Has President Tsai Ing-Wen seen any success in her government’s efforts to diversify Taiwan’s economy and lessen its dependence on China? How reliant is Taiwan on Chinese imports, like sand, that is used for semiconductors and construction?

Taiwan is one of the few countries which has a large trade surplus with the PRC on the order of $60 to $80 billion a year. While beneficial for Taiwan, it creates a major security vulnerability where the PRC could cut off purchases of Taiwanese exports. As you noted, President Tsai Ing-Wen and her government have viewed this relationship as much more of a security problem than an opportunity, in contrast to the previous administration. However, this is not something that a leader of Taiwan can really change without harming Taiwan's own economy. Most of the decisions are left with business executives regarding where best to invest and build the next generation of technology. The other critical piece here is that much of the trade surplus is not for stuff being sold for purchase in China, but sent there for final assembly and export to other parts of the world. 

The classic case is your iPhone. Much of that $60 to $80 billion surplus is in high-tech semiconductors, related parts, and software, being exported to the PRC and to a factory there, assembled, and then re-exported to every other place on earth. Tsai Ing-Wen cannot tell Apple “don't build iPhones in Shenzhen.” It’s really the final customer for those manufactured products who can influence how the supply chain is created and where things are made. President Tsai has emphasized trying to create a hospitable environment in Taiwan for investments with some partial success, but most of that has been driven by increasing costs and the pain of dealing with Chinese regulations and arbitrary switches in policy in China. Additionally, there is increasing customer concern from companies like Apple, Google, and others buying these products that they're overly reliant on a supply chain that runs through the PRC. So maybe they won’t shift all their production out of the PRC, but they certainly want a backup. That is not something driven by government policy in Taiwan but by downstream customers.

How much progress has there been in the Tsai administration’s plan of incentivizing Taiwanese manufacturers to leave mainland China?

Taiwan's comparative advantage right now is in its high-tech industry. They've already got the facilities, the labor force, and the infrastructure to manufacture high-tech items, particularly semiconductors. That's by far the biggest export by value of Taiwan. The Tsai administration has done well to ensure that semiconductors are still profitable to make in Taiwan and it makes the most sense to build them there. The expansion of production around the world is not in cutting-edge semiconductors. Taiwan right now is at seven nanometers and they're likely to go down to five soon, if they haven't hit that number already. TSMC and some of the other fabs are now investing in building two nanometers in Taiwan. However, around the rest of the world is investment in seven nanometers or above. There's so much demand for semiconductors and there's plenty of production capacity to go around. President Tsai has been successful in keeping the cutting-edge production in Taiwan. That is by far the largest and most profitable part of the manufacturing industry and by far the most strategically important. But it consumes a huge amount of resources. Copious amounts of water and a lot of energy. The administration has made keeping the semiconductor industry well supplied their top industrial priority. Taiwan's energy use, particularly of electricity is rising pretty dramatically right now because of the increased demands from the semiconductor industry. Recently they've been able to meet the increased demand, but it's less about incentives on the margins, such as a lower tax rate or higher subsidies for companies, and more about ensuring that you've got the land, labor, and the power to fab in Taiwan to live.

How is China’s declining economic growth rate impacting how it engages with Taiwan? What would the loss of Taiwanese imports mean for China’s economy? 

The items being banned to punish Taiwan, and signal to the United States that the PRC is unhappy with American policy on Taiwan, are very targeted. They harm specific importers, but they don't harm the larger economy very much. These items are a very small part of cross strait trade, but they're designed to have large political impact. Take, for example, the rise of shimuyu (虱目魚), or milkfish in English. The milkfish industry in Taiwan, which grew up over the last dozen years, is focused almost entirely on exports to China. If the PRC bans those imports, suddenly that industry collapses in Taiwan, and it can have a big impact on Taiwanese producers and farmers. For China, it’s only a drop in the bucket. However, China hasn’t dared go after stuff they actually need for the economy. The example I like to use is that you can still get on a plane in Taipei and be in Beijing in three hours. There are cross strait flights and China hasn’t interfered with that at all, even as they're doing these very high-profile military exercises in the strait.

China is also struggling to shift to a new economic model. Over the last 40 years, their model has largely been based on high savings, a very favorable demographic profile, and high government-led investments in mostly infrastructure and real estate. Now they need to shift to a more consumer driven model, but that requires actually allowing ordinary people to keep more of their income and allowing salaries to rise. The investment that is going into unproductive infrastructure comes from low-cost loans that are made from banks which have deposits from individuals. The banks put a tax on those deposits, in that they pay a very low rate for those deposits. Ultimately consumers are hurt by this policy. Therefore, China needs to move away from that in order to generate larger consumer spending. I don't see any signs that they're politically able to do that right now. Over the medium to long run, I think they're going to have a major slowdown because so much of their investment is going into unproductive activity. So, if China does something more dramatic than just stopping agricultural imports, the problem worsens. An underappreciated part of the economic integration between the two sides is that China is fairly vulnerable as well to anything that would disrupt the vast majority of trade. Hence, a Chinese blockade against Taiwan is not very credible as China will be harmed just as much as the other countries in the region. They're unlikely to do that, unless it's the precursor to a full invasion.

Can Taiwan’s “silicon shield” help protect it from Chinese military aggression? Has China managed to advance its own semiconductor chip industry as an alternative to TSMC?

China is certainly trying. SMIC (the Semiconductor Manufacturing International Corporation), which is the big Chinese state-run manufacturing company, has made some progress. The PRC has thrown an incredible amount of money at this problem to try to build up a domestic semiconductor industry that could compete with TSMC. From a strategic perspective, it is important in case China is cut off from Taiwanese chips. There's a lot of reason to obfuscate on both sides about the progress that they've made, but the most reliable information I've come across is that China is three to five generations behind where TSMC is in its manufacturing capability. What that means is no, China cannot provide the most advanced chips and cannot develop the processes that TSMC has already developed in a short period of time. 

Relatedly, the supply chain that TSMC relies on to build its semiconductors is really complex and involves inputs from at least a dozen different countries. There's a number of important chemicals that come from Japan, there's the extreme ultraviolet lithography machines that are made in the Netherlands. TSMC has access to all that stuff while SMIC may not. Were they to try to build a domestic industry that is completely self-sufficient and doesn't rely on any of those imports, it would include about 5,000 different things. Even if the Chinese talk a good game about how they are building that infrastructure and supply chain, it will be a much more expensive product if they build it domestically using domestic suppliers. I'm skeptical that they can do it in the near run or at all. It will come at an incredible cost to their efficiency and their cost per unit. The only way that makes sense is if they plan to cut themselves off from the rest of the world entirely.

There is a caveat to this. There are many devices, particularly in military hardware, that don't rely on seven nanometer chips. You could have chips that were cutting edge 10 or 15 years ago that are still important, and Taiwan doesn't build a whole lot of those chips. Japan and the US don't build a whole lot of those chips either. Where the danger lies is SMIC coming to dominate the market for chips that are two to three generations behind, but still critically important for a lot of consumer and strategically important goods. That is how diversifying the supply chain away from China to Japan, South Korea, Vietnam, the US and Europe can really make a difference. It's creating the incentives to build those older generations of chips that are still nonetheless strategically important.

Do you think Taiwan will follow China in also producing some of these older generation chips given the global reach of TSMC and its ability to produce them more efficiently and cost effectively?

TSMC, as a profit-oriented company, wants to build the chips that make it the most money. Older generations of chips do not generate a whole lot of money. On top of that, if China is going to flood the market with cheap chips, it may well lose money at that point. TSMC’s incentive is not to produce the older generations of chips, which is actually where the potential security issue lies. It's not that the US would be cut off from the most advanced chips, it's that we would be cut off from older chips that are all made in China. The Chinese can restrict the supply in a way that we cannot easily adapt to.

Do you think the “Silicon Shield” idea has validity in being a protection for Taiwan against Chinese military aggression?

Yes, but to a limited extent. It shouldn’t give people confidence that China will not direct military action against Taiwan. If the PRC is willing to use military action against Taiwan, the Chinese must be willing to pay the really large economic costs associated with the destruction of the Taiwan semiconductor industry. However, the “Silicon Shield” idea does have a certain stabilizing effect in a roundabout way. The PRC is heavily reliant on a supply of chips, so the costs are greater if they interfere with cross strait trade and destroy the industry. Second, it makes the threat that the US would intervene to stop an invasion, or otherwise to protect Taiwan, more credible. It is a core interest of the United States to maintain trade with our partners and allies in the region, and Taiwan is a critical supplier of stuff that we need. In that sense, the US pledge to maintain peace and stability across the strait is more credible to the Chinese than it would otherwise be if we weren't so heavily reliant on exports from Taiwan.

As Hong Kong become more directly under the control of the CPC, has Taiwan received an influx of new capital, as investors in Asia seek an alternative to Hong Kong?

One would hope so, but I haven't seen good evidence of that. Taiwan has an antiquated banking system and many of banking-related activities are still done on paper. It's very hard to open a new bank account, at least compared to the ease in Hong Kong or elsewhere. The kind of electronic banking services that Hong Kong and Singapore have are not widespread in Taiwan. That's actually hurting Taiwan’s ability to attract foreign capital. However, because Taiwan is so isolated from international economic institutions, it has had to be very conservative about how it regulates its banking system. Its central bank holds huge foreign currency reserves and monitors capital flows very carefully and imposes restrictions on flows. Thus, I'm pessimistic that Taiwan, in the near run, can take advantage of the decline in Hong Kong's image as a good place to do business as a way to be involved in “Greater China.”

On a related note, why do many multinational corporations that do have offices either in mainland China or in Hong Kong (or both), not have a base in Taiwan?

There are a couple of reasons for that. The one that the government likes to focus on is that Taiwan is not international enough and does not have enough English speakers. The Taiwanese are good at playing the middleman between the PRC and everybody else, but they're not really good at bringing people in to use Taiwan as a base to go everywhere else. This has been a consistent pattern for at least 50 years. The cultural and institutional legacy in Taiwan is really hard to overcome. The other reason relates to Taiwan's diplomatic isolation. It's not involved in most of the major trade arrangements or negotiations in the region. It has applied to be part of the CPTPP, but the PRC can probably exercise a veto over that application. It has made some progress on bilateral negotiations and has bilateral trade agreements with Singapore and New Zealand, but it would take something where Taiwan is included in a regional agreement that has some really strong mechanisms to force political, or at least regulatory change, domestically. The only way that happens, however, is if the US is involved in and throws its weight behind Taiwan's application. Right now, the United States is not really in the trade game. We're much less ambitious than we were during the Obama years with what was then TPP.

You recently wrote a blog post on your website about the upcoming Taiwanese local elections. Given the likely continued power of the DPP over the KMT in Taiwanese politics, how do you see this impacting the economic relationship between Taiwan and China?

Any effect will be very indirect. The local elections are really going to turn on the personal characteristics and quality of the candidates. The KMT has an advantage in the local elections, because it has a lot of incumbent mayors and county executives, most of whom are relatively popular. The indirect effect it might have is on the 2024 presidential election, and there are a couple of ways that the results of what I'll call these midterm elections will affect 2024. One is Tsai Ing-Wen, right now, is in her second term. She's the chairwoman of the DPP. Her ability to continue to influence DPP policy and potentially her successor in the party depends on the results of these elections. If the DPP candidates fare badly, she may have to resign as chair. It is not clear how the DPP would adjust to a power vacuum in the party and whether there will be the factional battles that you saw before Tsai Ing-Wen became chair in 2008. 

The flip side of this is a victory for the KMT. The KMT, however, is quite a bit behind in the generic party polls relative to the DPP. But if they do well in the local elections, I'd expect a bump up in popularity. Depending on which candidates do the best, you could end up with a popular KMT mayor running for president in 2024. If that happens, you may see a significant shift in cross strait relations again. Beijing is generally willing to cooperate with the KMT, but it hasn’t been willing to cooperate with the DPP regardless of what the DPP says.