Array
(
    [content] => 
    [params] => Array
        (
            [0] => /forum/index.php?threads/chip-globalization-is-over-and-sanctions-work-says-tsmc-founder.17611/
        )

    [addOns] => Array
        (
            [DL6/MLTP] => 13
            [Hampel/TimeZoneDebug] => 1000070
            [SV/ChangePostDate] => 2010200
            [SemiWiki/Newsletter] => 1000010
            [SemiWiki/WPMenu] => 1000010
            [SemiWiki/XPressExtend] => 1000010
            [ThemeHouse/XLink] => 1000970
            [ThemeHouse/XPress] => 1010570
            [XF] => 2021370
            [XFI] => 1050270
        )

    [wordpress] => /var/www/html
)

Chip Globalization Is Over and Sanctions Work, Says TSMC Founder

Daniel Nenni

Admin
Staff member
6e4fc3c45a16addb9785511b359a2c6a


(Bloomberg) -- Taiwan Semiconductor Manufacturing Co. founder Morris Chang declared globalization for the chip industry over and expressed support for US efforts to limit China’s tech advancement through export curbs and company sanctions.

“In the chips sector, globalization is dead. Free trade is dead,” Chang said at an event in Taipei Thursday. “Just look at the way China has been embargoed and the entity list. I agree with that.”

The 91-year-old industry pioneer said that the global chip supply chain will grow even more bifurcated as the US acts to curtail China’s access to the most advanced technology, and “I certainly support that part of American industrial policy to slow down China’s progress.”

Chang said China is at least five to six years behind Taiwan in chipmaking technology, but he also cautioned Taiwan should not be naive about its position relative to the US. When American leaders speak of “friend-shoring” high-tech manufacturing, Taiwan is not included in that policy, Chang said, as they’ve repeatedly voiced concerns about relying on Taiwan.

US Commerce Secretary Gina Raimondo described US dependence on Taiwan for advanced chipmaking as “untenable” in July, and her country’s Chips and Science Act is designed to entice companies like TSMC and South Korea’s Samsung Electronics Co. to set up operations within US borders. Taiwan and Korea have long been the epicenters of chipmaking — TSMC leads the way with cutting-edge logic processors like Apple Inc.’s bespoke Silicon parts while Samsung and SK Hynix Inc. dominate the memory market — which has led to increasing discomfort in the US and Europe.

Fears of China, which deems Taiwan part of its territory, attempting to seize the self-governing island by force have pushed some inside Washington to contemplate drastic measures. Among them, US officials have war-gamed scenarios where Taiwan’s sophisticated semiconductor industry would be wiped out to prevent it falling into China’s hands.

 
"TSMC has estimated chip costs in Arizona at 50% above its flagship production line in Taiwan, but the real level looks closer to double, said Chang." :ROFLMAO:

Either TSMC process engineers are the world's most incompetent, or we are about to see TSMC's margins rise to 70% off of the back of TSMC AZ.
 
"TSMC has estimated chip costs in Arizona at 50% above its flagship production line in Taiwan, but the real level looks closer to double, said Chang." :ROFLMAO:

Either TSMC process engineers are the world's most incompetent, or we are about to see TSMC's margins rise to 70% off of the back of TSMC AZ.

Obviously Arizona fab won't operate under the same level of cost as in Korea, Taiwan, or mainland China. Do you have some estimates or guess of a reasonable percentage, more expensive or cheaper, between Arizona and Taiwan?
 
I greatly respect Chang, but I hope he's wrong on free trade. Globalization seems to be on life support with grave consequences in tow, but cooler heads may prevail.
 
Obviously Arizona fab won't operate under the same level of cost as in Korea, Taiwan, or mainland China. Do you have some estimates or guess of a reasonable percentage, more expensive or cheaper, between Arizona and Taiwan?
My chemical engineer's intuition/general rules of thumb says less than 15% on the variable costs, 0% more expensive for tools (the single largest cost for new fabs), and probably significantly more to build the shell (but this depreciates over 29.5 years so the cost per year per wafer is pretty small).
 
Globalization isn't over at all, or even in decline. Trade and investment in or from China is declining. Western countries and companies are looking for alternatives to dependencies on China. China does not equal globalization.
 
My chemical engineer's intuition/general rules of thumb says less than 15% on the variable costs, 0% more expensive for tools (the single largest cost for new fabs), and probably significantly more to build the shell (but this depreciates over 29.5 years so the cost per year per wafer is pretty small).
This is an interesting and complicated topic for me. In many areas the cost difference between Taiwan and US are miles apart. For example the salary for a Taiwanese software or hardware engineer can be one third to one fifth of an American engineer. Healthcare cost is another significant one. A Taiwanese employee and his/her employer can pay one tenth (or less) of health insurance expense of their US counterparts.

All these cost difference can impact a fab's fix cost and operation cost. No matter it's sold to TSMC in Taiwan or TSMC in Arizona, an EUV machine from ASML might carry the same hardware and software price tag initially. But the price for installation, support, and consulting services can be drastically different. ASML might need to charge TSMC Arizona a much higher price in order to maintain the same level of profitability as it does in Taiwan.
 
This is an interesting and complicated topic for me. In many areas the cost difference between Taiwan and US are miles apart. For example the salary for a Taiwanese software or hardware engineer can be one third to one fifth of an American engineer. Healthcare cost is another significant one. A Taiwanese employee and his/her employer can pay one tenth (or less) of health insurance expense of their US counterparts.

All these cost difference can impact a fab's fix cost and operation cost. No matter it's sold to TSMC in Taiwan or TSMC in Arizona, an EUV machine from ASML might carry the same hardware and software price tag initially. But the price for installation, support, and consulting services can be drastically different. ASML might need to charge TSMC Arizona a much higher price in order to maintain the same level of profitability as it does in Taiwan.

Scotten did an analysis and, all things considered, I believe it is closer to a 10% wafer cost delta.

Morris has always been against building fabs outside of Taiwan and nothing will change that, especially at his age.
 
This is an interesting and complicated topic for me. In many areas the cost difference between Taiwan and US are miles apart. For example the salary for a Taiwanese software or hardware engineer can be one third to one fifth of an American engineer. Healthcare cost is another significant one. A Taiwanese employee and his/her employer can pay one tenth (or less) of health insurance expense of their US counterparts.
Okay so costing 5x more (and I guarantee you a technician or process engineer is not making 5x in the US over the ROC, but lets roll with that number) 3000 people spread across 600k WPY that amortizes to a 2.5% increase in labor expenses per wafer. Ontop of this labor is already a tiny percent of total wafer cost even for fully depreciated fabs (which won't be the case at AZ's production start).
All these cost difference can impact a fab's fix cost and operation cost. No matter it's sold to TSMC in Taiwan or TSMC in Arizona, an EUV machine from ASML might carry the same hardware and software price tag initially.
Hence the zero tool cost differnce
But the price for installation, support, and consulting services can be drastically different. ASML might need to charge TSMC Arizona a much higher price in order to maintain the same level of profitability as it does in Taiwan.
This goes into the variable cost bucket. As I also pointed out even with a huge increase to salary this has a minor impact given how many wafers there are per person. Given that N5 will be a very mature node by the time AZ ramps up TSMC AZ shouldn't require very much vendor support.
 
This is an interesting and complicated topic for me. In many areas the cost difference between Taiwan and US are miles apart. For example the salary for a Taiwanese software or hardware engineer can be one third to one fifth of an American engineer. Healthcare cost is another significant one. A Taiwanese employee and his/her employer can pay one tenth (or less) of health insurance expense of their US counterparts.

All these cost difference can impact a fab's fix cost and operation cost. No matter it's sold to TSMC in Taiwan or TSMC in Arizona, an EUV machine from ASML might carry the same hardware and software price tag initially. But the price for installation, support, and consulting services can be drastically different. ASML might need to charge TSMC Arizona a much higher price in order to maintain the same level of profitability as it does in Taiwan.

salary for a Taiwanese software or hardware engineer can be one third to one fifth of an American engineer.
It is true if you compare normal Taiwanese company engineers to top American company engineers.

TSMC just announced average salary of 2022. NT$3.17M (USD$104K)
On par with Intel.
 
Last edited:
Scotten did an analysis and, all things considered, I believe it is closer to a 10% wafer cost delta.

Morris has always been against building fabs outside of Taiwan and nothing will change that, especially at his age.

I believe you're talking about Scotten's old post in 2021:
Source: https://semiwiki.com/semiconductor-manufacturers/tsmc/303594-tsmc-arizona-fab-cost-revisited/

"The bottom line to all this is the cost for TSMC to make wafers in the US is only 7% higher than Taiwan if they built the same size fab complex in the US as what they have in Taiwan. Because they are building a smaller Fab complex the cost will be 17% higher but that is due to TSMC’s decision to build a smaller fab, at least initially."

So the cost difference between Taiwan and US in semiconductor manufacturing is only between 7% and 17%? IMO, it is extremally optimistic.

If this 7% to 17% cost difference is true, we don't even need to have the Chips Act or any discussion at all. The shrinking percentage of US domestic semiconductor manufacturing shouldn't have happened at all. Intel's fabs in Ireland and Israel, Texas Instruments' fabs in Japan and Germany, Micron's fabs in Taiwan, Japan, China, and Singapore, and Globalfoundries' fabs in Germany, Singapore are there for reasons. One of the major reasons is cost.

To many US companies, a 7% to 17% of cost saving won't even trigger them to consider a new vendor or to build new factory in a foreign country. It's not worthy it.
 
I believe you're talking about Scotten's old post in 2021:
Source: https://semiwiki.com/semiconductor-manufacturers/tsmc/303594-tsmc-arizona-fab-cost-revisited/

"The bottom line to all this is the cost for TSMC to make wafers in the US is only 7% higher than Taiwan if they built the same size fab complex in the US as what they have in Taiwan. Because they are building a smaller Fab complex the cost will be 17% higher but that is due to TSMC’s decision to build a smaller fab, at least initially."

So the cost difference between Taiwan and US in semiconductor manufacturing is only between 7% and 17%? IMO, it is extremally optimistic.
If anything it is pessimistic. Besides labor and building the shell, most large ticket items are not more expensive. The shell price is minor over the long life of that asset, and labor is a tiny fraction of total wafer costs. If you want to tell the world's industrial and chemical engineers that industry standard rules of thumb are wrong, you can always write a whitepaper on fab design for IEEE and chemical plant construction for AIChE. I would love to see ANY data that would even begin to suggest TSMC AZ wafer costs must be double what a similar greenfield site built in the ROC could do.

If this 7% to 17% cost difference is true, we don't even need to have the Chips Act or any discussion at all. The shrinking percentage of US domestic semiconductor manufacturing shouldn't have happened at all. Intel's fabs in Ireland and Israel, Texas Instruments' fabs in Japan and Germany, Micron's fabs in Taiwan, Japan, China, and Singapore, and Globalfoundries' fabs in Germany, Singapore are there for reasons. One of the major reasons is cost.
Most of the shrinkage was American companies becoming less pervasive. This logic also is not coherent either. If the US was 2x ROC, then why on earth would Germany, Israel, Ireland, or Japan be equal to the ROC? Also why does TSMC build in the ROC and not in Malaysia or Korea? The labor force is even cheaper there than in the ROC. Additionally the case you haven't noticed intel's AZ fab is way bigger than Ireland and a good bit bigger than Israel. If the US was a cost disaster why would intel's flagship fab be in AZ? California is a different beast, but there is a very good reason why everybody shuttered their CA fabs. As for Micron they make comedity products where price is the number 1 quality. That 7-17% difference in price is what could make the difference between 10% and 30% marketshare. In spite of this they still keep their US fabs open. Why you may ask? Because the cost difference is small enough that even in the hyper competitive DRAM/NAND industry, the cost difference is small enough that building a new fabs in SE Asia to replace their US/JP/ROC fabs wouldn't make any sense. The IRR for this project over continued use of a US fab would be on the time scale of centuries.

As for why CHIPS is needed look at the cost difference for greenfield sites vs expansions. The general rule of thumb that many within the chemical engineering community agree on is around 15% more expensive. Combine this with the excellent tax credits, access to cheap government loans, lower corporate income tax rate, and easier zoning in SK/the ROC is part of why you mostly only saw expansion to the fabs they already had at home. Besides cost there is also all of the extra difficulty of running a global manufacturing network. In spite of the FAR LOWER lower, building, and environmental compliance costs in the PRC you never saw large fab operations in the PRC... That is until they started paying for the whole damn fab. If low cost China needed to offer such huge incentives to get SK and Samsung to build up so heavily in the PRC, how is it shocking that the slightly higher cost US needs to offer competitive subsides for Samsung and TSMC to step outside their home turf to new greenfield locations?

2. Intel engineer's salary or US engineer's salary is generally much higher than Taiwan's. According to some surveys, the number can start with $130K or more.
No they don't. Software devs, sure. But software devs make obscene salaries. Fab people make nowhere near that kind of money.

pay.png
 
Last edited:
It is true if you compare normal Taiwanese company engineers to top American company engineers.

TSMC just announced average salary of 2022. NT$3.17M (USD$104K)
On par with Intel.

Two thoughts:

1. For US$100K annual salary (including benefits), TSMC and other top tier Taiwanese tech companies will certainly get a very capable and multi-years experience engineer who graduated from a top Taiwanese university.

For US$100K annual salary (including benefits), Intel will have some chances (if Intel is lucky) to recruit an entry level engineer from a top University.

2. Intel engineer's salary or US engineer's salary is generally much higher than Taiwan's. According to some surveys, the number can start with $130K or more.

 
If anything it is pessimistic. Besides labor and building the shell, most large ticket items are not more expensive. The shell price is minor over the long life of that asset, and labor is a tiny fraction of total wafer costs. If you want to tell the world's industrial and chemical engineers that industry standard rules of thumb are wrong, you can always write a whitepaper on fab design for IEEE and chemical plant construction for AIChE. I would love to see ANY data that would even begin to suggest TSMC AZ wafer costs must be double what a similar greenfield site built in the ROC could do.


Most of the shrinkage was American companies becoming less pervasive. This logic also is not coherent either. If the US was 2x ROC, then why on earth would Germany, Israel, Ireland, or Japan be equal to the ROC? Also why does TSMC build in the ROC and not in Malaysia or Korea? The labor force is even cheaper there than in the ROC. Additionally the case you haven't noticed intel's AZ fab is way bigger than Ireland and a good bit bigger than Israel. If the US was a cost disaster why would intel's flagship fab be in AZ? California is a different beast, but there is a very good reason why everybody shuttered their CA fabs. As for Micron they make comedity products where price is the number 1 quality. That 7-17% difference in price is what could make the difference between 10% and 30% marketshare. In spite of this they still keep their US fabs open. Why you may ask? Because the cost difference is small enough that even in the hyper competitive DRAM/NAND industry, the cost difference is small enough that building a new fabs in SE Asia to replace their US/JP/ROC fabs wouldn't make any sense. The IRR for this project over continued use of a US fab would be on the time scale of centuries.

As for why CHIPS is needed look at the cost difference for greenfield sites vs expansions. The general rule of thumb that many within the chemical engineering community agree on is around 15% more expensive. Combine this with the excellent tax credits, access to cheap government loans, lower corporate income tax rate, and easier zoning in SK/the ROC is part of why you mostly only saw expansion to the fabs they already had at home. Besides cost there is also all of the extra difficulty of running a global manufacturing network. In spite of the FAR LOWER lower, building, and environmental compliance costs in the PRC you never saw large fab operations in the PRC... That is until they started paying for the whole damn fab. If low cost China needed to offer such huge incentives to get SK and Samsung to build up so heavily in the PRC, how is it shocking that the slightly higher cost US needs to offer competitive subsides for Samsung and TSMC to step outside their home turf to new greenfield locations?


No they don't. Software devs, sure. But software devs make obscene salaries. Fab people make nowhere near that kind of money.

View attachment 1092

I wish Scotten's 7% to 17% cost difference between Taiwan and US is true and I wish you are correct that the cost disadvantage of a US fab is very limited.

But in reality, US and non US semiconductor companies haven't believed this for the past 20+ years. They have expressed their different opinions clearly by using their checkbooks and on their investment destinations. Those didn't fully understand this reality are currently in big troubles.
 
Two thoughts:

1. For US$100K annual salary (including benefits), TSMC and other top tier Taiwanese tech companies will certainly get a very capable and multi-years experience engineer who graduated from a top Taiwanese university.

For US$100K annual salary (including benefits), Intel will have some chances (if Intel is lucky) to recruit an entry level engineer from a top University.

2. Intel engineer's salary or US engineer's salary is generally much higher than Taiwan's. According to some surveys, the number can start with $130K or more.

Agree.

But engineer salaries cost have very little impact to Fab's competitiveness. UMC engineers probably only get 70% of TSMC engineers but gives UMC no advantage.

In my opinion, cost difference came from so many different small factors which ultimately makes huge cost difference.

For example, 3-6 months difference to build a new Fab. 3-6 months difference to improve yield for new nodes. For daily operation, almost all TSMC supplier are 24h stand by to fix the problem. Even the same 24h stand by duty, TSMC's culture (peer pressure) may force you to work extra harder to solve a little problem today instead of wait for tomorrow. Actually former TSMC top executive Konrad Young predicted that those TSMC employees moving from Taiwan to US. Afterworking with local American coworkers, they will find out there are more important stuff in life other than work so the culture will change immediately and will no longer work as hard as they do in Taiwan.

Intel Foundry Services have to dance with many customers at one in the future. Those small factors may result in a even larger cost difference.
 
Last edited:
Do we know what State subsidies are involved with the TSMC AZ fab complex? If TSMC gets a decent piece of the CHIPS Act that should also reduce the cost of building the fabs and maybe even running them.

Personally I do not think the AZ fab wafer costs will be more than 10% and it is clear that TSMC's top customers are behind the project so they will pay. Last time I heard TSMC would produce N4 and N3 wafers in AZ. Those will be VERY long nodes.

The TSMC Symposium is next month so we will know more then.
 
Do we know what State subsidies are involved with the TSMC AZ fab complex? If TSMC gets a decent piece of the CHIPS Act that should also reduce the cost of building the fabs and maybe even running them.

Personally I do not think the AZ fab wafer costs will be more than 10% and it is clear that TSMC's top customers are behind the project so they will pay. Last time I heard TSMC would produce N4 and N3 wafers in AZ. Those will be VERY long nodes.

The TSMC Symposium is next month so we will know more then.

Is this 10% cost difference you mentioned just for the 12 inches wafer material or the overall manufacturing cost?
 
Back
Top