To me it's almost more surprising that with that high degree of automation they are not splitting the factory up into smaller ones and building them per-continent.
Shipping stuff is pretty cheap, but distributing the supply chain and creating multiple factories making the same product is incredibly expensive.
Let's say you have a hard drive supplier. Now rather than the supplier shipping to 1 factory, there are 5. Freight rates are going to be higher because you have a lower volume. Increased demand on one continent means you need the right balance of material going to that factory rather than simply upping the order for your single factory. Incorrect sales forecasts now mean you have inventory in Europe but need it in the US. You either need to pay to move it, try to get your suppliers to reroute things (always for a fee), or build finished units and ship them to the US, potentially facing higher tariffs than shipping in components [1]. Inventory costs are higher while you figure this all out, and you have increased logistics costs.
You're paying a (bigger) team of people to manage this, and expediting shipments as needed (likely at 3-10x normal cost depending on how expedited, market demand for shipping, etc).
Then there's product consistency. Even if you order identical equipment and keep all the same suppliers, you'll have different issues at each facility. Sharing best practices around the world is challengings and frequently done poorly. [2]
Then you have tariffs and government regulations. If you were 0.25% of GDP for a country, you can go to the government and lobby harder than if you're 0.05% of GDP in 5 different countries. [3]
-- [1] https://hts.usitc.gov/current offers some light trading if you want to find tariffs for the US. They're incredibly detailed, and the smallest difference could change from a 2% tariff to 60%.
https://en.m.wikipedia.org/wiki/Chicken_tax is also interesting.
[2] I know of a distributed manufacturing setup where two "identical" factories have significantly different quality outputs. The best rationale they've come up with is a difference in altitude and climate.
Corporations who have distributed manufacturing like this often have a corporate team who travel between locations checking how well locations are adhering to the company defined best practices.
[3] Definitely don't discount this step! I've worked for 4 manufacturers professionally, and I know for a fact 3 of them have pushed for tariffs and exemptions from regulations. #4 likely did too, but I don't have a specific example in mind. In each case, the difference in profits were 7 or 8 digits long.
For the product consistency, I assume you are talking about non-fully-automated factories? With full automation "adherence to best practices" hopefully shouldn't be as hard to ensure, I hope.
For tariffs, if we assume that a big part of the specific industries playes are distributed internationally, the size of the industry in general should be the much more significant factor, as you are not lobbying individually anyway, right? If you can even go as far as fluidly shift production between countries you should be to negotiate better lobbying deals, which I understand is how the automotive industry operates to some degree?
Automation certainly helps, but there are quirks and hiccups you could run into. Let's say you built a factory every 5 years. Along the way some machinery vendor goes out of business. You might or might not have the machine design, the rights to make another version, etc.
As you build out, you figure out things that work well or don't work well, and you make adjustments to your process accordingly. Building every factory identically or at the same time is probably a bad choice unless everything is really well fleshed out. The marginal benefits of upgrading older factories to the new standard is rarely economically viable.
>For tariffs, if we assume that a big part of the specific industries playes are distributed internationally, the size of the industry in general should be the much more significant factor, as you are not lobbying individually anyway, right? If you can even go as far as fluidly shift production between countries you should be to negotiate better lobbying deals, which I understand is how the automotive industry operates to some degree?
The actions I've seen (not limited to former employers) include CEO's meeting with high level federal government officials, getting local property tax exemptions for expansions bringing jobs, and collaborative industry group action about unfair dumping of product on the US market.
Remember each manufacturer in an industry group has a different structure. Some things are good for the industry as a whole, and some are more beneficial for one manufacturer than another. CEO's advocate for their own company too, not limiting public actions to trade group actions.
--
Previously I hadn't covered capital expenditure. Building factories is super expensive and tooling factories is also super expensive. I wouldn't be surprised if the facility and tooling would run 100M-500M.
Even in manual assembly, you'd have jigs and tools to buy, go/no-go guages, etc.
I have two Crucial BX500 SSDs here.
One made in Thailand - The 960GB version.
One made in Mexico - The 1TB version.
Minor point on this great post, it would actually be 0.01% instead of 0.05%, because the amount in each country is divided by 5x, but the GDP is also increased 5x, for a 25x percentage decrease.
Not really. You went from negotiating with one country to negotiating with one country of the same size (just 5 times). In an individual negotiation you are still facing a single equivalent country (with no GDP increase).
People may think that simple, generic items may gravitate to competitive markets, but with modern manufacturing equipment, even very small factories can have sky high productivity.
At around 2014, I was working on arranging manufacturing for one netbook design, and I went to South China to pick suppliers for mechanical parts.
A company called RKX hinge was the biggest notebook, and cellphone hinge maker in the world with more than half of global market share. To my surprise, they did not have a website at the time, not even a sign on the door.
They were a small, 2 storey factory in Baoan, with an office on the factory floor, and "a warehouse space" made of dangerously tall pyramids of boxes in the corner of shop floor.