The manufacturing ecosystem in Shenzhen (for example) is unique. Any individual element of it is probably replaceable (at a considerable cost), but the presence of all those elements in the same place is not.
The agglomeration benefits of this are HUGE, and losing them would introduce a thousand small frictions into the manufacturing process. Issues that can be sorted out in hours or days in Shenzhen (finding alternate suppliers, repairing specialized machinery, etc.) would take weeks or months anywhere else. That's for individual issues. A complex piece of equipment might have a dozen such issues while scaling and optimizing its manufacturing processes. In Shenzhen, this might cause a few weeks of delay. Anywhere else, the delay would be measured in years.
There are several other such manufacturing ecosystems in China; Shenzhen is just the one I'm more familiar with. Yes, the individual parts may be replaceable -- but the whole is significantly greater than the sum of its parts. Building replacement ecosystems elsewhere would take decades.
It took China roughly two decades to build this kind of ecosystem. Its primary resource for doing so was the ~4 million STEM graduates in produces every year. That's on par with the rest of the world put together. So don't underestimate the difficult of replicating this.
I’m sure there were synergies in steel and auto production in US cities connected by rail between Pennsylvania and Michigan.
I’d say the key misunderstanding is that you’re conflating China with “markets” in “it took China roughly two decades to build this kind of ecosystem.” If you read it as, “Markets took two decades to build this kind of ecosystem,” it should be clear that the relationship between prices, time, volume, supply and demand aren’t determined by political prerogatives, but by market ones.
In other words, if your thesis is correct (that there’s something special about Shenzhen), the thing that’s special about it can be reproduced elsewhere. It will probably happen faster than two decades, because it’s proven to be a lucrative setup.
Sure, there's nothing mystical about Shenzhen, and I'm sure it can be replicated elsewhere -- but on a decadal timeframe at best.
There are many different factors which go into producing such an ecosystem. Raw number of STEM graduates is a very crude proxy, but it's probably as good as anything else for representing the "feedstock" of such an ecosystem, and gives an rough indication of how difficult the task might be.
China produces ~8x as many STEM graduates as the US. Therefore, all else being equal, if it took China 20 years to develop such an ecosystem, it should take the US roughly 160 years to do the same.
"But wait!", you say, "The US can be a fast second-mover, learning from Shenzhen and not repeating its mistakes. Plus, we may have fewer STEM graduates, but on average they're better-educated and more entrepreneurial, which would allow us to move even faster". (Note, I'm not endorsing these claims, just saying that it's an argument one could plausibly make).
Fine, granting all that, let's say that each American STEM graduate is worth 10 Chinese STEM graduates. In that case, it would only take 16 years to begin to compete with Shenzhen.
That analysis is almost certainly far too generous towards the US, however -- and even if it were well-founded, what American politician would pursue the necessary policies, given a best-case 16-year ROI? And what American polity would accede to the requirements of such policies?
Answer: none. This just isn't going to happen in the US. India has the scale to pull it off, but not the organisational capacity; Germany has the organisational capacity but not the scale; most other places have neither. Anybody who thinks that it'd be easy to replace China is utterly fooling themselves.
This entire statement is predicated on STEM graduates being a proxy, but you provide no evidence that's the case. Building complex market ecosystems has very little to do with supply of STEM graduates. Having some supply is certainly a prerequisite, but there is no indication that the current supply rate is the prerequisite.
I suggest you read the mythical man month, it illustrates the fallacious thinking behind throwing more bodies at projects to get them done faster.
I've read the Mythical Man Month, and it has no bearing on what I'm talking about here. The problem is that "building Shenzhen" isn't a "project", and never can be a "project". The key feature of an industrial ecosystem like this is its enormous diversity -- hundreds of thousands of companies, large and small, doing complementary and competitive and most importantly different things.
That is not something that can ever be matched by a monolithic "project". But it is something that absolutely requires large number of both bodies and brains. STEM graduates is, as I say, a poor proxy, but I can't think of a better one. Feel free to suggest.
(Hint: it isn't GDP or anything like that, otherwise China would never have been able to do this in the first place.)
>The key feature of an industrial ecosystem like this is its enormous diversity
A bunch of engineers all graduating from the same Chinese school system under the same censorship regime != "diversity".
>and most importantly different things.
Some are doing different things, but that has nothing to do with the number of STEM graduates. Your incorrect presumption here is that a STEM graduate is magically innovative and entrepreneurial, which there is no evidence of.
>STEM graduates is, as I say, a poor proxy, but I can't think of a better one.
So don't use it at all. If STEM graduates were enough, why do you think that China hasn't displaced Silicon Valley in the software game?
>Feel free to suggest.
Incentives, market conditions, and government support. In other words, stuff that can only partially be controlled. Silicon Valley is a direct product of research into military technology driven by the Cold War. No number of STEM graduates can replicate that.
That’s the worst possible example, because they truly don’t make cultural products like movies and music almost anywhere but Los Angeles.
If it was about market forces, writers would get paid a bajillion times more. On the contrary, they’re paid terribly, relative to the value their product provides. Same with visual effects artists, just not as bad.
Or music: an ecosystem unmatched anywhere but in LA. Even all those Swedish pop artists wind up there.
Before you google counterexamples, just ask someone who actually works in film or music. You can’t just throw money at “it.” What seems to make good writers and musicians concentrate in LA doesn’t seem to be 100% governed by conventional market forces.
Yes, it’s been difficult to replicate Hollywood! I think experiments with subsidies like for VFX in Vancouver, documentaries in Massachusetts and everything in Savannah, Georgia are interesting and nurture communities at different rates of success.
In contrast, Wisconsin gives a $3 billion handout to Foxconn, Wisconsin gets a Foxconn factory. Because it will save money, things adjacent to the plant will be built.
Writers cost nothing! There’s no money to save by locating them in LA, the savings on airfare and lodging and communications for a literal single human being is de minimus compared to the budget of a feature film.
The scary thing about Shenzhen is all the applause for abstract things like skilled workforces and no appreciation for how different that “IP” is in comparison to e.g. Disney IP, like Mickey Mouse. They are not at all the same knowledge economies, electrical engineering and films.
> The Shenzhen special sauce is the high density of complimentary industries and specialized workforce.
> It's certainly possible to replicate, but history has shown that it can be difficult. It would be like trying to supplant Hollywood.
There's now motivation to take on that difficult job, and the motivation grows the longer the trade war drags on.
Companies may even rediscover the benefits of diversification. Right now Shenzhen is unique, but I'm sure there'd be much less anxiety right now if there was a Shenzhen II in Vietnam, Mexico, or even the US where production could be shifted too in times of crisis.
SZ's special sauce used to just be its proximity to Hong Kong. To think that the role SZ plays has changed so completely in less than 30 years is evidence that it might not be that difficult to replicate.
The agglomeration benefits of this are HUGE, and losing them would introduce a thousand small frictions into the manufacturing process. Issues that can be sorted out in hours or days in Shenzhen (finding alternate suppliers, repairing specialized machinery, etc.) would take weeks or months anywhere else. That's for individual issues. A complex piece of equipment might have a dozen such issues while scaling and optimizing its manufacturing processes. In Shenzhen, this might cause a few weeks of delay. Anywhere else, the delay would be measured in years.
There are several other such manufacturing ecosystems in China; Shenzhen is just the one I'm more familiar with. Yes, the individual parts may be replaceable -- but the whole is significantly greater than the sum of its parts. Building replacement ecosystems elsewhere would take decades.
It took China roughly two decades to build this kind of ecosystem. Its primary resource for doing so was the ~4 million STEM graduates in produces every year. That's on par with the rest of the world put together. So don't underestimate the difficult of replicating this.