5 ms·

One thing I'm curious about is how you define "critical manufacturing"? Manufacturing seems like a directed graph where each node is a product or material and t

One thing I'm curious about is how you define "critical manufacturing"? Manufacturing seems like a directed graph where each node is a product or material and the edges are dependencies or inputs to that product. So if you want to onshore manufacturing of a particular product, you need to also onshore all of its dependencies. My contention is that the graph is so densely connected that you end up having to bring entire supply chains onshore and this effectively means every country needs to manufacture everything.

jl6 5y ago
> One thing I'm curious about is how you define "critical manufacturing"? It’s a good question. For me, critical means you can’t walk away from a deal because you need the product, meaning you end up being extorted. Food and energy meet this criteria. Do chips? Maybe. And if so, then yes, you’re absolutely right that you would need to secure the supply chain/graph too. Maybe not the whole graph if you work back to the point where there is a diversity of competitive sources.
sigstoat 5y ago
> this effectively means every country needs to manufacture everything. depending on your exact goals, it could be worthwhile to bring a little bit of every node on the graph onshore. you retain critical experience, and have at least some hope of ramping up. getting a bunch of functioning, complicated businesses to expand is way easier than creating them from scratch.
- alooPotato 5y ago
  oh that's interesting. You maintain a skeleton crew of an industry as a hedge in case you need to ramp it up....
  - javajosh 5y ago
    This is, in fact, why I think its great people still have an interest in low-tech methods for doing and making. In a Carrington Event[0] we will need people with that knowledge. Those without it will be much worse off. [0] https://en.wikipedia.org/wiki/Carrington_Event https://en.wikipedia.org/wiki/Carrington_Event
JamesBarney 5y ago
I don't think this is true. You really only have to focus on parts of the supply chain that are required to produce, fuel, food, medical, and military equipment that also are not produced anywhere in the world except by a potential a adversary or next to a potential adversary. It's a big undertaking sure, but it's really only onshoring pieces that are produced exclusively in China or next to it.
- alooPotato 5y ago
  "fuel, food, medical, and military equipment" and "exclusively in China" seems like it'd literally bring in the entire dependency graph.
  - kube-system 5y ago
    I can't think of much now that the US relies on for fuel, food, medical, and military equipment that is exclusively made in China. Especially the latter.
rgrieselhuber 5y ago
For countries that want to maintain a high degree of sovereignty and have the resources to do so, it’s probably the way to go. 100% is impossible of course but directionally the smart move.
- alooPotato 5y ago
  I think what I'm getting at is that unless you do 100% of the graph (at least with your allies) you're eventually going to be dependent on an adversary. Like you might not solve anything by bringing chips onshore if your only supply of silicon for those chips is offshore. So you're just pushing the problem down the stack...
  - samus 5y ago
    Silicon is one of the most common minerals on Earth. Rare metals are a different story.
    - alooPotato 5y ago
      ok that was a bad example but the point still stands
jimmaswell 5y ago
If we can at least make sure "the west" has it covered that's a great start, and the US is definitely capable if the will is there. Probably unrealistic for every tiny South American or European country to do literally everything.
SkyMarshal 5y ago
Not necessarily, if you find a way to restrict supply chains to allied nations only that you don’t have risk of war with.
- alooPotato 5y ago
  so cut out china? now bring onshore everything to allied countries.... still seems like a large portion of the graph..
  - SkyMarshal 5y ago
    It's a huge proportion of the graph, it would be like an economic Manhattan project, but 10x bigger. But if things continue on the current trajectory with China's rapid military build-up, base-building around the SCS, threatening Taiwan and the other Asian nations, it may become necessary.
samus 5y ago
Some dependencies are fine to leave off-shore if there is a diverse set of suppliers or if they are located in safe and friendly countries.
kansface 5y ago
Some manufacturing can presumably be spun up with not too much effort if you have other heavy industry around and access to raw materials. Chip fabs require way more than retooling a line. We could quantify the problem as the amount of time it would take to bootstrap a domestic industry if the current suppliers turned off the spigot.
- alooPotato 5y ago
  Ahh this makes a lot of sense. So you rank each node in the graph by how many other nodes depend on it PLUS how hard it'd be to spin up that industry if we need to - then tackle the high value ones
javajosh 5y ago
*> ...how you define "critical manufacturing"? This is a good question, and IMHO it's about threat-modelling. Three events come to mind: pandemic, war with China, and a Carrington Event[1]. The last one is interesting because it would destroy most electrical infrastructure, and it's not clear that the US retains the ability to manufacture those components. I'm not sure, but I don't think huge transformers have a ton of components, so your other (valid) concern about dependencies wouldn't apply in that case. Fabs are at least somewhat similar in that they don't recombine a huge variety of components in different ratios like Shenzen does. [1] https://en.wikipedia.org/wiki/Carrington_Event https://en.wikipedia.org/wiki/Carrington_Event

jl6 5y ago

> One thing I'm curious about is how you define "critical manufacturing"? It’s a good question. For me, critical means you can’t walk away from a deal because you need the product, meaning you end up being extorted. Food and energy meet this criteria. Do chips? Maybe. And if so, then yes, you’re absolutely right that you would need to secure the supply chain/graph too. Maybe not the whole graph if you work back to the point where there is a diversity of competitive sources.

sigstoat 5y ago

> this effectively means every country needs to manufacture everything. depending on your exact goals, it could be worthwhile to bring a little bit of every node on the graph onshore. you retain critical experience, and have at least some hope of ramping up. getting a bunch of functioning, complicated businesses to expand is way easier than creating them from scratch.

alooPotato 5y ago
oh that's interesting. You maintain a skeleton crew of an industry as a hedge in case you need to ramp it up....
- javajosh 5y ago
  This is, in fact, why I think its great people still have an interest in low-tech methods for doing and making. In a Carrington Event[0] we will need people with that knowledge. Those without it will be much worse off. [0] https://en.wikipedia.org/wiki/Carrington_Event https://en.wikipedia.org/wiki/Carrington_Event

JamesBarney 5y ago

I don't think this is true. You really only have to focus on parts of the supply chain that are required to produce, fuel, food, medical, and military equipment that also are not produced anywhere in the world except by a potential a adversary or next to a potential adversary. It's a big undertaking sure, but it's really only onshoring pieces that are produced exclusively in China or next to it.

alooPotato 5y ago
"fuel, food, medical, and military equipment" and "exclusively in China" seems like it'd literally bring in the entire dependency graph.
- kube-system 5y ago
  I can't think of much now that the US relies on for fuel, food, medical, and military equipment that is exclusively made in China. Especially the latter.

rgrieselhuber 5y ago

For countries that want to maintain a high degree of sovereignty and have the resources to do so, it’s probably the way to go. 100% is impossible of course but directionally the smart move.

alooPotato 5y ago
I think what I'm getting at is that unless you do 100% of the graph (at least with your allies) you're eventually going to be dependent on an adversary. Like you might not solve anything by bringing chips onshore if your only supply of silicon for those chips is offshore. So you're just pushing the problem down the stack...
- samus 5y ago
  Silicon is one of the most common minerals on Earth. Rare metals are a different story.
  - alooPotato 5y ago
    ok that was a bad example but the point still stands

jimmaswell 5y ago

If we can at least make sure "the west" has it covered that's a great start, and the US is definitely capable if the will is there. Probably unrealistic for every tiny South American or European country to do literally everything.

SkyMarshal 5y ago

Not necessarily, if you find a way to restrict supply chains to allied nations only that you don’t have risk of war with.

alooPotato 5y ago
so cut out china? now bring onshore everything to allied countries.... still seems like a large portion of the graph..
- SkyMarshal 5y ago
  It's a huge proportion of the graph, it would be like an economic Manhattan project, but 10x bigger. But if things continue on the current trajectory with China's rapid military build-up, base-building around the SCS, threatening Taiwan and the other Asian nations, it may become necessary.

samus 5y ago

Some dependencies are fine to leave off-shore if there is a diverse set of suppliers or if they are located in safe and friendly countries.

kansface 5y ago

Some manufacturing can presumably be spun up with not too much effort if you have other heavy industry around and access to raw materials. Chip fabs require way more than retooling a line. We could quantify the problem as the amount of time it would take to bootstrap a domestic industry if the current suppliers turned off the spigot.

alooPotato 5y ago
Ahh this makes a lot of sense. So you rank each node in the graph by how many other nodes depend on it PLUS how hard it'd be to spin up that industry if we need to - then tackle the high value ones

javajosh 5y ago

*> ...how you define "critical manufacturing"? This is a good question, and IMHO it's about threat-modelling. Three events come to mind: pandemic, war with China, and a Carrington Event[1]. The last one is interesting because it would destroy most electrical infrastructure, and it's not clear that the US retains the ability to manufacture those components. I'm not sure, but I don't think huge transformers have a ton of components, so your other (valid) concern about dependencies wouldn't apply in that case. Fabs are at least somewhat similar in that they don't recombine a huge variety of components in different ratios like Shenzen does. [1] https://en.wikipedia.org/wiki/Carrington_Event https://en.wikipedia.org/wiki/Carrington_Event