Host (5:26)

Great. I think, you know, one of the things that I found really interesting because, you know, in Our lifetime, we've never really seen tariffs. And I think one of the, you know, we wanted to start with this first, first chart that you opened with, which shows the historical tariff rate in the US going back to like 1820 or something like that. And you know, can you just explain sort of what we're sort of seeing here from a historical context?

Kai (5:51)

Yeah, so the, the latest data point, this is as of April 15th. So this was after, so the liberation aid announcements came out first with these so called reciprocal tariffs. You might remember, like Vietnam was higher than China at the time. And then at some point Trump decided to remove these tariffs or pause them at least on other countries, but kept the Chinese ones intact and in fact hike them to 145%. So I believe this dot here reflects the tariffs as of that date. I should mention we're recording on Monday, May 12, and as of this morning it's been changed again. So we'll see where it lands. But at least at the time, it's pretty interesting just because if you put this into historical context, this shows the tariff rate for U.S. imports over the past almost, I guess, 200 years. So you can see this over many, many political, presidential administrations through world wars, et cetera. And this, this shows, I think, just pretty obviously how big a deal this actually is and that the fact that the market is paying a lot of attention to tariffs is, is actually totally rational. Right? To the extent where the average tariff rate were to be raised to levels not seen for 120 years since 1900, that is a significant, that is a really big deal. And so I think this gives impetus to the analysis. And you know, it also gives credence to, you know, I guess in, in the intro I, I kind of ripped on investors a little bit for, you know, kind of getting spooked by the tariffs. But look, it's, it's, it's completely rational. I mean, to the extent these, these things do remain, it will create a lot of friction for companies attempting to do global trade.

Host (7:27)

So when Trump and the administration was sort of prepping the market for this, it was, there was kind of like three things at different times they were saying that they were trying to accomplish with the tariffs. One was like a fairness issue, like other countries had have tariffs on our goods and so it's not fair, so we should have tariffs on their goods. Another was, you know, sort of supply chain and bringing industries that are important to national security sort of back. So sort of like a deep global globalization sort of idea. And then also the, like these Trade deficits are, are sort of bad. Like, if we have a big trade deficit, that's like kind of, again, like not good and unfair, and we should be trying to balance that out. So do you have any. I'm kind of grouping all these things because at different points in time, they were kind of making different arguments, which I think was one thing the market was a little bit unclear about. But just in general, you know, commenting, what are your thoughts on those things? This isn't a chart specifically. I'm just wondering what you kind of think on that.

Kai (8:26)

Yeah, look, look, I think, I think it's a challenging thing because as you're pointing out, there are, like, a lot of different objectives in play here and there's a coalition of political support for, you know, Trump's policies that is, you know, pretty, pretty fragmented. There are additional things as well to what you mentioned, say, like raising, like, revenue so we can now get rid of the income tax or whatever. Right. Like, there's a lot of different, you know, things that we're trying to accomplish with these tariffs. And the problem is that, you know, it's not really possible in many cases to do all of them together. Right. In many cases, they conflict with each other, you know, and I think it becomes a somewhat partisan issue. So, you know, I don't know if we should get into too much the details here. Right. Like, because you have to remember, even under Biden, there is industrial policy as well, the CHIPS act and such. I think it was a goal of his as well, to reshore at least the strategic industry, such as chips and drugs, to some extent. So, yeah, I mean, the reason behind it almost doesn't matter because at this point for investors, it is what it is. And I think the goal that many investors have kind of on the more macro side to try to predict what's going to happen next, I think is really, really hard. You know, I don't know. You know, it doesn't even feel like just watching, you know, the, the cabinet members speak that there's even a cohesive understanding within the Trump cabinet, let alone amongst, you know, us, you know, less informed, you know, kind of average public market investors. So, yeah, I feel like that's a challenging thing. And so the best we can do is to attempt to build resilient portfolios that will, you know, be exposed to the tailwinds of globalization, which I do expect to remain, but, you know, while also mitigating some of the, some of the risks in various tail scenarios.

Host (10:10)

So in this next chart, which I think it's the top 10 largest companies in the S and P at the time. Correct me if I'm wrong on that.

Kai (10:18)

But I snuck in TSMC at the end, but yes.

Host (10:21)

Okay. Yeah, there they are. Okay, so, yeah, just what were you. How did you define this? And you know, talk about, you kind of mentioned these multinationals are maybe higher quality. But anyways, I'll let you explain sort of what we're looking at here.

Kai (10:35)

Yeah, so, yeah, this is the top 11 largest companies by free flip market cap. So, you know, Saudi Aramco does not make the cut here. And I think what was. What's interesting is, you know, these are basically like, these are like the kind of winners of the current era of global capitalism. These are the companies that, you know, for those of us who invested in Berkshire or Apple or any of these companies, like, they have generated tremendous wealth for investors. And, and I've obviously created, to the extent many of the companies are American, a lot of jobs and other things. But what's interesting here is if you actually go through the names one by one, what you find is that with the exception of Berkshire Hathaway, these companies are all multinationals. So Apple, Microsoft, Nvidia, Amazon, on average, 50% of the revenue of these companies is derived from outside the US or abroad. About 58% of the production occurs outside of each country. Each company's home country and 43% of employees are, you know, located outside of each country. So tsmc, the reason why I kind of snuck this in is kind of interesting is because it's an example of a company that has, you know, a global footprint with regards to revenue, but most production is done within its home region, Taiwan and China. But if you focus on the US Companies, yes, they are very multinational. Amazon's interestingly, one of the outliers here in as much as their foreign revenue exposure is only 31%. So it's still high, but less than its kind of peer group, despite having foreign production, obviously they import a lot of goods from other countries, so they kind of be considered a borderline importer. But 31%, at least by the arbitrary threshold we chose of 25%, still makes them a multinational, you know, according to our definition.

Host (12:27)

And then you were able to. So you did this across all stocks in the major indices, both US S&P 500 and some international indices. And you found. I'll let you kind of comment on, you know, what we're looking at here with the indexes.

Kai (12:46)

Yeah. So the point here was, you know, the, the prior exhibit we showed like the world's leading companies. But you know, obviously we want to make sure we can generalize to, you know, the rest of the stock market, which is why showing this aggregate data, while maybe less visually compelling, compelling is obviously very important. And so what I do here is look at the S&P 500, the first column. Right. So 100% of the index, about 80% is global firms, defined as either a multinational, such as those we saw, an exporter or an importer. And then if you go to msci, ifa, which is the developed international index, it's pretty similar, as is acwi, which is the, you know, developed. So the world index, including emerging now emerging markets itself, and the third bar is a little bit lower and it's kind of hard to make out basically on colors. But what's interesting here is that emerging markets, they tend to, you know, have more exporters as a share of their total market, which I guess makes sense intuitively because that's where a lot of the production, you know, all the manufacturing occurs and the richer countries tend to be more net importers of goods.

Co-host (13:51)

So in this next one we're seeing how this has changed over time, which as you can see, it's rising rapidly.

Kai (13:58)

Yeah, and look, this, this speaks to the success of, you know, the multinational or global business models that these companies taking advantage of, you know, gains from trade and, you know, specialization, you know, lower lowering labor costs, you know, getting access to specialized production, say with the SMC has generally been a good strategy. And as a result, not only have we seen the market caps of the companies that have employed this strategy increase, we've also seen other more kind of traditional business start to move more globally to, you know, keep up with their peers.

Co-host (14:29)

You mentioned these different categorizations and this is, this is probably a good time to maybe define those a little bit better. You, you came up with a two by two matrix here. Can, can you talk about how you define exporter, multinational, importer and domestic and then maybe talk about this matrix.

Kai (14:42)

Yeah, so look, the most obvious way a company is exposed to international trade is by its customers. If I sell my goods to the European market or to the Japanese market, right now I have exposure to tariffs in the event that they were to put them on my company. But there's another dimension that's really important too, which is production. To the extent where I rely on, say production in China, that's an important source of potential risk. And this can come in two forms. It can either be internal. So in other words, if I have Factories located in China, right. I still need to ship them across the border to the us or it could be the case that it's an outsourced situation. So for example, Apple, right, they use Foxconn to make a lot of their phones. Foxconn, a third party supplier. So they don't necessarily show up internally on Apple's org chart. That being said, of course, to the extent where there's supply chain disruptions in China, that's gonna be a problematic for Apple. And so what I do here in this matrix is I say, look, there's two dimensions of trade risk. So let's look at customers. For example there I say there's two categories. There's domestic and there's global. And I basically arbitrarily define as a threshold here 25%. So any company who derives 25% or more revenue from abroad is considered global and anyone that doesn't is considered domestic. And then the same on the production side. And then what I do is I look at the two by two matrix and construct four quadrants. So let's start with the global. Global. So a company with a global production and global customers, I call it multinational. That's not precisely the definition of multinational, but let's use that for now. So think like Novo Nordisk is a good example. On the opposite extreme, there's purely domestic businesses that have both domestic customers and domestic production. UnitedHealthcare is a good example. And then on the off diagonals, let's look at exporters. So exporters are companies that, you know, produce domestically but then export globally. A lot of the European luxury brands fall in this category. And then importers. I mentioned RH earlier. I also have this example of Cognizant, which is a outsourced IT firm. They have about 3/4 of their employees in India, yet are based in the US is another example of a firm who kind of does a production abroad but then sells into a local domestic market.

Co-host (16:54)

This next one I think is great because it really gets at the issue you're talking about here. And you know, people want to classify automakers as where they're from. So you know, Ford is a US automaker and Hyundai is a foreign automaker. But when you get into details about, you know, where they're making these cars, where the parts are coming from, where the employees are, where they sell them, I mean, this is kind of a big mess in terms of what is a, you know, where these companies are based out of. So could you talk about this a little bit?

Kai (17:16)

Yeah, I mean that it's very well said. Like it's, you know, companies are more than just where they happen to be based or, or even listed what a stock exchange you're listed on. And, and if you think about it not in that dimension, but instead along this dimension of these, this quadrants, you get like a different picture. And so like, you know, on the multinational side, you might find like, Stellantis and Toyota, BMW, these are companies that, you know, both have global production and global customers domestically. Harley Davidson is a good example of a company that's, you know, kind of made in America and obviously is targeted towards American consumers. A lot of the Indian carmakers are also for protectionist reasons in this quadrant. You mentioned Ford and gm, they are kind of borderline importers, at least based on these definitions. And the reason why is that they primarily sell trucks to American things. But a lot of the parts are created outside the US and then conversely on the export side. Right. You have, I guess, Subaru or Ferrari is another example, kind of the same as your European luxury brand. And they make nice Italian sports cars and sell them to everyone who will buy one. But I think what's interesting about this exhibit is the auto industry is kind of exceptionally international. So compared to other industries, like, let's say utilities being the other extreme. And we do find that in general, the automotive industry is pretty global in its, in its construction. But obviously there's a lot of variance within the industry and that's what makes.

Co-host (18:40)

These tariffs so complicated. Because if you think about all the parts that are going into the cars and where they're coming from and where they're assembled, and it's got to be a mess behind the scenes to try to implement tariffs on something like this.

Kai (18:50)

Yeah, absolutely. And I think the uncertainty is, is the main major killer. I was talking to someone this morning who works in the import export business, and that's the big thing, which is, you know, if we just knew where the tariff would be, we could start like, you know, preparing for that future. But, you know, because it changes so rapidly, it's kind of a wait and see for now at least.

Co-host (19:12)

So this next one gets to what you talked about at the beginning, which is this evidence that global firms do outperform domestic firms. So can you talk about that?

Kai (19:18)

Yep. So in this case, I defined domestic firms as that lower left quadrant and then global firms as any of the other three quadrants. So it could be a multinational, it could be exporter or an importer. And what we do here is we say, let's now build Portfolios of companies in each of these categories we will rebalance monthly. I think we market cap weight here. And again, this is a backtest, it includes transaction costs, you can't invest in these things, so on and so forth. But we do find significant relative outperformance of global versus domestic firms over this period. And this kind of goes to the point I was making before that these are outperforming companies. And to the extent one was to say let's switch from the blue line to the red line moving forward, that's not what got us to where we are now. And so it's maybe less obvious once you see this exhibit that that should also be the path forward.

Co-host (20:13)

And one of the questions you hear all the time these days is whenever you see anything that works better than something else, we've got to test it on equal weighted and market cap weighted because we've been in such a market cap weighted dominated environment. So what did you find there?

Kai (20:23)

Yeah, look. So this shows the relative performance of global versus domestic companies on a market cap versus equal weighted basis. And so the general idea is that both things go up more recently. The MAG7 have been magnificent. Seven stocks have been dominant the past few years especially. And so you do see that the outperformance for the market cap weighted version of the strategy was extra good the past five years or so. But either way, I think the point stands that you know, even if you're a small or mid cap domestic firm, sorry, global firm, you would have outperformed your domestic peers over this time period.

Host (21:01)

Another thing is as we talk about.

Co-host (21:02)

The market cap weighted indexes that I think a lot of people misses when we're talking about trade is this idea that it's not just goods, it's services. And so for the US we are an importer, a net importer of goods, but we are an exporter of services and those are high margin services. So it's important, I think to keep all of that and take all that into account when you're looking at this overall trade thing.

Kai (21:20)

Right? Absolutely. I mean that's that the Apple's the best example, right? The largest stock, which is, you know, Apple's made a decision at some point in through time to outsource the kind of, let's call it low, lower end manufacturing pieces of their business and instead they retain kind of the highest value pieces like the design and their ip, et cetera. And so I think that there is definitely something to be said there to the extent where the, the Trend, the past 15, 20 years has been outsourcing production to emerging markets and retaining the highest value aspects in their home countries. I think that definitely is a important trend that we've seen and I expect to continue. But obviously we are definitely facing a bit of an inflection point with regards to that.

Co-host (22:13)

This next one is eye opening for me because I didn't know a lot of this, but here you're looking at.

Kai (22:16)

The global firm share of, of all.

Co-host (22:18)

The country stock indexes across the world. And there's some interesting stuff in here. So was there anything that stood out to you?

Kai (22:23)

Well, I think the main reason why I, I kind of included this chart was, you know, I, I kind of spent the first, you know, few exhibits of this paper talking about like, you know, the US multinationals where everyone knows Apple and Nvidia. But when you actually step back and look at the data, you know, more holistically, interestingly what you find is that it's actually Europe that is the most, most multinational. What I did here was I said for each country we do the classification I mentioned and then I say what percentage of each country's individual index is composed of global as opposed to domestic companies. And the US falls as number 18. So it's not even that high really. It's the EU companies that are the highest. Interestingly, China and India end up being pretty low. And I think that's largely due to their kind of very massive consumer markets, at least relative to many of the other countries, kind of within their peer group.

Co-host (23:11)

And so here we look at it by sector. So this was interesting as well. What'd you find here?

Kai (23:15)

So I think as everyone knows, it's kind of the IT sector and in particular computer hardware that is the most exposed to global trade, right? These, these hardware supply chains, can they wind through, you know, Taiwan, China, Vietnam, et cetera. But it's not just the, it's not just it. Materials, industrials, healthcare, consumer discretionary and staples all constitute very global sectors. On the other hand, I'd say that utilities, real estate, financials, energy and communications are considered more domestic. Communications is a little bit weird because you have like, you know, Google in there, but then you also have your more traditional telecoms which are of course, you know, much more domestic.

Co-host (23:56)

And this gets into what we were talking about before in terms of services being really important. You know, if you look at it being the most globally exposed industry.

Kai (24:03)

That's right. I mean, again, like it is kind of weird. I mean all these are weird, right? Because you can always decompose these sectors into Industry groups and industries which get to the different levels of granularity. And even here you do find bifurcation. Right, because software can't really be tariffed. Like how would you take care of your software? I mean, they have like digital services taxes in the EU now, but for the most part they're not really able to be tariffed in the same way hardware like an iPhone would be able to be. And so if you were to break this out further, the exposure might change a little bit. Semiconductors, of course, are another also captured in it. We all know that semiconductors are very, not only a strategic industry, but also one where in the US we do want to try to reshore to the extent we can because of our reliance on TSMC and Taiwanese and Chinese production of semiconductors.

Co-host (24:52)

So once you realize that there is outperformance among these global firms, one of the things we always do in the factory world is we try to break that down and figure out what's driving that outperformance. And so here you're getting at that, looking at stock selection versus industry selection. So what'd you find here?

Kai (25:04)

Yeah, one of the really important things to do is to kind of make sure that this effect isn't solely due to a sector bias. In other words, okay, so you have more. So global firms tend to be more tech forward. So therefore maybe that's just explaining the entire thing. Right? Because what we find is that yes, it is the case that say in technology, the technology stocks tend to be more global than say, utilities. That being said, within any industry there's also dispersion. So for example, intel is more domestic than Qualcomm, or I have an example of say Disney being more domestic than Netflix within the media space. So there's a lot of dispersion within the industry as well. And so what we can do is we can do an attribution where we say these excess returns which we found, are they attributable to the fact that the strategy selected the industry that then went on to do better, or they selected the best companies within each industry and that actually led to the outperformance? And we find relatively even contributions from bulls. So in other words, yes, being in kind of more global industries was helpful, but also being in the most global stocks within each industry was also helpful.

Co-host (26:13)

So we're going to talk here a little bit about why you talked about global firms outperforming and we're looking at some fundamentals and talk about why. But just I want to ask first, at a high level why would we expect that? Like, why would we expect global firms to be better business wise than firms that are not global?

Kai (26:28)

Yeah, I think there's like three things. So, so first is just like the, the fact that, you know, global markets are larger and so firms that can then spread their fixed costs over a larger customer base can generally get better economics, better unit economics. The second thing is on the production side, on the cost side, there's a couple of advantages. So first, you know, if I were to outsource my production to a cheaper country, right off the bat we get cost savings. But the other kind of more interesting aspect is, you know, there are, you know, countries and companies outside the U.S. let's say, that are, you know, very specialized. So I mentioned TSMC with, with semiconductors. You know, also Swiss washers come to mind where it's not just about saving costs. It's the fact that these countries or companies had developed very, very kind of like niche specialties in certain areas. And it's not even a matter of saving cost. At no cost currently could we make a cutting edge semiconductor in the US at this point in time. Right. It's not even just about saving money, it's just that's a lot of the human capital and tooling has developed around these countries. It takes some time to rebuild it. I'm not saying it can't be done, it's just that there's that component as well. And the third thing, and the third reason why I think these companies may have outperformed is simply due to selection bias, which is that if you were a top tier company, then you're going to want to play on the global stage. If you compete on a global stage, the returns to being successful are higher, but also the competition is higher. Whereas if you are a kind of less good company, maybe you would just become comfortable competing in your own narrow domain and not having to face as much competition from the big boys abroad.

Co-host (28:11)

The TSMC thing is interesting. We're not going to talk about politics here, but like as a country you have your own national security concerns. Like it's not great for us that these chips that we need are being made in another country. But then you've obviously got the cost thing, which is it's going to be very costly and challenging and like you said, we can't even do it right now in terms of making those things here. So that's an interesting balance, you know, with that type of industry.

Kai (28:31)

Yeah. And I think, you know, there's a reason why that's one of the explicit goals both of Biden and Trump. So this is a bipartisan issue, right? Is to reshore the strategic industries such as, you know, defense related stuff, semiconductors, maybe energy and drugs. Whether or not we want to restore Nike apparel factories from Vietnam, that's I think a more debatable question. But I guess it just comes down to kind of how each politician and how each party seeks to accomplish this. There's different means to that same end, but it does feel like it's a relatively bipartisan thing at this point.

Co-host (29:06)

So you talked about the benefits of globalization and here we're kind of seeing it. So what are we seeing here in terms of the profitability of global firms?

Kai (29:12)

Yeah, so we see across the board that global firms, this is as of the most recent date, so March 31, 2015, are significantly more profitable than domestic firms. So look at The ROA is 30 versus 15, ROA is 10 versus 5. Net profit margin 17 versus 15. And then free cash flow margin 14 versus 4. Interestingly, if you bring this back through time, it's pretty consistent over time. So, you know, global firms have actually widened their advantage somewhat. But even say 15 years ago, they were still better quality firms. And if you adjust by sectors, which I don't show here, or by countries, you kind of find the same effect, which is these global firms do tend to have higher quality businesses than their domestic peers.

Co-host (29:59)

So this next one gets to what we talked about at the beginning, but you can really see the magnitude of this. So what is the Trade Policy uncertainty index? Can you explain that? And then what's happened recently with it?

Kai (30:08)

Yeah, so it was a bunch of Fed economists that actually put this together. And what they did was they said let's comb through a bunch of very kind of well known news sources. I think like New York Times and Wall Street Journal fall into this. And they say, let's look for mentions of trade policy uncertainty, specifically U.S. trade policy uncertainty, and kind of create some sort of index based on how high that number is as a percentage of the total mentions of various things. And you find kind of episodic spikes around various trade shocks, Japan, NAFTA, et cetera. And then what you see is in 2018, a huge spike around the first Trump trade war with China. And then in 2025, an even bigger spike. And just for those who don't have the video, we're talking about a trade index of, let's say that peaks around 100 for about 50 years and then it peaks at 250 or 300 in the first Trump trade war and then peaks at boy, it doesn't peak. It exceeds at least 1,000 in the current period after Liberation Day.

Co-host (31:11)

And so this next one gets this idea that you would expect global firms to underperform during these periods of trade shocks. And as you alluded to in the paper, I think it's probably pretty challenging to do here because there aren't that many. But what did you find in terms of global firms suffering during trade shocks?

Kai (31:25)

Yeah, so what I did here was I said let's look at periods when this index, trade policy uncertainty index spikes, you know, a lot. And what we find is in those periods, the returns of global versus domestic stocks on an annualized basis. Right. And this is again, a few days annualized isn't super meaningful, but the return is negative 33% versus being positive in all other time periods. So I think like, look like this is there aren't that many episodes of big trade spikes. So I think we have to take this with a grain of salt. That being said, I think it does provide some evidence for this idea. So let's step back for a second. What is the idea we're trying to assess here, which is look, so if it is the case that empirically global firms have outperformed domestic firms, why would that be the case? Because in an efficient market we would expect that valuations would fully bake this in. However, empirically what we've discovered is that they haven't because obviously there's an excess return for global firms, which implies that markets have, markets have allowed global firms to trade at a discount relative to their profitability, adjusted intrinsic value. And so the question is why do they trade at this discount? And there are many reasons. You could argue that that could be the case. I think one very plausible argument would be that they embed a risk premium for geopolitical risk, that investors are not dumb. They recognize that these companies, if they have a supply chain that winds, say through China or Russia, that that is a potentially risky thing to be doing and they should be compensated for that. And again, it's very difficult to know, to disentangle all the different variables. But this at least ex post analysis maybe supports potentially this idea that maybe there is a geopolitical risk premium. And that could be one of the reasons, perhaps among others, that these stocks have outperformed over the past several decades.

Co-host (33:18)

So one of the big questions people have been asking recently is exposure to China because obviously, although as you mentioned, the China tariffs came down temporarily over the weekend, obviously the most aggressive trade policy has been directed towards China. So you looked at China exposure by industry and what did you find?

Kai (33:32)

Yeah, so first I should mention how we do this because I think it's kind of an interesting idea. So the first thing is, as I mentioned, what we're trying to do here is figure out which countries companies have exposure to on the production side. Now the problem is that you can't just go into a 10k and look this up. For the most part, there's no, like, you know, some companies do disclose their suppliers in a structured format, many don't, most don't. And so what we had to do was compile mentions not just from 10Ks, but from corporate documents, like millions of different documents. And this is where some of the large language model tools that, you know, I've written about and talked about in prior podcasts, episodes and such are put into play. So the idea would be that if I were sitting here trying to manually do this myself, I would basically, you know, for a given company, let's say Tesla, I would like look through every single mention in the news and third party sources, trade publications, et cetera, for mentions of their suppliers in different countries. Obviously that's a very tedious task. Fortunately, that's where large language models can do it in a fraction of a second and much more efficiently. So that's one way we do this. The other way we do it is just as a way of triangulating is by looking at the employee, the employment data. So we can say for any given company, say Tesla, go on LinkedIn, where do their employees live? Do they live in the us they live in China, in Germany, and from there form some kind of a proxy for the, you know, the, the footprint of a company that works pretty well, except not when the companies outsource. So to the extent you're talking about Apple, who outsources their production to Foxconn, that's not going to show up. But to the extent you have a company that, you know, pretty much internalizes all its production and vertically integrates, that'll be a decent proxy. The point being that because these are both have their strengths and weaknesses, we kind of blend them to create a kind of composite measure of exposure. So now with that in mind, we can then go through every single company through each point in time and say, what is it as exposure to country X, Y and Z at this point in time? So this is the snapshot you can see today in Exhibit 24 for the current point in time. What we do is rather than show every single company, we aggregate by industry. And what we find is, and this is for US Companies only, we find that the industry with the most exposure to China is technology, hardware, followed by semiconductors, automobiles, retail, consumer durables, consumer services, pharma and biotech, capital goods, media, entertainment, materials. And so, yeah, I mean, this is again, pretty much lines up with intuition that it's really these tech, tech, hardware, semiconductor conductors, and then autos, retail and apparel that are heavily exposed to Chinese production.

Co-host (36:09)

What's interesting, I don't know if you see it in the data, but is the ability of firms to ship that production to other foreign countries. So Apple, for instance, I believe, has been shipping the majority of its US iPhones from India since this happened. They had at least enough capacity there to do that temporarily and shipping to everywhere else from China. So that ability to pivot is probably very key. And that's probably hard to see in standard data. What, what firms have the ability to do that.

Kai (36:31)

Yeah, one of the nice things is that the LinkedIn data in particular go back pretty far in time. Right. Because, you know, people will post in their bios, oh, I used to work in Hong Kong, then I moved to, you know, China, et cetera. And so you can actually look at each point in time what say non US, non Chinese companies, employee share in China was. So in other words, like, you know, think like a given company, what percentage of their employees were in China at each point in time. And what you find is that that number steadily increased for many years and then peaked as early as 2009. So right after the financial crisis, companies executives started to realize, yeah, maybe we have a kind of unhealthy dependency on China. Why don't we start diversifying? Right. So this was the kind of China plus one idea where they said, let's, you know, start putting production in India, in Vietnam, Mexico. And then, you know, that accelerated again during the COVID period when we had this kind of like Covid zero policy in China, which is very disruptive to supply chains. And so, you know, companies, you know, the idea was this idea called French shoring as well, right. This, when this kind of idea came into play and companies said not only should we diversify China for cost and logistical reasons, there's also a geopolitical element too. And this is again following the first Trump China trade war and then obviously today. And so what's really interesting is if you look at the data on employees I mentioned, they peaks in 2009. It's been steadily decreasing since then. And I think this does point to the kind of Flexibility of these multinationals to be able to kind of redeploy supply chains, you know, through the countries that make the most sense, whether for cost, logistical or political reasons. Again, this is a slow process. Right. It's, it's been going on now for I guess 15 years that companies have been gradually diversifying away from China. And that's the problem. Right. Which is, you know, companies are certainly able to do this. It's just a matter of like, you know, having enough certainty to be willing to make these multi year investments. You know, when, you know, at this current point in time we don't, as a, you know, if you're a CEO of a company, you don't really know where things are going to end up landing. Yeah.

Co-host (38:29)

And I would think if anything this would be a pretty big accelerator of that trend. Regardless of how this plays out. I think this maybe was a, you know, a shot over the bow to some companies to say like, we've got to diversify our supply chain to multiple countries in the future. I mean, obviously like you said, it's going to take time to do it though.

Kai (38:42)

Yeah, I mean, I agree with you. But I think what's interesting about the data is that it suggests that companies have already been doing this. Right. It's not like, you know, a lot of companies have already been doing it. Now that Chukanj was, you know, you had all these companies that say, left China and went to Vietnam and then, you know, Liberation, Liberation Day comes around and the Vietnamese tariffs are higher than the Chinese tariffs. Right. So that's never fun. It's a bit of a bait and switch. But you know, that's, I seem to have been gone away and at least for the time being it does seem like the, you know, tariffs will be or are highest on China and then next on, I guess Canada, Mexico and then the rest of the world.

Co-host (39:14)

So you were also able to look at your returns of global firms. But to do it comparing all global firms to X China global firms. So what did you find there?

Kai (39:23)

Right, so what I did here is I took the original exhibit with the blue and the red line. So global firms versus domestic firms and we saw outperformance for global firms. Then I said let's create a third line where what we do is we say take the universe of global firms and exclude those with the most exposure to China and then reallocate that say to the remainder of the portfolio. And interestingly what I find is that it's basically no change in returns. Like it's like no, no Difference, more or less. So, which is a little bit counterintuitive. Right. Because you would expect it to be the case that, you know, to the extent where you are, you know, there should be an opportunity cost, I guess, of excluding firms that do business with China. Right. But it appears to, at least over this historical period, which again may not be representative moving forward, that that has not been the case.

Host (40:09)

You also looked at supply chains. So you have, well, first of all, how did you measure, you have in this chart resilient supply chain. So how did you measure that? And, and what did you find with like resilient supply chain firms versus the rest of the universe?

Kai (40:26)

Yeah, so what I did here is I did two things. First is I said it's probably the case that companies that have Expo, have all their suppliers in one country are heavily exposed. That's probably a bad thing. Right. And conversely, companies that have a very diversified supply chain, say, have production across many countries are in a better position because if any one country gets tariffed or some disruption occurs, you just ramp up production in, you know, a factory, you know, that's in an unaffected area and that's, that's helpful. But the problem of course is what happens if we get across the board tariffs like we have. Then what you need to do is, you know, follow a strategy of localization. Right. So this is the best example will be Tesla. So Tesla, you know, most of their cars are produced in the US of course, but then around the time of the Chinese China trade war under Trump won, they actually ended up building a gigafactory in Shanghai and then one in Berlin. And the idea there, among others, right, is to, you know, produce Chinese cars for Chinese consumers, produce European cars in Germany for European consumers. Right. And there that allows you to get around the tariffs that the EU or China might impose on US manufactured cars. And so that sort of strategy can be applied anywhere and it's called localization. And so that's the idea here, which is companies that have a distribution of production and sales that are similar are in theory more robust than companies that have a mismatch. Like if you do all your production in one country and all your sales in another country, because then now you're relying on there being no barriers to trade between those two countries. So we can measure that using a measure called the Hirschman Index hhi, looking at the region level production versus revenues for different companies. And so anyways, the point being that companies that have diversification in supply chains and then have an alignment or high similarity between their production and sales footprints are considered resilient. And if you go to the exhibit, what you find is that, you know, the, the companies that are global but also resilient tend to actually outperform historically, just simply their global peers.

Host (42:39)

One of the things that you're most widely known for is this idea of intangible value and the way that you measure it in the ratios that you create at the company level and sort of how you build portfolios. That's how you actually pick the stocks that you own in your ETFs. But this next chart shows the difference between global firms from their intangible value and the ratios that you used versus the domestic firms. So what did you find here?

Kai (43:06)

Yeah, so, I mean, from a high level, the finding was that global firms are significantly more intangible intensive than their domestic peers. So take one example, like so, R and D divided by price. In other words, for each dollar, as an investor you put into a company, how much R and D capital do you get out? It's about 2.5 for global and 1.1 for domestics over 2x. The same goes for patents and PhD employees and marketing expenditures and a variety of other metrics. I think the next question to ask is why this might be the case. I think the most obvious thing is that intangible assets have unique properties of which one of the most salient is scalability. Oftentimes you must invest high costs upfront to develop, say, a patent on a drug or software. Yet then the marginal cost of distributing that product is close to zero. Right. And so to the extent where a company needs to, you know, make the expenditure either way to create the intangible asset, they might as well sell to the broadest customer base in the world, which of course implies a global customer base. Like, why would you spend all the work doing this and then just sell it to, you know, people who live in New York? And so that's, I think, one important reason. The other one is slightly more minor. But there's a significant tax advantage for many companies who are both intangible and operate globally. So, for example, it's quite common to see companies, you know, park their IP in, say, Ireland and then license it to their subsidiaries abroad. And again, if you are domestic, you don't have the ability to take advantage of some of these loopholes, I guess.

Host (44:40)

So then you were able to kind of say, okay, let's look at high and low intangibles in the global universe and the domestic universe. And let's look at the returns going back to to 2007. So can you talk to the findings in this chart?

Kai (44:52)

Yeah. So this is again another backtest, but the idea was let's separate the world in the quadrants this time on the dimension of global versus domestic and then high versus low intangible value. And so if you look at just the blue lines, what you find is that within this global universe there's a pretty significant spread between high intangible and low intangible value companies. And then if you go to the domestic universe, it's the same. You know, there's also a widespread, in absolute terms it's kind of wild to see this, which is that in real return space in US dollars across a global index, basically there's been no returns for low intangible value companies, whether in global or domestic over the past 15 years. All the returns have been generated by your high intangible value companies. And this is not in the paper, but of course it's not just Apple and Google, there's other companies in here. And of course there's also the other dimension which is important too, which is that even if you're only investing in high intangible value, those that are in the global universe have even done better than those that are just in the domestic universe. And I think this, this, you know, I think this chart's important because the, the thick blue line, the, the kind of highest line here, you know, these are the companies, these are these high intangible value companies within the global universe that I think are, you know, potentially the best positioned moving forward for a kind of environment of trade uncertainty and tariffs. Because this, which is as I already argued and shown, global stocks have outperformed domestic stocks. They're higher quality businesses, you generally want to be in them. The problem is that now you're sitting here and you're afraid of tariffs which could legitimately be harmful for companies that produce physical goods. The thing is that intangible companies, you can't really tariff an intangible asset. Purely digital businesses, services businesses as I mentioned earlier, are not really able to be tariffed. Again there are different ways of imposing non tariff barriers. Like you know, if you're China, you just steal someone's ip, right? That's a kind of a cost imposed, a tax imposed on companies doing business in China. But you know, that's obviously less egregious than you know, 145% tariff. So and I think one other other interesting thing I found was, you know, there was some research done by some economists who said, who found that when geopolitical uncertainty is high, it's the case that companies that have a lot of physical capex tend to cut back on their investment. So if you're a company that builds factory, it has lots of factories, you're not going to build a new factory if you don't know what the geopolitical situation is going to be. But they find that companies that are more intangible intensive, companies that rely more on intangible assets, they tend to continue to invest through the geopolitical episode because their assets are mobile. So the idea would be that in an environment where trade barriers are, barriers to physical goods are coming up, but obviously the barriers to intangible assets continue to be. It's hard to impose anything that the advantage for intangible as opposed to tangible companies may actually increase more than they already have over the past 15 years. And so in addition to the other things I've mentioned, if you're looking for a way to own global companies but then also have, you know, more resilience in the face of tariffs, the intangible subsets of these global companies could be an interesting place to look.

Host (48:08)

All right, this next chart, which, you know, at the time you wrote it was, you know, non US Stocks were outperforming US stocks. Now that's still true year to date, but US Stocks have come back a lot as the administration has walked back some of the tariffs. But I think it just presents an interesting sort of idea that, you know, U.S. has outperformed international for so long but that you, this year, you know, that trend has sort of flipped. So I guess and you know, you kind of ended the article with talking about international stocks and the valuation, I guess, the discount, if you will, that, that. And this is the next chart, exhibit 40 that non US firms have had over US firms. So can you just talk to this idea or trend?

Kai (48:59)

Yeah, so obviously this is the setup, okay. US Stocks that outperform global stocks, so that's either developed or emerging market non US stocks by significant margin for the better part of 15 years. And as a result, most investors, especially US investors are heavily overweight. US equities versus international. And investors have at many points in time over the past 15 years said, all right, now is the time things to turn around. But there's never been a catalyst. This year has been interesting, right, because you have, and I'm going to get this number wrong, but the US market's basically flat and non US is up like 12% or 14% or something. So you finally see some resilience and through the period over which US Stocks took a bath after Liberation Day, international stocks held up really well. So you're actually starting to get some true diversification from being outside the U.S. and now, you know, for any allocator who's saying, all right, well like that maybe now's the time, maybe I should shift some money from US to international but. And what are the two buts? So the first but is that, you know, US companies are just more dynamic than global companies. And I think that's true on average, of course. Right. But the average isn't always the most important thing to look at because, you know, and I talk about it in a different paper, but I have found that it is true that, you know, non US companies have had lower earnings per share growth than US companies. But I find that that can be largely explained by the fact that they have under invested in intangible assets. So they have not invested the same way that US companies have in say R and D, employee training, marketing and sales. If you were to adjust that away, it explains a large percentage of the gap. And as I just showed, global companies tend to be more intangible 2x as intangible as domestic companies in the international space. The second objection you often see is, well, I expect growth in say Japan and Germany and Europe to be pretty low weak GDP growth and that could be a problem. But the thing is that for these non US global companies, they sell mainly to a global audience, right? So think, you know, companies like, you know, Novo Nordisk or Airbus, right? They compete with not other European companies, but they compete with Eli Lilly and Boeing and in some cases are actually the superior, you know, choice within each industry. So, you know, the case I'm making in this final section here is not specifically for non U.S. companies in general, but non U.S. companies that are also global companies. So non U.S. multinationals, you know, you mentioned like ASML, Samsung, Sony, Toyota, right? These are companies that are as good, if not better in some cases than their US counterparts. Yet they traded a huge discount simply because of the accident of where they're domiciled and where they're listed, what stock exchange they're listed on. This chart here, exhibit 40 shows that across a variety of metrics, price to earnings, price to sales, price to book, EBITDA, EBITDA, you find that the non US global firms trade at a 46% discount on average relative to global firms, US global firms. And so I think to the extent where these non US global firms have first of all demonstrated resilience through the current crisis resilience relative to say US companies that have more idiosyncratic political risk, yet they trade at a 50% discount roughly while still maintaining competitiveness on intangible intensity or other, any other metric you choose to look at. It's definitely an interesting place to be, you know, especially kind of given that, you know, we're probably not out of the woods yet and there may yet be some additional, you know, rounds of, of tariff related anxiety moving forward.

Host (52:44)

Kai, thank you for letting us use all your hard work and images for this, this discussion. It's, it's been great and I think that investors are going to learn a lot from it. What would you say if you there to be a conclusion to this? What are the biggest takeaways in your mind that investors should be thinking about given all this empirical data that you sort of outlined for us today?

Kai (53:04)

Yeah, I mean it's, it's funny. I think this is a common theme across a lot of the, a lot of my work and the work of any kind of empirical, you know, researcher within, you know, finance or economics, which is that like once you take a historical, historical view on things that have happened, you kind of get a bit more space and you can kind of step back from, you know, the news headlines and a lot of the, you know, kind of crazy ups and downs, right? You think about the past three weeks like there have been like days where the market's gone up like you know, 8% and then down the next day and then up then down like these kind of crazy gyrations. And you know, at least as of today we're kind of in the same place, at least in the US we started. And so, you know, taking kind of longer term view I think helps get the context and confidence for an investor to say, you know, maybe a lot of this is noise and you know, of course there are meaningful things that are happening. But taking a longer term view, staying the course, continuing to focus on what's got us, what's led to success so far, which are these kind of high intangible global oriented companies that have taken advantage of a lot of the benefits of free trade and such. As I pointed out earlier, I think free trade is here to stay whether we like it or not. And the reason really is that just the cost of not having it is too high. Right. Like 100 plus years ago when we had smooth Harley tariffs, like yes, you know, there was a wave of de globalization but the world was way less connected back then. Today we have the Internet, we have containerization we have modern logistics, we have more people, right? So the network effects the network value. The potential upside of globalization is just so much higher than it was even a hundred years ago. So for the world to fully turn its back on that would be even higher opportunity cost than what it was before. So I think just the political appetite for such a move, to me would seem unlikely. I think what's more likely is what they call reglobalization, Just the kind of rewiring of global trade through different regions and kind of geopolitical blocks, which doesn't destroy trade, it just kind of reorients it. And that definitely creates a lot of geopolitical risk and friction, which is hence the reason for this paper and the reason why I have, you know, these four suggestions for how investors can, you know, while still maintaining exposure to global trade kind of tilt, make. Make tilts to hopefully be more resilient in the face of some of these stressors while still, you know, focusing on the, The. The big picture, which is look like we've all benefited, we being shareholders, sorry, not. Not everyone, to be very clear, have benefited from, you know, exposure to global trade. And as an investor in a stock market like that, that's. That's been the golden goose. Let's not. Let's not kill it. Great.

Host (55:42)

Thank you very much, Kai. We really appreciate it, man. Thanks.

Kai (55:45)

Thanks for having me. Thanks so much for tuning in to this episode. If you found this discussion interesting and valuable, Please subscribe on YouTube or your favorite podcast platform or leave a review or a comment. We appreciate it. No information on this podcast should be construed as investment advice. Securities discussed in the podcast may be holdings of the participants or their clients.