I thought about this a couple of years back and as far as I can make out, we are approaching the situation where you could just about fit a digitised general engineering factory and materials store into a high street shop unit. At that point you could robotically fabricate a car to order and just drive it out the front onto the street when it is finished.

But isn't China's advantage the established (and efficient?) chain of materials supply, logistics along with the availability of cheap labor when needed?

It's about having your supply chain as close to your customer as possible so you can pull (as in lean) on demand. What then happens is the customers' needs iterate faster and faster and you need to keep up. Fit the 3d printer in the shop, print whatever the consumer needs in less then it takes to drink a cup of coffee and you're the winner. Of course you need to make sure you have enough raw materials. That's something China has to import a lot. So unless China builds up domestic demand to much higher levels, western countries will gain more from 3d printing than China.

It isn't just 'print whatever the consumer needs'. It's

- Dig it out of the ground

- process it

- Fabricate the unprintable bits

- Store all of this

- Print whatever the consumer needs in less time than it takes to drink a coffee.

So unless this make believe "shop" is fronted onto a processing/fabrication plant, which is itself, fronted on a magical mine that you can dig up any resource you like. Your scenario can never happen.*

* Or we could invent teleportation.

Yes, but no.

Yes, there's a ton of raw resource extraction, processing, and transport.

But no, because a Nissan Altima or Apple Macbook (or Moto X!) are not commodities in the same way that gold or lumber or silicon are.

There's a reason that the Nissan Altima was the sixth best-selling car last year. I don't purport to know what that reason is, but it's not because it's comprised of raw materials any different from any other car.

Cars have ridiculously much electronics in them and you cannot "print" those just like that in a small shop. Have you seen how big wnd expensive a wafer fab is?

I have seen what people are developing. Besides, cars do not need to have micro electronics.

http://reprap.org/wiki/MetalicaRap

http://medtechinsider.com/archives/24967

http://www.nthdegreetech.com/printed-semiconductors.php

That said, there will always be some things for which economies of scale win out against customisation, and I suspect that many electronics components are in that bracket. Luckily we have things like FPGAs, so you don't need to keep a huge stock of specialised chips to make a wide variety of devices.

I don't think there is a market for cars without micro electronics. Especially not with the rise of electric cars.

FPGA's are several orders of magnitude more expensive than mass-produced electronics.

Modern cars contain more than just controller electronics. What about image sensors, for instance, for automated parking systems? You can't 3D-print an image sensor. Not without a clean room at least.

What's preventing all of those electronics from becoming dumb terminals and abstracting everything away to the central processing unit?

Distributed intelligence is more robust. And the electronics need to be fast for your airbag and ABS.

Presumably you don't have to "print" all parts there.

... So there is just a never ending supply of raw material also sitting in this high street shop?

You keep a stock of base materials and things like chips that you can't fabricate onsite.

Is the same as keeping stock for a normal shop, you keep track of what is being used and order more when it gets low.

They just get a hookup to the Feed, of course.