TiB 193: The bear case on China; the bear case on the US; unknown unknowns; and more...

Welcome new readers! Thoughts in Between is a newsletter (mainly) about how technology is changing politics, culture and society and how we might do it better.

Forwarded this email? Subscribe here. Enjoy this email? Forward it to a friend.

One popular way of thinking about the future of US-China competition goes something like this: today China remains a middle-income country on a per capita basis, but it’s growing and catching up with America fast; so, given its vastly bigger population, it’s only a matter of time before it’s by far the world’s biggest economic power. But what if that’s not true? Noah Smith has a good post that argues that China’s catch up growth may be coming to an end, with important geopolitical consequences.

The core argument is that it’s well documented that “catch up” growth is a phenomenon that tends to stall around “middle income”, as the marginal output from new capital investments falls and the benefits from adopting technology and other best practices from richer countries slow down. China’s path, Smith argues, will be no different. It’s now being hit by a perfect storm of four interrelated economic headwinds: a real estate crisis (see here), an electricity crunch, the CCP’s crackdown on tech (see TiB 176), and the pandemic.

Smith’s view is that this really shouldn’t be much of a surprise. No country has reached high income per capita without free markets and secure private property rights, and Xi Jingping’s decision to take a much more dirigiste approach than his recent predecessors is now generating the inevitable consequences (It’s worth reading this in conjunction with Smith’s recent piece, “What if Xi Jingping just isn’t that competent?”). I suspect this remains a minority view, for now, (though see George Magnus here in agreement), but I’m sure Smith is correct that if he’s proven right it won’t be long before “the ‘China can do no wrong’ narrative of the past few years will be replaced with the equally simplistic ‘China can do no right’”. Watch this space.

That said… even if you take a a very bearish view on China’s medium term economic prospects, the likely path of US-China rivalry remains troubling. As Tanner Greer argues in this excellent new essay, the nature of this competition is not merely economic (or even technological), but ideological - and the Taiwan question gives it an immediate and dangerous military dimension. As we’ve discussed before (see also the TiB podcast episodes with Meia Nouwens and Shashank Joshi), direct military action by China against Taiwan remains by far the world’s gravest geopolitical threat.

What can and should the US do? Greer argues that we’re at risk of being misled by historical analogies. It’s tempting to see the US-China relationship as a sort of Cold War redux - and to think that the strategies the US deployed successfully against the USSR will enjoy the same results against China today. Greer thinks that’s dangerously wrong. Riffing on this superb piece by former senior DOD official Keith Payne (which really is a must-read if you’re interested in this sort of thing), Greer argues that “strategic ambiguity” and deterrence, core pillars of the US’s Taiwan policy, rely on foundations that were present during the Cold War but are missing today.

In particular, Payne notes that strategic ambiguity only works when (a) you can convince your opponent that you might lose control of events and do something that, on the face of it, is irrational (“Mr. Chairman, you will have to take into account the possibility we Americans are just [expletive] fools”, as Dean Rusk told Khrushchev) and (b) you have such military dominance that the uncertainty your ambiguity creates imposes greater costs on your opponents than on yourself. Payne (and Greer) argue that (b) does not apply today in the domain of US-Taiwan policy. Payne concludes:

Scary times.

One of my all-time favourite papers on investing is Richard Zeckhauser’s classic essay on the “Unknown and Unknowable”, which does an excellent job of delineating risk and uncertainty and explaining why exceptional returns often require situations where we not only don’t know what will happen, but we don’t even know the possible states of the world, never mind the probabilities of them occurring. I was reminded of this this week by this excellent post by Alex Wang, co-founder of Scale AI, on “Betting on Unknown Unknowns”.

Wang asks why technological progress sometimes happens unreasonably quickly. The obvious recent example is the COVID vaccines. Wang notes that Fauci and others were seen as implausibly optimistic at the start of the pandemic when they speculated that a vaccine might be available by August 2021. They were wrong, of course, but were actually too pessimistic. Wang cites Moore’s Law as another case study: at the time it seemed a wildly optimistic prediction because there was no clear path to achieving it. It required “unknown unknowns” - breakthroughs that couldn’t be imagined at the time - to happen.

Of course, optimistic tech predictions don’t always come true, so when is it right to bet on “unknown unknowns” coming to the rescue? Wang says “You can only bet on unknown unknowns near the frontiers of human tenacity and creativity”. That seems necessary, but not sufficient. I think unknown unknowns are more likely to occur when you have (a) a big, well defined prize, (b) multiple teams with diverse approaches trying to win it, and © a shared “reason to believe” that it should be possible (We talked about the idea that Moore’s Law is a social, not a scientific, phenomenon back in TiB 30). If that’s right, finding ways to engineer more “reasons to believe” in important domains might be an underrated way to accelerate progress.

Thanks for reading all the way to the end. If you enjoy Thoughts in Between, there’s a really easy way to help: forward this email to a friend or share it on Twitter.

Feel free to hit reply if you have comments, questions or recommendations; I’m always happy to hear from readers.

Until next week,

Matt Clifford