[Leftcoast]

Ok .... Quick view of my world through the late 80s, 90s and very early 2000s as a recently minted mid-80s Stanford IE.

My Senior year I joined the company I had interned with - a small but rapidly expanding networking company that would grow from $40M the year prior to my joining and peak just under $10B and then back to ~$400M.   I rode with it most of the way up and part of the way down.  You learn a lot both ways but going up is much, much more fun.

Initially we outsourced to several small subcontractors that would later merge and become big, big EMS conglomerates.  It tended to be expensive and slow (not nimble) so we decided to bring manufacturing inhouse.  We grossly underestimated how hard that would be and were pretty lousy at internal manufacturing the first year.  Second year we doubled down, got much more professional and were able to eventually move everything in house at higher quality levels and lower cost.  Our business doubled twice and our volume was now 10-20x what we started with (prices had decreased so revenue and volume were increasing at separate rates). 

We needed to expand rapidly in the early 90s and much of the growth was coming overseas.  China was now on the roadmap as a brobdingnagian potential market but still seen as a risky investment - maybe we should choose to work with a nearby and stable country ... like Singapore.  We really wanted to be part of the growing European common market as well. Do you build at the current US site, build your own factory overseas near your new customer base or do you go back to subcontracting for your overseas sourcing to reduce risk?   We were confident in our own capability so we decided to build factories right sized for the three major markets - US, Europe and China.  Reduce logistics, reduce taxation (near 0% for our new factories given government grants, plus payroll help and land use rent free made this an easy choice.).  Over the next 5-10 years this was our strategy but then the late 90s internet bubble burst and our top selling product lines shrank.  Reducing manufacturing cost was the absolute goal and each factory's existence was questioned and needed to be separately justified.

Plus our Asia plant wasn't in China (Singapore) which was becoming a very high price place for manufacturing.  We needed experience with China and our board pointed out correctly that the risk was much lower now, good partners were available and we needed to be in China.  During the previous 10 years most of our suppliers had moved to China and the first step in every supply chain became doing as much in China as possible before finishing in the local factories where you still had them.  Silicon Valley had closed their very last IC plant (Intel), the once prolific Printed Circuit Board industry in the US had all moved to Asia.  Disk Drives were in Thailand and Indonesia (remember Seagate, Shugart and Connor Peripherals were once all here).  IC tech was mostly in Taiwan (UMC, TSMC).  Coastal China seemed to have an inexhaustible supply of labor coming in from the internal provinces and it had become harder and harder to find skilled workers in the US even slightly interested in manufacturing.  

It wasn't just high tech.  Television manufacturing and home appliance manufacturing that once dotted Appalachia had also moved out of the US.  They moved to Mexico because transferring a TV or washing machine by boat from China is 1000s of times more expensive than an iPhone.  For us our labor rates were 8-10x China and even if we used about 50% less for the same jobs the math didn't work out.  Plus in the intervening years most companies were now sourcing in the place that focused on manufacturing and provided the best options - and then shipping to the rest of the world.  The pendulum had swung from regional manufacturing to choosing the best low cost, focused option and OUTSOURCING rather than owning your own production.   And, yeah, that place was often coastal China.

Now when you are looking at large investments to keep up, your future is much less certain than it used to be and lowering cost immediately is the highest priority the cards are stacked against a capital equipment intensive renovation that will take 2-5 years to payoff.  

I could go over the analysis here but I'll just summarize.  Any good analysis will look at total cost and not just labor.  To be honest, the labor differential is discussed most in the press but that is not as big as it used to be and in 2020 is no longer the biggest savings.  In most high tech manufacturing analyses materials are a bigger driver so sourcing and combining your volume with the purchasing power of an EMS giant is worth something for small to medium/large companies.  Most will buy their A level strategic parts themselves and leave COTs parts to their EMS.  While direct labor is a driver you also look at the skilled labor to sustain a large plant and factor in your company's willingness to commit to manufacturing being treated as a strategic differentiator.  (Hint - Most high tech companies would prefer to differentiate on tech innovation and design).  Against this is the greater logistics cost of shipping from China to your markets and the difficulty of supporting NPI in Asia which often involves sending a large engineering contingent to China for several weeks for each new product. You also want to load factories at capacity and not turn them off which also favors EMS/Subcontractors with multiple customers.   At one large company I later worked for 75% of their volume was in 3.5 months of the year with the other 25% spread across the remaining 8.5.  Investing in dedicated equipment that would only be busy for 30% of the year is difficult to justify.   

I will mention again that everything from plastic molding, printing, packaging, IC production, Printed Circuit Board design, sheetmetal, and COT (common off the shelf) components had moved to Asia during the 18 years I had spent supporting internal manufacturing.  We had already moved a big part of our purchasing, component sourcing and planning to Asia just to stay close to our suppliers.

So ... in an uncertain future with a changing product line and smaller commitment from management we now faced a critical decision point.  The numbers showed a 30% cost advantage of going to Asia which would also mean shutting down internal factories.  As important to the C-Level management was the lower initial investment and lower sensitivity to a changing business model (read that as management's doubt in their future forecasts).

My last two years at that company I shutdown three factories I'd been involved in building from green fields 10-15 years earlier.  When I got sick of that, I changed my profession to supporting EMS, working on China to the US logistics and supporting small start-ups that needed someone who really understood manufacturing and NPI.  Turns out there are not that many people who have hands-on knowledge in the US any more.  While I like small companies most of my old team are working in manufacturing at the big 4 hardware tech companies at higher salaries than we ever got in our rearguard action.  

We'd fought a long campaign to keep our internal factories competitive but found when we finally lost that the other side needed us much more than you might have expected.

So can you restart US manufacturing now?  Maybe, but you will fight regions that now have a 30 year head start, a skilled workforce that sees manufacturing jobs as the best way to join the middle-class and a network of 1st, 2nd and 3rd tier suppliers that no longer exist in the US.   Better to win this at the NEXT manufacturing transition, with a technology that is still evolving and for where growing logistics costs work in your favor.  Don't try to win with the iPhone which has absolutely none of those three non-Asia advantages.

[Leftcoast]

Ok .... Quick view of my world through the late 80s, 90s and very early 2000s as a recently minted mid-80s Stanford IE.

My Senior year I joined the company I had interned with - a small but rapidly expanding networking company that would grow from $40M the year prior to my joining and peak just under $10B and then back to ~$400M.   I rode with it most of the way up and part of the way down.  You learn a lot both ways but going up is much, much more fun.

Initially we outsourced to several small subcontractors that would later merge and become big, big EMS conglomerates.  It tended to be expensive and slow (not nimble) so we decided to bring manufacturing inhouse.  We grossly underestimated how hard that would be and were pretty lousy at internal manufacturing the first year.  Second year we doubled down, got much more professional and were able to eventually move everything in house at higher quality levels and lower cost.  Our business doubled twice and our volume was now 10-20x what we started with (prices had decreased so revenue and volume were increasing at separate rates). 

We needed to expand rapidly in the early 90s and much of the growth was coming overseas.  China was now on the roadmap as a brobdingnagian potential market but still seen as a risky investment - maybe we should choose to work with a nearby and stable country ... like Singapore.  We really wanted to be part of the growing European common market as well. Do you build at the current US site, build your own factory overseas near your new customer base or do you go back to subcontracting for your overseas sourcing to reduce risk?   We were confident in our own capability so we decided to build factories right sized for the three major markets - US, Europe and China.  Reduce logistics, reduce taxation (near 0% for our new factories given government grants, plus payroll help and land use rent free made this an easy choice.).  Over the next 5-10 years this was our strategy but then the late 90s internet bubble burst and our top selling product lines shrank.  Reducing manufacturing cost was the absolute goal and each factory's existence was questioned and needed to be separately justified.

Plus our Asia plant wasn't in China (Singapore) which was becoming a very high price place for manufacturing.  We needed experience with China and our board pointed out correctly that the risk was much lower now, good partners were available and we needed to be in China.  During the previous 10 years most of our suppliers had moved to China and the first step in every supply chain became doing as much in China as possible before finishing in the local factories where you still had them.  Silicon Valley had closed their very last IC plant (Intel), the once prolific Printed Circuit Board industry in the US had all moved to Asia.  Disk Drives were in Thailand and Indonesia (remember Seagate, Shugart and Connor Peripherals were once all here).  IC tech was mostly in Taiwan (UMC, TSMC).  Coastal China seemed to have an inexhaustible supply of labor coming in from the internal provinces and it had become harder and harder to find skilled workers in the US even slightly interested in manufacturing.  

It wasn't just high tech.  Television manufacturing and home appliance manufacturing that once dotted Appalachia had also moved out of the US.  They moved to Mexico because transferring a TV or washing machine by boat from China is 1000s of times more expensive than an iPhone.  For us our labor rates were 8-10x China and even if we used about 50% less for the same jobs the math didn't work out.  Plus in the intervening years most companies were now sourcing in the place that focused on manufacturing and provided the best options - and then shipping to the rest of the world.  The pendulum had swung from regional manufacturing to choosing the best low cost, focused option and OUTSOURCING rather than owning your own production.   And, yeah, that place was often coastal China.

Now when you are looking at large investments to keep up, your future is much less certain than it used to be and lowering cost immediately is the highest priority the cards are stacked against a capital equipment intensive renovation that will take 2-5 years to payoff.  

I could go over the analysis here but I'll just summarize.  Any good analysis will look at total cost and not just labor.  To be honest, the labor differential is discussed most in the press but that is not as big as it used to be and in 2020 is no longer the biggest savings.  In most high tech manufacturing analyses materials are a bigger driver so sourcing and combining your volume with the purchasing power of an EMS giant is worth something for small to medium/large companies.  Most will buy their A level strategic parts themselves and leave COTs parts to their EMS.  While direct labor is a driver you also look at the skilled labor to sustain a large plant and factor in your company's willingness to commit to manufacturing being treated as a strategic differentiator.  (Hint - Most high tech companies would prefer to differentiate on tech innovation and design).  Against this is the greater logistics cost of shipping from China to your markets and the difficulty of supporting NPI in Asia which often involves sending a large engineering contingent to China for several weeks for each new product. You also want to load factories at capacity and not turn them off which also favors EMS/Subcontractors with multiple customers.   At one large company I later worked for 75% of their volume was in 3.5 months of the year with the other 25% spread across the remaining 8.5.  Investing in dedicated equipment that would only be busy for 30% of the year is difficult to justify.   

I will mention again that everything from plastic molding, printing, packaging, IC production, Printed Circuit Board design, sheetmetal, and COT (common off the shelf) components had moved to Asia during the 18 years I had spent supporting internal manufacturing.  We had already moved a big part of our purchasing, component sourcing and planning to Asia just to stay close to our suppliers.

So ... in an uncertain future with a changing product line and smaller commitment from management we now faced a critical decision point.  The numbers showed a 30% cost advantage of going to Asia which would also mean shutting down internal factories.  As important to the C-Level management was the lower initial investment and lower sensitivity to a changing business model (read that as management's doubt in their future forecasts).

My last two years at that company I shutdown three factories I'd been involved in building from green fields 10-15 years earlier.  When I got sick of that, I changed my profession to supporting EMS, working on China to the US logistics and supporting small start-ups that needed someone who really understood manufacturing and NPI.  Turns out there are not that many people who have hands-on knowledge in the US any more.  While I like small companies most of my old team are working in manufacturing at the big 4 hardware tech companies at higher salaries than we ever got in our rearguard action.  

We'd fought a long campaign to keep our internal factories competitive but found when we finally lost that the other side needed us much more than you might have expected.

So can you restart US manufacturing now?  Maybe, but you will fight regions that now have a 30 year head start, a skilled workforce that sees manufacturing jobs as the best way to join the middle-class and a network of 1st, 2nd and 3rd tier suppliers that no longer exist in the US.   Better to win this at the NEXT manufacturing transition, with a technology that is still evolving and for where growing logistics costs work in your favor.  Don't try to win with the iPhone which has absolutely none of those three non-Asia advantages.