the trajectory we might see in the 0:05:37.760 --> 0:05:41.880 Taiwan Strait was take a look at sort of baseline conditions, 0:05:42.000 --> 0:05:45.960 what are the sort of structures in place there that 0:05:46.080 --> 0:05:50.920 lean towards stability or that lean towards increasing tensions and 0:05:51.279 --> 0:05:54.440 even to the extreme of a crisis or conflict, And 0:05:54.480 --> 0:05:58.360 we examined all those key variables and their trajectory and 0:05:58.400 --> 0:06:01.600 what weight they might have to be candid. This is 0:06:01.760 --> 0:06:03.720 an area of a lot of debate among people who 0:06:03.800 --> 0:06:06.359 focus on the Taiwan Street. You know, to what extent 0:06:06.680 --> 0:06:10.520 does Beijing wield the most agency over it versus Taiwan, 0:06:10.720 --> 0:06:14.080 versus the United States, which has historically been a key 0:06:14.160 --> 0:06:17.599 balancing power in the street. And then from there we 0:06:17.760 --> 0:06:21.680 kind of developed our sense of probabilities. But again, this 0:06:21.760 --> 0:06:27.200 is a really complex and dynamic geopolitical environment. The election 0:06:27.520 --> 0:06:30.799 that Taiwan will be having in advance of this recording 0:06:31.279 --> 0:06:34.080 is an example of how things can really shift at 0:06:34.160 --> 0:06:37.159 quick notice depending on how events like that turn out. 0:06:37.880 --> 0:06:41.599 But overall, we think our sense of what the risks 0:06:41.640 --> 0:06:44.159 of crisis or conflict in the Strait are are meant 0:06:44.160 --> 0:06:46.640 to be, you know, standing the test of time and 0:06:46.680 --> 0:06:49.200 looking out over the next five years. And then in 0:06:49.279 --> 0:06:52.359 terms of how we study what the outcome of a 0:06:52.480 --> 0:06:55.680 war would be were one to happen. This is a 0:06:55.760 --> 0:06:58.760 little bit more of a rigorous process, and there are 0:06:58.760 --> 0:07:02.560 a lot of think tanks, a lot of internal government entities, 0:07:02.600 --> 0:07:06.080 not just in the United States, but Japan, Taiwan, all 0:07:06.120 --> 0:07:08.680 around the world that are increasingly looking at how a 0:07:08.800 --> 0:07:12.080 crosstright conflict could play out, and it often is very 0:07:12.160 --> 0:07:15.920 much focused on the military dynamics. That's historically the origin 0:07:15.960 --> 0:07:19.840 of these war games is testing military tactics and battle 0:07:19.880 --> 0:07:24.720 plans and playing out how different scenarios can produce different 0:07:24.760 --> 0:07:28.120 outcomes depending on sort of what you go into the 0:07:28.160 --> 0:07:33.120 battle with and how different conditions can change those outcomes. 0:07:33.560 --> 0:07:37.200 So for that particular exercise, there's a lot of different 0:07:37.320 --> 0:07:40.640 variations of war games related to Taiwan out there, and 0:07:40.840 --> 0:07:43.360 a lot of them produce a variety of results. It 0:07:43.400 --> 0:07:46.400 really depends on kind of the assumptions going into it on. 0:07:46.960 --> 0:07:49.280 You know, for example, how does the worst start, is 0:07:49.320 --> 0:07:52.200 there any warning of it, how prepared is Taiwan to 0:07:52.240 --> 0:07:56.040 defend itself? Does the US get involved? But what all 0:07:56.080 --> 0:07:59.640 of them point to is a war would be incredibly 0:07:59.680 --> 0:08:02.320 costly at all sides. Most of the war games focus 0:08:02.400 --> 0:08:05.760 on sort of the military and the humanitarian impact of that. 0:08:06.320 --> 0:08:09.200 But we looked at also the economic impact, and what 0:08:09.240 --> 0:08:14.520 we found is especially for China, Taiwan, and the United States, 0:08:14.560 --> 0:08:17.559 but also the global economy, the toll would be quite 0:08:17.560 --> 0:08:21.080 immense were the cross street tensions to erupt into a 0:08:21.120 --> 0:08:22.320 broader conflict. 0:08:22.600 --> 0:08:24.800 Maybe this is obvious, but I think it's worth pointing 0:08:24.800 --> 0:08:27.440 out that we're dealing with something that is mostly beyond 0:08:27.480 --> 0:08:29.640 the historical sample set. So normally, when you're trying to 0:08:29.640 --> 0:08:31.840 model something, you think of what's happened before. You think 0:08:31.880 --> 0:08:34.960 of like Tetlock's book Super Forecasters, where they talk about 0:08:34.960 --> 0:08:37.720 outside in analysis, right, you look at what's happened before 0:08:37.720 --> 0:08:39.880 they might be analogous and say, what are the chances 0:08:39.920 --> 0:08:41.960 that would apply in this case. Of course, there are 0:08:42.000 --> 0:08:44.480 many wars in history, including wars that are happening right now, 0:08:44.920 --> 0:08:47.600 but there aren't that many that involve China, and there 0:08:47.640 --> 0:08:50.520 are none that involve cross rate conflict. And there aren't 0:08:50.520 --> 0:08:54.160 actually that many to involve a major essentially continental power 0:08:54.240 --> 0:08:57.720 seizing an island power, right, And so it actually is 0:08:57.720 --> 0:09:00.719 a quite limited sample set. And what we essentially do 0:09:00.800 --> 0:09:03.360 with think of plausible outcomes and then kind of back 0:09:03.400 --> 0:09:07.200 in relative probabilities at least compared to other outcomes, and 0:09:07.200 --> 0:09:09.719 then say, okay, that tells you, sort of cumulatively, what 0:09:09.800 --> 0:09:11.680 the odds are of these of these scenarios. 0:09:12.160 --> 0:09:14.200 Can you tell us a little bit more about how 0:09:14.240 --> 0:09:18.760 you choose which scenarios to focus on, because obviously you 0:09:18.840 --> 0:09:23.280 can't you can't examine the entirety of possibilities when it 0:09:23.320 --> 0:09:27.320 comes to, to Jared's point, like a relatively unknown and 0:09:27.480 --> 0:09:32.000 unpredictable conflict. But there are clearly some scenarios that are 0:09:32.000 --> 0:09:36.360 seen as more likely than others. And I remember when 0:09:36.400 --> 0:09:39.120 I was working in Hong Kong, we did a big 0:09:39.120 --> 0:09:41.679 piece on the news side of things about, you know, 0:09:42.040 --> 0:09:45.400 this is what a war between China and Taiwan might 0:09:45.440 --> 0:09:47.440 look like. And our source for that was we went 0:09:47.480 --> 0:09:51.120 out and we spoke to geopolitical experts such as yourselves, 0:09:51.200 --> 0:09:53.679 and we tried to get their sense of what the 0:09:53.800 --> 0:09:57.280 most likely scenarios were. But when you guys are doing it, 0:09:57.480 --> 0:10:00.000 who are you talking to, what are you looking at? 0:10:00.679 --> 0:10:03.560 Part of it is a constraint on how you're trying 0:10:03.559 --> 0:10:06.480 to answer the question. So in this case, we're trying 0:10:06.520 --> 0:10:09.840 to model economically what these outcomes are. The range of 0:10:09.840 --> 0:10:11.840 outcomes is more of a gradient. I mean, in a 0:10:11.880 --> 0:10:14.240 technical sense, there could be thousands of outcomes, right, but 0:10:14.280 --> 0:10:16.320 you can't actually model all of that. So what we 0:10:16.400 --> 0:10:18.679 tried to do was decide what are the sort of 0:10:18.800 --> 0:10:22.400 key outcomes that are more or less analogous within themselves 0:10:22.480 --> 0:10:24.520 than try to model that. But I should also say 0:10:24.520 --> 0:10:27.120 we're putting out other pieces on the terminal that are 0:10:27.120 --> 0:10:30.439 actually more detailed and maybe more nuanced than you might 0:10:30.480 --> 0:10:32.520 have seen on the sort of econometric side in the 0:10:32.520 --> 0:10:33.000 big take. 0:10:34.080 --> 0:10:35.839 No, I think this is exactly right to kind of 0:10:35.920 --> 0:10:40.200 underscore the point on some of these scenarios have more 0:10:40.840 --> 0:10:44.800 direct economic implications than others. So, for example, there's a 0:10:44.800 --> 0:10:47.480 lot of different variations of what a major crisis industry 0:10:47.559 --> 0:10:50.000 could look like short of a war. We chose to 0:10:50.040 --> 0:10:53.520 model a blockade because that's the scenario, even though it's 0:10:53.720 --> 0:10:55.960 one of the more unlikely ones for a lot of 0:10:56.000 --> 0:10:59.400 reasons we can get into that would have more direct 0:10:59.440 --> 0:11:03.600 implications for the global economy versus something like a seizure 0:11:03.640 --> 0:11:07.000 of one of Taiwan's outlying islands, where it would be 0:11:07.360 --> 0:11:11.440 incredibly escalatory and risky from a geopolitical perspective, but the 0:11:11.480 --> 0:11:15.200 impact to markets and to the global economy directly might 0:11:15.240 --> 0:11:19.040 be less so because it's not necessarily directly affecting trade. 0:11:19.679 --> 0:11:21.480 But all that to say, you know, in terms of 0:11:21.480 --> 0:11:23.600 what we looked at for the big take in the 0:11:23.640 --> 0:11:28.520 overall exercise was looking across the spectrum of possibilities from 0:11:28.880 --> 0:11:31.840 sort of best case scenario, by which we define that 0:11:31.920 --> 0:11:35.840 as sort of an enduring or stable piece, to absolute 0:11:35.880 --> 0:11:39.160 worst case scenario, by which we define that as an 0:11:39.559 --> 0:11:44.120 all out conflict, including a conflict between the United States 0:11:44.160 --> 0:11:48.000 and China, and then everything in between, the status quo 0:11:48.120 --> 0:11:51.079 kind of being the base case, something slightly worse than that, 0:11:51.160 --> 0:11:54.120 which we define as increased tensions, and there there's a 0:11:54.160 --> 0:11:57.160 lot of variation and what specifically that could look like, 0:11:57.559 --> 0:12:01.439 but the economic implications are probably not all that different 0:12:01.600 --> 0:12:07.319 across those various inter scenarios, and then major crisis short 0:12:07.360 --> 0:12:09.960 of a war. Again, a lot of different variation in there, 0:12:10.000 --> 0:12:11.880 but the one that we think would have the most 0:12:12.040 --> 0:12:14.600 impact on the global economy would be something like a 0:12:14.640 --> 0:12:16.840 blocking that directly interrupts Tree. 0:12:33.360 --> 0:12:36.240 So I should just say both Jared and Jenny have 0:12:36.400 --> 0:12:40.400 mentioned a big take, which is basically a Bloomberg piece 0:12:40.520 --> 0:12:43.440 based on the report that they did for Bloomberg Economics, 0:12:43.640 --> 0:12:48.160 where they're estimating the exact or rough price tag of 0:12:48.440 --> 0:12:51.680 some sort of military action between China and Taiwan. They're 0:12:51.800 --> 0:12:55.080 estimating the price tag at about ten trillion dollars, which 0:12:55.080 --> 0:12:59.000 would be equivalent to ten percent of global GDP, which 0:12:59.000 --> 0:13:03.560 would be more more than the loss from Russia's invasion 0:13:03.600 --> 0:13:07.120 of Ukraine, the COVID pandemic, and the global financial crisis. 0:13:07.200 --> 0:13:09.920 So some pretty big numbers there. But Jenny, I'm curious 0:13:09.960 --> 0:13:13.079 why you say that the blockade is in some respects 0:13:13.520 --> 0:13:17.480 a less likely scenario here, because my impression was that 0:13:17.480 --> 0:13:20.400 that was kind of becoming I don't want to say consensus, 0:13:20.400 --> 0:13:23.600 but like it was certainly a realistic possibility of what 0:13:24.040 --> 0:13:25.640 military action could look like. 0:13:26.240 --> 0:13:28.600 No, that's right, and I think it still remains a 0:13:28.800 --> 0:13:35.080 realistic possibility. I think though, changing military dynamics suggests that 0:13:35.240 --> 0:13:39.040 it is probably one of the tactics that is likely 0:13:39.200 --> 0:13:43.680 lower on Beijing's list of preferred options. And I say 0:13:43.679 --> 0:13:47.680 that because a blockade is technically an act of war, 0:13:48.240 --> 0:13:52.120 would be incredibly risky in terms of tipping into a conflict. 0:13:52.240 --> 0:13:55.320 For example, if China were to put in place a 0:13:55.360 --> 0:13:58.760 true military blockade of Taiwan and Taiwan we're trying to 0:13:59.200 --> 0:14:03.400 challenge that Beijing would be forced to either use military 0:14:03.520 --> 0:14:06.640 force to keep that blockade in place or to have 0:14:06.800 --> 0:14:09.600 to remove it. So it has a really high chance 0:14:09.760 --> 0:14:12.880 of tipping into a conflict, but at the same time, 0:14:13.080 --> 0:14:17.640 it's not guaranteeing an outcome. Right. If Beijing was going 0:14:17.679 --> 0:14:20.600 to risk war, it would seem to make more sense 0:14:21.120 --> 0:14:26.280 to go for the all out amphibious invasion, which has 0:14:26.800 --> 0:14:30.400 the chance of securing control over Taiwan, whereas a blockade 0:14:31.160 --> 0:14:33.480 isn't likely to do that on its own. All that 0:14:33.520 --> 0:14:38.520 being said, most of our understanding of what the Chinese 0:14:38.680 --> 0:14:42.160 military strategy might be for Taiwan and in general how 0:14:42.240 --> 0:14:45.920 amphibious campaigns are carried out, we think that a blockade 0:14:45.920 --> 0:14:49.560 would be likely part of their campaign, part of a 0:14:49.680 --> 0:14:53.040 multi step process. Were they to try and undertake an 0:14:53.080 --> 0:14:57.040 amphibious invasion of Taiwan to try and keep enemy combatants 0:14:57.080 --> 0:15:01.200 from leaving, to try and keep any third part from intervening, 0:15:01.560 --> 0:15:04.080 it would be part of it, but to launch it 0:15:04.160 --> 0:15:08.720 in advance would also give a lot of time in 0:15:08.880 --> 0:15:14.080 warning and the tendency and more recent military campaigns has 0:15:14.120 --> 0:15:17.800 been to try and take advantage of the element of surprise, 0:15:18.080 --> 0:15:20.440 and you lose that with something like a blockade, which 0:15:20.440 --> 0:15:24.560 is a very obvious demonstration of force. But all that 0:15:24.600 --> 0:15:27.800 being said, it could still very much be a realistic possibility. 0:15:28.160 --> 0:15:31.480 Particularly one of the things that we've talked about too 0:15:31.840 --> 0:15:34.640 is it might not take the form of an overt 0:15:34.680 --> 0:15:38.440 campaign or an overt blockade. It could take the form 0:15:38.480 --> 0:15:41.400 of something that looks more like what we call lawfare 0:15:41.600 --> 0:15:46.120 or legal warfare, enacting, for example, a customs regime where 0:15:46.160 --> 0:15:49.960 you're trying to subject all craft that are coming in 0:15:50.000 --> 0:15:53.480 and out of Taiwan to Chinese customs. But you'd still 0:15:53.520 --> 0:15:55.760 need to enforce that in some way, shape or form, 0:15:55.880 --> 0:15:58.680 So it still bears a pretty high risk of tipping 0:15:58.680 --> 0:16:00.320 into something more connect. 0:16:00.800 --> 0:16:04.480 I have a very i don't know, maybe rudimentary question. 0:16:04.640 --> 0:16:07.640 So I'm looking at this wargame scenario that was conducted 0:16:07.680 --> 0:16:11.400 by the Center for a New American Security and in 0:16:11.440 --> 0:16:14.320 their report they have like this map that looks like 0:16:14.320 --> 0:16:16.960 a big risk board and they even have dice, so 0:16:17.120 --> 0:16:18.920 like it's like, let's say, is that really like a 0:16:18.960 --> 0:16:22.760 part of these sort of geoeconomics analyst toolkit to actually 0:16:22.800 --> 0:16:25.360 have these sort of dice games and maps because you 0:16:25.400 --> 0:16:27.800 see them in movies and the general has that stick 0:16:27.840 --> 0:16:29.760 if they push things around. But is that actually like 0:16:30.000 --> 0:16:31.640 a real part of how this is done or is 0:16:31.640 --> 0:16:34.000 that just an affect for the report? 0:16:34.480 --> 0:16:38.720 The war gaming industry is skewed heavily towards actual military scenarios. 0:16:39.000 --> 0:16:42.040 The actual economic war games are quite rare. I mean 0:16:42.040 --> 0:16:44.880 that's growing in part because of Taiwan issue. I have 0:16:45.080 --> 0:16:47.720 played in a number of these, including think tanks and 0:16:47.760 --> 0:16:51.600 in US government. The economic side is typically less sophisticated 0:16:51.720 --> 0:16:53.560 or treated as sort of just an effect or something 0:16:53.600 --> 0:16:56.000 in the background. And I think when you're doing military 0:16:56.040 --> 0:16:59.400 scenarios you can do things like saying, roll the dice. 0:16:59.440 --> 0:17:00.920 If you're fighter jets, take out the other fund. 0:17:01.000 --> 0:17:03.640 They really do that. So there is a serious military 0:17:03.680 --> 0:17:06.840 analyst rolling the dice on a glorified risk board. 0:17:06.960 --> 0:17:10.760 Yes, but it's much with economics. There are far more 0:17:10.960 --> 0:17:14.240 actors involved. And it's not like national decision making is everything, 0:17:14.280 --> 0:17:16.840 because how do you simulate quote the market right? Yeah, 0:17:17.000 --> 0:17:19.840 and or supply chains for that matter. And so the 0:17:19.880 --> 0:17:23.080 body of let's say, economic war gaming literature when it 0:17:23.080 --> 0:17:25.960 comes to Taiwan is much much smaller than the military 0:17:25.960 --> 0:17:26.920 wargaming literature. 0:17:27.920 --> 0:17:30.520 Well, why don't you talk to us about the economic 0:17:30.680 --> 0:17:33.600 modeling aspect of this, because it seems hard enough to 0:17:33.760 --> 0:17:37.600 game out what military action could actually look like, but 0:17:37.680 --> 0:17:40.760 then when you start trying to estimate the cost of 0:17:40.880 --> 0:17:44.560 various types of action, that seems even more complicated. But 0:17:44.640 --> 0:17:47.880 you've put a number on it, ten trillion dollars, So 0:17:48.040 --> 0:17:49.560 how do you actually go about doing that? 0:17:50.160 --> 0:17:52.720 Sure? So, first I want to give credit to our 0:17:52.760 --> 0:17:56.879 outstanding modeling team in Europe, mostly as part of Bloomberg Economics. 0:17:56.880 --> 0:17:58.760 They did most of the actual mathwork which Jenny and 0:17:58.800 --> 0:18:02.080 I were doing was helping to define the scenarios and 0:18:02.080 --> 0:18:05.119 then working with them to come up with essentially like 0:18:05.240 --> 0:18:08.159 scenario inputs that will be relevant to those scenarios that 0:18:08.200 --> 0:18:12.480 would have relevant economic outcomes. And so we already talked 0:18:12.480 --> 0:18:15.159 about how we develop scenarios. When it comes to the 0:18:15.200 --> 0:18:19.600 actual modeling, you know, it's essentially layering on three things. 0:18:20.440 --> 0:18:23.560 The first is a sort of standard wto trade shock 0:18:23.640 --> 0:18:25.920 model which people use for tariffs and other stuff like that. 0:18:26.400 --> 0:18:29.360 We played with tariff rates in some ways to simulate 0:18:29.359 --> 0:18:31.480 the effect of sanctions. These models don't really have a 0:18:31.560 --> 0:18:35.080 sort of sdning the PBOC effect button in them, but 0:18:35.119 --> 0:18:37.320 you can get some of the trade shocks in there. 0:18:37.440 --> 0:18:39.920 There's also a financial model that is layered on top 0:18:39.960 --> 0:18:41.919 of you play with things like the vics and you 0:18:41.960 --> 0:18:45.080 can see how market uncertainty or capital and flows move 0:18:45.440 --> 0:18:47.960 and what the economic outcome or effect is of that. 0:18:48.280 --> 0:18:50.560 But actually, in this case, because we're dealing with Taiwan, 0:18:51.000 --> 0:18:54.480 the most important factor in terms of the cumulative economic 0:18:54.520 --> 0:18:57.200 impact was supply chain impacts. And I know you guys 0:18:57.200 --> 0:18:58.600 have talked a lot about this on the show, the 0:18:58.640 --> 0:19:02.760 importance of Taiwan, seeing particular for advance semiconductors. And what 0:19:02.760 --> 0:19:06.040 we did with that was talk to various industry research 0:19:06.080 --> 0:19:09.879 professional people, including at Bloomberg, who work on these things, 0:19:10.359 --> 0:19:14.119 and using the OECD trade and value added data and 0:19:14.160 --> 0:19:16.840 input output tables, so essentially estimate at the end of 0:19:16.880 --> 0:19:19.880 the day, how much of certain sectors, and these are 0:19:19.880 --> 0:19:22.360 generally high level sector, it's just how the data is presented, 0:19:22.720 --> 0:19:25.280 how much of those would get knocked out if, for example, 0:19:25.600 --> 0:19:28.960 the world did not have access at TSMC's chips. You 0:19:28.960 --> 0:19:30.680 could debate how bad it would be, and in fact 0:19:30.680 --> 0:19:32.479 we got a range of estimates. We took sort of 0:19:32.520 --> 0:19:35.679 something in the middle that is like awful but not 0:19:36.359 --> 0:19:38.920 completely apocalyptic. But I think if you look at the 0:19:38.920 --> 0:19:40.960 bottom of the big take, there's a note explain this 0:19:41.040 --> 0:19:44.000 that essentially we knocked out in the war scenarios something 0:19:44.040 --> 0:19:47.000 like eighty five percent of global electronics production, so things 0:19:47.040 --> 0:19:50.200 like smartphones and all that computer's laptops, and then something 0:19:50.280 --> 0:19:53.040 like sixty two percent of global autos and transport. So 0:19:53.400 --> 0:19:56.040 that's a massive shock, and it's actually doing a lot 0:19:56.040 --> 0:19:58.800 of the work in terms of the simulation of economic impacts. 0:19:59.119 --> 0:20:01.080 Yeah, and just to add to that, shout out to 0:20:01.119 --> 0:20:04.360 you our Bloomberg Intelligence colleagues who helped us better understand 0:20:04.440 --> 0:20:07.440 the role that these chips play in different industries. And 0:20:07.840 --> 0:20:09.800 the point that Jarred noted on what we call the 0:20:09.840 --> 0:20:13.399 golden screw problem of trying to determine you know, a 0:20:13.560 --> 0:20:16.600 chip itself might not cost that much, but it's hard 0:20:16.720 --> 0:20:20.119 sometimes to know for sure how essential it is to 0:20:20.160 --> 0:20:22.720 a particular product. You know, for example, could the car 0:20:22.880 --> 0:20:26.240 be made without that chip? Is there the possibility that 0:20:26.280 --> 0:20:30.320 the manufacturer could substitute for something else, And as Yard noted, 0:20:30.520 --> 0:20:33.560 we worked with our BI colleagues to kind of estimate 0:20:33.560 --> 0:20:37.439 within that range of how essential these chips are. But 0:20:37.840 --> 0:20:42.280 our rough estimate is that chips we know are really 0:20:42.280 --> 0:20:46.080 important to high end products like smartphones and PCs, but 0:20:46.119 --> 0:20:49.879 they also play a really critical role in other major 0:20:49.920 --> 0:20:54.080 industries autos, home electronics, et cetera. And I think there's 0:20:54.119 --> 0:20:56.760 a huge question of not just what happens if those 0:20:56.800 --> 0:21:00.480 products can't get produced, but everything that relies on those products, 0:21:00.680 --> 0:21:04.520 in particular in our more advanced service economies, what that 0:21:04.680 --> 0:21:07.280 means going forward if you can't get new smartphones, if 0:21:07.320 --> 0:21:09.960 you can't get new autos, and if you can't get 0:21:10.040 --> 0:21:14.440 new PC. So I think this is an important figure, 0:21:14.560 --> 0:21:18.240 the ten tillion dollar one and the impact that we're 0:21:18.320 --> 0:21:21.399 estimating to advanced economies, But it is in some ways 0:21:21.400 --> 0:21:24.119 the beginning of the conversation, because there's more work that 0:21:24.119 --> 0:21:27.119 can be done to really refine those numbers and then 0:21:27.480 --> 0:21:29.359 estimate the downstream impact. 0:21:29.760 --> 0:21:31.720 I want to ask a question that actually takes this 0:21:32.080 --> 0:21:34.600 step way back. And you know, as Tracy mentioned in 0:21:34.640 --> 0:21:36.600 the intro, and I didn't know that you wrote a 0:21:36.640 --> 0:21:41.160 thesis almost almost twenty years ago predicting an imminent. 0:21:42.320 --> 0:21:48.280 Now I feel old, old and wrong, sorry, but why now? 0:21:48.520 --> 0:21:51.159 Why are we this seems to go in waves of 0:21:51.160 --> 0:21:53.879 people concerned, why do you like set the stage for 0:21:53.960 --> 0:21:55.720