iiinova — where your digital self keeps evolving

In aggregate, 3.4B passengers a year translates into ~45 terawatt-hours (TWh) per year. In '24, China will consume around ~10,000 TWh. So this is equivalent to ~0.5% of total power consumption in China a year. This compares with est. ~78 TWh for the entire light NEV fleet.

070

One under-appreciated energy efficiency advantage of HSR vs. EVs is higher effective seat-yield: HSR in China is typically ~80% seat-yield vs. <1/3rd for passenger cars. A fully packed EV can be nearly as efficient as HSR but cars are most often used by a single person/driver.

281

This compares to ~210 watt-hours per km for a typical electric vehicle. An EV with two passengers would be ~105 watt-hours per px-km.

HSR can transport passengers at an avg. electricity cost of ~38 watt-hours per px-km. The typical ~348 km trip takes up ~13.2 kWh, or around half a days worth of typical power consumption.

120

The total build cost for ~45,000 km was ~$1T, or ~$24M per km. Amortized across 124B passenger rides comes out to $8 per ride and 2.3 cents per km. Avg. mfg. wage is ~$8/hour today, so the amortized upfront build cost per ride is about 1 hour's worth of avg. mfg. work today.

121

If the HSR network were frozen today at current run-rate of ~3.4B px/year for the next 30 years, this would aggregate to 124B passengers traveling ~43T kilometers, or ~7.1 light years. This gets you to Luhman 16, the third-closest known star.

120

Since 2009, China's HSR network has served ~22 billion passengers traveling an aggregate ~7.6 trillion kilometers (~348 km/trip). This is equivalent to ~26k round trips to the sun, or ~0.80 light years.

1156

I am not that familiar with it but it seems promising.

There really is much to be optimistic about in the future for our kids and grandkids. We just have to get their in one piece so we can deliver it to them.

020

And I should emphasize that this should not just be about 🇺🇸+ 🇨🇳 but everyone on the planet. Where this ends up is a much harder question to resolve but as long as people don't get so emotional and nasty about it, I am confident we will be able to collectively figure it out.

🇺🇸+ 🇨🇳 will continue to drift apart economically and that's perfectly fine! (again, as long as it is done peacefully) The world will need to find a new equilibrium point — and new (or modified) organizational structures.

Thus 🇺🇸-🇨🇳 trade & tech war is simply the manifestation of the underlying reality that 🇺🇸 + 🇨🇳 are drifting apart in terms of economic compatibility. Levelheaded recognition of this reality (i.e. not freaking out about it) is key to managing a peaceful transition.

120

Why dedicate this increasingly scarce pool on meeting foreign demand when there is still work left to be done at home?

Meanwhile, China is increasingly less thrilled about being valued only for its labor. Its blue-collar labor pool is shrinking and becoming more of a precious commodity and wages rising, commensurately.

Within this framework, 🇺🇸 has responded rather predictably by effectively restricting or banning imports of everything ranging from Tiktok to Huawei smartphones to BYD cars. While various reasons are given (reciprocity, blanket "national security"), economic competition is really at the heart of it.

🇺🇸+ 🇨🇳 are becoming less economically compatible over time. Understandably, 🇺🇸 did not want to give up the economic "high ground" that powered a comfortable life for the vast majority of citizens. Also understandably, 🇨🇳 wanted to bring such a life to its billion-plus citizens.

Ultimately, what has really changed the equilibrium was China quite rapidly moving up the value chain (faster than most expected) and starting to develop higher-value tech/IP/brand-driven products that were competing with U.S. brands and MNCs.

111

This trading relationship dominated over two decades from the mid-90s when China first started emerging as a major manufacturing destination to the mid/late-2010s when the current trade war started to brew.

100

This reciprocal trade relationship was fundamentally based on the respective strengths & weaknesses of both sides. China had a large, productive and inexpensive labor pool. The U.S. had the largest modern consumer market, leading technology and advanced human capital.

100

The U.S.-China trade relationship of the first part of the 21st century was at its core predicated on 🇺🇸 retaining brand/tech/IP value and outsourcing production to 🇨🇳 mfg. partners. In return, 🇨🇳 opened up its domestic market to these high-value 🇺🇸 products.

111

The mistakes came from poor strategic & tactical decisions on whether or not to re-invest windfall profits back into the business, in an attempt to get ahead of new technologies like electrification that could disrupt existing business models.

There are lessons to be learned here. The lesson isn't that GM should have never entered China. It earned tens of billions of incremental dollars for doing almost no work and putting very little capital at risk.

Its Chinese competitors did, including its JV partner SAIC. And once EV technology came of age, the writing was on the wall in China. It was only a matter of time.

100

It decided to mostly distribute to shareholders through dividends, debt paydowns and buybacks. Where it did reinvest back into the business, with the benefit of hindsight, we now know it did not do it particularly well or efficiently.

100

GM China provided a lifeboat for GM when it underwent restructuring during the GFC. For a period of time, GM China generated more net earnings for GM shareholders than the rest of the business combined.

100

These charges are the official recognition of what's been known for a few years by now: After nearly three decades, GM won't be selling that many cars in China anymore. Off its initial investment of <$1B, it made $17B ($22B minus $5B in charges). At least a 17x return. Not too shabby.

100

The writedown is a non-cash charge and the restructuring costs are presumably going to be covered by ongoing sales as the operations wind down. The $5B represents cumulative equity income from around mid-2018 to today.

100

GM to take more than $5 billion in charges on China operations

Last week, GM announced that it was going to be taking $5B in charges on its China joint ventures, split between a writedown of $2.7B on its China JVs + another $2.6-2.9B in retructuring charges related to "plant closures and portfolio optimization". www.reuters.com/business/aut...

reuters.com

The company has struggled to compete with carmakers in China,

100

This was analogous to a landlord collecting rent checks. GM had little operational responsibility as again, its main goal for the JV was to monetizing existing R&D selling to Chinese households.

100

Over the last two decades from 2004 to 2023, off that initial investment, GM generated cumulative equity income of ~$22B. Even last year (2023), GM earned $446 million on ~1M unit sales.

100

These profits were reinvested into expanding manufacturing capacity over the years, from a few hundred thousand in the early years to over 4 million at its peak This meant GM did not have to put that much more incremental capital at risk than the initial registered capital amounts.

100

From GM's perspective, the joint venture was a way to monetize existing R&D investments. The GM-SAIC joint venture was profitable from the start: it is not hard for a car company to be profitable when it does not have to reinvest significantly back into R&D.

100

GM originally setup its JV with SAIC-GM in 1995 with its 50% contribution to registered capital of ~$800M The equity component might even be less depending on how much the buildout of the plant was funded by local debt.

https://www.economist.com/business/1997/06/05/china-adopts-the-chaebol

130

This plus geographic proximity to SE Asia gives it a decisive advantage over others including Japan, Korea and the U.S. in how supply chains and new manufacturing clusters will develop in the region. We should expect its influence to only continue to rise in the decades ahead.

China will play an outsize role in how manufacturing supply chains develop in ASEAN, given its existing industrial base supplying both upstream components and a rising market share in capital equipment, which sit at the heart of industrial FDI.

How this battle for foreign capital plays out with India and other regions is the other related story. We can see already how FDI in ASEAN dwarfs India (green). We are already seeing these themes play out with India competing with Vietnam and Indonesia for a slice of Apple's manufacturing.

Note that the chart above does not mean there is more manufacturing investment in ASEAN vs. China now. It just focuses on FDI. In China, business GCF (especially industrial) is growing as a share of GDP, picking up the slack for property and traditional infrastructure declines.

Foreign MNCs aren’t investing in China anymore. Meanwhile Chinese MNCs are driving FDI growth in ASEAN, tightly integrating SE Asia into existing supply chains centered around China. ASEAN is at the frontlines of a major shift in global mfg in the coming decades. 🧵

150

But this is not supported by operating data - utilization is rising even as the network grows in length. bsky.app/profile/gluk...

I remember that Nikkei article from 2022. I read it again and do not see where it referenced cost overruns. Again we can calculate $/km pretty accurately from the publicly disclosed capex numbers.

000

Costs have not come in higher and may have even been lower than forecast. Projects are typically completed within budget and on time. Total buildout cost can be extrapolated from China Railway’s publicly reported financials. FAI is reported monthly.

100

Look Grady, the goal here is to replace that Gallium, which is a critical component in a variety of commercial and defense sectors, in a sustainable way. Do I care how its done or if it is painful?

100

Billy, what's going on here? Why can't we just stick to relying on tariffs and discovering new greenfield mines? It worked for oil, remember? We figured out the fracking and the shale stuff? This just sounds painful.

100

We need to partner with existing aluminum producers, retrofitting them to extract gallium as a byproduct. We don't care where or how old these plants are. We don't have time to get through regulatory approvals. Brownfield is the only way we can get the unit economics to work.

100

This is what we really need: Chemical, mechanical, materials science, civil ... every type of engineering and technical resource we can find at home or abroad. We don't care if they are young, old, white, black, yellow, or purple ... as long as they can build & not just bark.

100