Major media outlets, financial twitter (fintwit) and social media are abuzz with comparisons between Russia invading Ukraine and the possibility of China invading Taiwan. These comparisons are (thankfully) off base for many reasons which I will explain in this post. Specifically, I will touch on:
1. The differences between the economies (specifically, exports) of Russia/Ukraine and China/Taiwan
2. The level of interconnectedness of each with the Western/Global economic machines
3. The incentives of the various players involved who are capable of influencing how a conflict plays out
Image from Japan Times
It’s a MAD, MAD world…
On July 16th, 1945, the first atomic bomb was tested in the New Mexico desert. In the minds of many Americans, The Bomb was an answer to prayer. There would no longer be any need for a seaborne invasion of the Japanese mainland. The Bomb would save thousands of American and Allied-Forces’ lives. Ten days later the Potsdam Declaration was delivered to Prime Minister Kantaro Suzuki, requesting the unconditional surrender of all Japanese armed forces. Suzuki responded dismissively, and in so doing signed the death warrants of nearly 300,000 of his countrymen. On August 6th, a U.S. B-29 bomber nick-named Enola Gay dropped Little Boy on Hiroshima, killing an estimated 80,000 people instantly. Including deaths from radiation and other aftermath, the death toll escalated to nearly 200,000. Fat Boy was dropped on Nagasaki three days later.
The Bomb hasn’t been used since. From Shane Parrish’s blog post on mutually assured destruction:
After the Us dropped the first atomic bombs on Japan, other countries raced to develop their own. The USSR had hydrogen bombs within 8 years. Both developed their technology to the point where either of them had the ability to basically decimate the entire world if the leaders chose to. It goes without saying that a nation had never held that type of power before.
By the 1960s, the concept of mutually assured destruction (hereafter referred to as MAD) was crystallized. Both the US and the USSR could bring about the end of humanity (including themselves), but neither wanted to. This lead to a stalemate, essentially stating “I won’t if you don’t”. For either to attack would mean their own destruction, defeating the purpose of war. Ironically enough, the concept of MAD has led to relative peace between countries with nuclear capabilities…
I want to stress that I am not making the case that China will never invade Taiwan (or force Taiwan to come under their governance in some other way). What I am going to explain is that if there was a conflict between the West and China that was handled in the same way as the Russia-Ukraine conflict, it would result in economic mutually assured destruction. Therefore, an escalation like we are seeing in the Russia-Ukraine conflict is extremely unlikely – even if China were to invade Taiwan militarily.
To understand how bad it would be if the US and her allies responded to China invading Taiwan in the same way as they have to Russia, let’s take a walk down hypothetical lane.
It’s May 22nd, 2025. China invaded Taiwan 11 months ago. In response, the United States and her allies placed draconian sanctions on China, similar to those which have been placed on Russia. China now controls Taiwan’s manufacturing capabilities, which include all of their semiconductor factories (henceforth fabs). However, the fabs are not making any chips because equipment suppliers from the US, Europe and Japan have been prohibited from shipping them equipment or honoring service agreements. With that backdrop, we’re going to look at “a day in the life of Jane”.
Even though the stock market is down 85% Jane Doe, a government employee, still has her job. Jane’s daughter Jan, 18 years old, is about to head to college. Jane and Jan are shopping for three things: a car, a phone, and a new laptop.
Jan loves Apple products so the first thing they try to buy is a new iPhone. When they get to Apple’s website there is a popup that says the following: “Due to the ongoing conflict between China and Taiwan, and the sanctions placed by the United States and her allies, Apple is not currently able to make phones or computers. Apple is working hard to identify new suppliers capable of making the high quality components we require to build the world’s best products, and is committed to resolving this problem as quickly as possible. Please check back with us on a regular basis for updates.”
Jan and Jane are disappointed but haven’t lost hope. Besides, it’s unpatriotic to complain about a justified economic war. Jan is willing to use a PC and an Android phone if that’s what she has to do to support her country.
Jan goes to Bestbuy.com and searches for “Android Phones”. To her surprise, a similar message appears telling her that Bestbuy doesn’t have any new phones available but hopes they will get stock as soon as the conflict between Taiwan and China is over. Bestbuy suggests purchasing a refurbished phone, which they do have in stock. Jan looks at the price of an iPhone 11, a phone five generations old. It costs $2,500. Jan looks at Jane apprehensively, and Jane says “you need a new phone sweetie, go ahead and buy it.”
Finding a computer is a similar experience. Dell, HP, and every other computer manufacturer have their computers assembled in China. China is under an export block, so there are no new computers available for purchase. As with the phone, Jan and Jane end up spending 3X the original list price for a 4 year old laptop.
Buying a car is an even more depressing experience. Jan wanted an electric vehicle (EV), but the United States, Europe and Japan are only producing 25% of the EVs they produced in 2022 because EVs require more chips and other components that come from China than internal combustion engine (ICE) cars. The used vehicle price spike of 2021-2022 pales in comparison to the price spike that has taken place in the past six months, with fully half of global auto capacity having been taken offline. Jan ends up buying a 2017 Honda Accord with 100,000 miles on it for $35,000.
I’m not being hyperbolic. It’s actually worse than this. Netflix users would start seeing the service go down periodically because Amazon Web Services (henceforth AWS) would be unable to offer permanent uptime to its customers without a new supply of chips for its datacenters. Shopify - which runs on Google cloud - would start experience rolling downtimes. And, even when it was up, its chat service (Twilio) and ability to receive payments (Stripe) would only work intermittently (both run on AWS).
The National Security Agency and CIA would have to start limiting the amount of data they collect on our adversaries (and the amount of digital spying they engage in), because that data (and their spying programs) rely on chips made by Intel, AMD and Nvidia - all of whom as of 2023 had nearly 100% of their cutting edge chips manufactured in Taiwan. Worse, the cybersecurity companies relied on by US companies and the government to defend against cyber attacks would also become unreliable, leaving the US vulnerable to attacks by Russia, North Korea and Iran.
Every single software-as-a-service business that runs on the cloud would start experiencing downtime. Companies running on-premise servers would be capped at whatever usage they had already built out. No growth would be possible…
And so far we’ve only talked about the collapse of technology. China being shut out of the Western economy would result in a total collapse of virtually all supply chains. Nike’s clothes, the plastic bottles all consumer packaged goods go in - there is no end to China’s integration in global supply chains. I’ll stop providing anecdotes for the sake of brevity, but their supply is endless.
Data on the differences between Russia/Ukraine and China/Taiwan
Russian GDP was 1.71 trillion USD in 2021. Ukraine’s GDP was 164.5 billion.
For perspective, the GDP of New York State was $1.9 trillion (bigger than Russia). Kansas’ GDP was 175.5 billion (bigger than Ukraine). The budget of the NYPD is larger than the budget of the Ukrainian military.
Taiwan’s GDP was estimated at 689 billion. China’s GDP was 17.7 trillion. The United States’ GDP in 2021 was 23 trillion.
Beyond the vastly different scale of the economies is the type of products they export – which speaks to the interconnectedness they have with global supply chains.
Ukraine export composition in 2020:
1. Cereals: US$9.4 billion (19.1% of total exports)
2. Iron, steel: $7.7 billion (15.6%)
3. Animal/vegetable fats, oils, waxes: $5.8 billion (11.7%)
4. Ores, slag, ash: $4.4 billion (9%)
5. Electrical machinery, equipment: $2.5 billion (5.2%)
6. Machinery including computers: $1.9 billion (3.9%)
7. Oil seeds: $1.8 billion (3.7%)
8. Food industry waste, animal fodder: $1.6 billion (3.2%)
9. Wood: $1.4 billion (2.9%)
10. Articles of iron or steel: $877.8 million (1.8%)
Total exports in top 10: 37.4 billion. $49 billion in total exports.
Here’s a breakdown of the top ten individual products:
Note that every single product is a commodity which can be sourced elsewhere. Yes, shutting down Ukraine’s exports can cause price spikes and inflation, supply chain quagmires, etc – but it is possible to get all these materials from other sources. It is also possible for other countries to ramp production of any of the above products in a relatively short period of time (a new planting season would start in 6 months or less, for example). They are commodities.
One final point before moving to Taiwan’s exports. Let’s assume that Russia takes over Ukraine completely and causes the US and her Allies to sanction Ukrainian exports as well as Russian exports. Does that halt the production of the above materials? Unlikely. Ukraine and Russia would still be able to get seeds and plant and harvest corn. They will still be able to mine iron ore and make steel. If the West doesn’t buy these commodities someone else will (granted, possibly at a large discount). In the end, these products will still find their way into some part of the global economic machine. This is important because even in a world blanked by sanctions, Russia selling oil to China still means China is buying less oil from Brazil and other countries who can shift their own supply elsewhere.
Now on to Taiwan.
Taiwan export composition in 2020
1. Electrical machinery, equipment: US$174.3 billion (50.3% of total exports)
2. Machinery including computers: $44.7 billion (12.9%)
3. Plastics, plastic articles: $18.6 billion (5.4%)
4. Optical, technical, medical apparatus: $16.8 billion (4.8%)
5. Vehicles: $10.1 billion (2.9%)
6. Iron, steel: $7.5 billion (2.2%)
7. Organic chemicals: $7.2 billion (2.1%)
8. Articles of iron or steel: $7.1 billion (2.1%)
9. Mineral fuels including oil: $6.3 billion (1.8%)
10. Copper: $4.1 billion (1.2%)
Total exports in top 10: $296.7 billion. $346.6 billion in total exports.
It should be obvious that producing the products listed above requires a completely different magnitude of specialization, expertise, and supply chain complexity/interconnectedness. For the sake of brevity and because it is sufficient to make my point, I am only going to look at the largest company exporting integrated circuits: Taiwan Semiconductor Manufacturing (henceforth TSM).
TSM would no longer be able to make chips. TSM is reliant on equipment produced by the West. Four of TSM’s primary suppliers (ASML, Lam Research, Applied Materials and KLAC) are based in the West (and more). Making cutting edge chips is the single most complex manufacturing feat humans have accomplished, and it requires constant back and forth between TSM and its suppliers.
ASML’s last generation EUV machines cost upwards of $150 million per unit. From Intel’s website: Delivering just one of these tools takes three Boeing 747 cargo planes, 40 freight containers and 20 trucks. The school bus-sized machine comprises 100,000 parts and weighs nearly 200 tons. Sounds a bit more complicated that corn and wheat, eh?
Crucially, making sure these machines run optimally is a team effort that requires ongoing collaboration, which TSM pays for through service contracts.
A next generation EUV machine will be even more complex and is expected to cost upwards of $300 million. Just for a single machine.
If TSM shut down tomorrow, could Intel simply takes its place? No, for a variety of reasons. First, Intel cannot make cutting edge chips anymore – it’s simply not good enough. It announced recently that it will be moving its own cutting edge chips to TSM to manufacture. Even if Intel had the technical capability, it would take decades for it to build out the infrastructure TSM already has in place. A modern semiconductor manufacturing facility costs $20 billion and takes 3-4 years to build. Once it’s finished – best case – it’s making 100,000 wafers per month (1.2m per year). TSM’s total capacity as of 2020 was north of 13 million wafers.
Intel’s own stated objective - which it is unlikely to accomplish – is to catch up to TSM’s manufacturing expertise in the 2024-2025 range. Given that they are far behind and now have far less resources to invest in next generation manufacturing processes, it’s likely that they will never catch up.
One final point. Even if Intel had the expertise and capital to replace TSM’s manufacturing facilities, and even if they could build out capacity in less than ten years, prices of chips would skyrocket. Engineers in Taiwan are paid less than in Europe and the United States. They work longer hours. Taiwan is not only more efficient at building manufacturing capacity, it is far more efficient at running manufacturing capacity. Re-printing this snippet from a previous post:
…the entire country of Taiwan is one gigantic semiconductor subsidy. Semiconductors require an enormous amount of water to manufacture. If Taiwan runs out of water due to a weird weather pattern, guess what happens… Chips are prioritized above all else. During a recent draught domestic use of water was rationed and water was diverted from farmers to make sure they could keep producing chips! Obviously, this would never happen in the US. Another example is Taiwan’s high-speed rail network. Each district with a fab can be accessed quickly and inexpensively from any other district, making it easy and cost effective for engineers to be where they are most needed.
The reality is that TSM has a monopoly on the production of cutting-edge chips. This is evident when you look at their operating margins.
By 2023 when Intel starts producing their 3nm chips at TSM, the following companies will ALL have their most cutting-edge chips being produced by TSM:
1. Apple
2. Intel
3. AMD
4. Nvidia
5. Qualcomm
6. Broadcom
Where do these chips end up?
HPC means high performance computing – i.e. Cloud.
The following products would either be taken offline – or would cease being able to grow if TSM were no longer exporting chips:
Apple’s iPhone and Mac computers.
Amazon, Microsoft and Google’s core businesses that run on their own data centers which rely on chips from TSM.
Amazon, Microsoft and Google’s cloud businesses. Here is a very short list of companies that run on AWS:
Adobe, Airbnb, Autodesk, BMW, Bristol-Myers Squibb, Disney, ESPN, McDonald’s, Slack, Salesforce, Twilio, Twitch, Stripe, Zoom…
One of the key features that makes cloud services like AWS work is the flexibility in their capacity. In a world where SaaS businesses are growing 30% a year, and more companies than ever are moving to the cloud, an inability to expand means Armageddon. No more 100% uptime.
Without Qualcomm’s chips new mobile phones would stop getting made (Android included). And, everyone’s already heard about the problems automakers are having due to constrained supply. If TSM/Taiwan was taken offline, auto sales would plummet.
To summarize, the cloud would stop growing (or worse), Apple wouldn’t be able to make any products, everyone would be stuck with their existing mobile phone indefinitely, and car sales would evaporate.
There are more examples but the above seem sufficient to prove what an apocalyptic scenario a shutdown of Taiwan would be – just from losing access to chips made by a single company. What about China?
Top exports of China are broadcasting equipment ($208B), computers ($141B), integrated circuits ($108B), office machine parts ($82.7B), and telephones ($54.8B). These are exported mostly to the United States ($429B), Hong Kong ($268B), Japan ($152B), South Korea ($108B), and Germany ($96.9B).
73% of Russia’s exports are energy and metals (raw materials, commodities). China is not Russia.
It’s the incentives, stupid
The good news is that the world runs on incentives.
A case can easily be made that Russia was incentivized to invade Ukraine. Russia has been stating at least since 2008 on a near-annual basis that the continued expansion of NATO to their doorstep would cause a war. Russia was unequivocal in calling NATO membership for Ukraine a red-line.
Image from Transnational.live
Zelensky is a pro-West president who was actively seeking NATO membership and weapons. Russia took Crimea in 2014, and Russia was/is supporting the regions of Donetsk and Luhansk – which want to separate from Ukraine. Remember, there has been a civil war in Ukraine since 2014 in which nearly 15,000 people have died, many of whom are ethnic Russians. Russia had amassed tens of thousands of troops at its border with those regions.
Beyond the military and political dynamics, it is unlikely that Russia had any idea how fierce and coordinated the global response to their invasion would be. With their massive central bank reserves (most of which are now useless) they probably assumed it would be easy to wait out whatever sanctions were placed on them. They might have even thought they would end up better off, guessing that higher oil and gas prices would more than make up for the economic impact of sanctions.
All this being said, when considering the difference between Russia/Ukraine and China/Taiwan – I think the most important incentives to consider are not those of Russia/Ukraine and China/Taiwan – rather, it is the incentives of the West and her allies.
The United States in particular has much to gain and little to lose from the Russia/Ukraine conflict (unlike Europe). It will be able to sell more arms/weapons. The military industrial complex is real and influential. It is largely responsible for the US having spent more than a decade begging Europe to increase its defense spending. Germany has now committed to more than doubling their defense budget from 47 billion Euros in 2021 to 100 Billion euros in 2022, and has further committed to a minimum of 2% of GDP being spent on defense thereafter (~70 Billion+ - representing an ongoing increase of nearly 50%). Politically speaking Russia is unpopular and the establishment wings of both parties are always in agreement around trying to weaken Russia’ economy. I’m aware this may seem less true now because of Trump, but Trump is not representative of the Republican Establishment (and to be fair, he did eventually increase Sanctions on Russia and approve the sale of lethal arms to Ukraine). Romney is more representative of the Republican establishment, and he named Russia our primary geopolitical foe during his race against Obama. The United States was heavily against the Nord Stream 2 pipeline, and will now likely start exporting more fossil fuels to Europe. Lastly, there is also the somewhat conspiratorial view that the US establishment would like to see Russia caught in a quagmire similar to our own situation in Afghanistan. A conflict in Ukraine will weaken Russia by draining both time/focus and resources. The only thing the US stands to lose from a conflict between Russia and Ukraine is additional pressure on inflation – which it (right or wrong) sees as being transitory.
China is also a big winner from the Ukraine/Russia conflict. They know they will not end up on the receiving end of draconian sanctions. They have been focused for decades on securing supply to natural resources. They are already receiving large discounts on Russian energy. The more Russia is cut off from the west, the more reliable and inexpensive becomes the supply of energy and raw materials from Russia.
To summarize… Russia, the United States, and China are all currently acting in a way that can be understood by looking at incentives.
Any incentives the US may have to act against China during a conflict with Taiwan (and vice versa) are dwarfed by the incentive to avoid economic calamity.
I’ll close by stressing again that I am not stating that China will never invade Taiwan. The point is that if China were to invade Taiwan, the US and her allies would be incentivized to respond differently – in a way that does not bring about the economic equivalent of a nuclear winter.
Thankfully for the world, the US/her allies and China/Taiwan have already reached a level of economic integration that creates a mutually assured destruction scenario - history tells us that is good for stability.