Array
(
    [content] => 
    [params] => Array
        (
            [0] => /forum/index.php?threads/semiconductor-investments-won%E2%80%99t-pay-off-if-congress-doesn%E2%80%99t-fix-the-talent-bottleneck.15998/
        )

    [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
)

Semiconductor Investments Won’t Pay Off if Congress Doesn’t Fix the Talent Bottleneck

Daniel Nenni

Admin
Staff member

National security experts who have reviewed the semiconductor situation have converged on international talent as a key area for reform. The National Security Commission on Artificial Intelligence, the GOP China Task Force, the bipartisan Future of Defense Task Force, and the President’s Council of Advisors on Science and Technology have all published reports identifying semiconductors as a critical need for U.S. national security and international talent recruitment as a vital part of any strategy to promote U.S. leadership. The artificial scarcity of semiconductor talent is already a major challenge in growing U.S. semiconductor capacity and will only get worse if the U.S. funds new fabs without expanding the talent pipeline. High-skilled STEM immigration will be necessary for success.
 
Dan, do you think platforms could be developed or current ones expanded to greatly leverage the available talent as is being done in other fields? Semis of all types have leveraged talent and resources greatly thanks to the semi revolution which has given us numerous tools to leverage talent greatly. Could there be a "Platform Revolution" coming to all fields of human endeavor that advances based on the advances in the semi ecosystem that has already leveraged many talents and resources to levels once considered impossible? Thanks
 
National security experts who have reviewed the semiconductor situation have converged on international talent as a key area for reform. The National Security Commission on Artificial Intelligence, the GOP China Task Force, the bipartisan Future of Defense Task Force, and the President’s Council of Advisors on Science and Technology have all published reports identifying semiconductors as a critical need for U.S. national security and international talent recruitment as a vital part of any strategy to promote U.S. leadership. The artificial scarcity of semiconductor talent is already a major challenge in growing U.S. semiconductor capacity and will only get worse if the U.S. funds new fabs without expanding the talent pipeline. High-skilled STEM immigration will be necessary for success.

US semi salaries are already up to 4-5 times higher than Taiwanese.
 
US semi salaries are already up to 4-5 times higher than Taiwanese.

High salaries means more Americans will come to semiconductors which is good. In my opinion if you have to actually go to work versus work at home salaries will have to be higher. The downside is (as Morris Chang has said many times) that manufacturing semiconductors in the US will be more expensive than Taiwan or China or Korea and consumers around the world will not pay more for products just because they are made in the USA. We have tried the "Born in the USA" marketing versus foreign made products but still US households are filled with non USA products because they are cheaper and the quality is good enough and sometimes better.

And now that inflation is running amok it will be even harder to justify buying more expensive US products. So good luck reshoring manufacturing for higher costing electronic products. Just my opinion of course.
 
High salaries means more Americans will come to semiconductors which is good.

High salaries? Coding HTML in Google pays more for junior software developer than working in the process engineering for 5 years. Any proper software development + 10 years experience push it to $200k — more than unit GMs get in most semi companies.
 
High salaries? Coding HTML in Google pays more for junior software developer than working in the process engineering for 5 years. Any proper software development + 10 years experience push it to $200k — more than unit GMs get in most semi companies.

I have been hearing about layoffs looming for Google, Facebook, Twitter, etc... I don't know about the rest of the world but things are changing here in Silicon Valley. Cost of living is skyrocketing. Violence, suicides, and drug related deaths are at all time highs. Changes will have to be made, absolutely.

New fabs will have to hire and the locations will have a lower cost of living and a better quality of life? We shall see....
 
High salaries? Coding HTML in Google pays more for junior software developer than working in the process engineering for 5 years. Any proper software development + 10 years experience push it to $200k — more than unit GMs get in most semi companies.
A programmer with four years of experience got a job at Facebook last year and the total compensation is about 220k.
 
US semi salaries are already up to 4-5 times higher than Taiwanese.
I didn't believe your post, but I just did a bunch of searches on engineering salaries and TSMC manufacturing salaries in Taiwan, and if anything your estimate of the multiplier is a little low. It looks to me more like 6:1, based on comparing NT$ to US$. Thanks for the perspective.
 
High salaries? Coding HTML in Google pays more for junior software developer than working in the process engineering for 5 years. Any proper software development + 10 years experience push it to $200k — more than unit GMs get in most semi companies.
The engineer salary for TSMC and Intel are very similar. The direct labors (not engineers) in TSMC earn $30K only which is a lot less than Intel direct labors.

While Faang and 100+ software companies pay 3 times more than Intel. That's the real problem.
 
The engineer salary for TSMC and Intel are very similar. The direct labors (not engineers) in TSMC earn $30K only which is a lot less than Intel direct labors.

While Faang and 100+ software companies pay 3 times more than Intel. That's the real problem.

Actual process RnD Senior Eng. was $34k at UMC just 10 years ago. This is why I never went to process, or semiconductor industry as such. Even as a self-taught EE/ME/FWE/WebDev carpetbagger, and even as a random engineering stuff salesman, I earned way more in my early career than some people who also started 10 years ago earn today.
 
I'd have a good laugh at this with my millennial semiconductor friends, but most have left the industry for software or finance to make 2x more money doing half the work, so they'd just be laughing at me.

Congress could go after rentiers and healthcare companies that make ~50% of cost of living in America, a staggering amount for any first world country. But nooo instead they just encourage more neoimperialism to back out of a mess made by neoimperialism.

An immigrant engineer who does whatever their manager asks of them (if they want to be an American) is not the same as any ol' engineer who freely enjoys the craft and working with others to build stuff.
 
The cost of healthcare in the US is staggering in cost and low in quality. Logic and fairness currently have no place in many labor markets. The first thing anyone must learn is "Life is not fair". All one can do is seek their own path and hope for the best.
 
The engineer salary for TSMC and Intel are very similar. The direct labors (not engineers) in TSMC earn $30K only which is a lot less than Intel direct labors.

While Faang and 100+ software companies pay 3 times more than Intel. That's the real problem.

A large chuck of FAANG's TC come from stocks, J Powell is fixing this problem by crashing the stock market.
 
3 times difference.

Downtown Taipei is as expensive as most of first tier world cities, but even comparing TP to Shenzhen, you feel the later just provides fancier life, and feels "fresher"

Taiwan as a whole feels away from hustle, and bustle of the rest of East Asia

My feeling nothing changed there since the Asian crisis

So, if the choice was in between Shenzhen, and Taipei, and the little problem of Shenzhen being situated in a totalitarian dictatorship wasn't there, the choice would've been Shenzhen by a big margin
 
Last edited:
The engineer salary for TSMC and Intel are very similar. The direct labors (not engineers) in TSMC earn $30K only which is a lot less than Intel direct labors.
VCT, you state "engineer salary for TSMC and Intel are very similar"--> Can you state the role and the salary?

FWIW I believe Morris Chang, in the Brookings Institution interview, when he stated "the same product, the Oregon cost is about 50% more than the Taiwan cost" (32:20)

Earlier in that interview, which is a series "Vying for Talent", he mentions that from the 1950s to the mid-1970s the US had manufacturing talent, in his estimation, in semiconductor manufacturing. After about 1980 though the top people went into higher paid fields like finance and consulting, judging by where the top schools send their graduates.

I have something to add to Mr. Chang's analysis of TSMC Oregon costs. I worked at Qimonda up until 2008, when the fabs shut down in Richmond. For a year prior to that we had McKinsey onsite improving our processes, using Lean Six Sigma, Factory Physics, and other modern techniques. We reduced our cycle time but our costs remained too high to compete with Samsung in DRAM and so had to shut down.

Qimonda Richmond (RIP), TSMC Oregon, Samsung Austin all had/have this same problem, where we need to pay more to attract better people, but in Korea and Taiwan they don't have to, they get their talent without the finance and consulting syphon, which drives up cost.

So salaries are pretty key, and they need to be higher to compete with consulting and finance, but that makes cost higher too, which is already too high. It's a catch 22.
 
Last edited:
Qimonda Richmond (RIP), TSMC Oregon, Samsung Austin all had/have this same problem, where we need to pay more to attract better people, but in Korea and Taiwan they don't have to, they get their talent without the finance and consulting syphon.
As I pointed out in a previous post in another thread, TSMC Oregon is in Camas, WA, which is a really dumb place to locate a fab from a recruitment perspective. It's almost as if it was sited to fail. Hillsboro is where the Intel Oregon fabs are, and over 20,000 Intel employees work. Camas is a city of 26,000 people about a 60min drive away, on a Sunday afternoon in May. Imagine the commute in the winter, when ice is common in the Columbia River corridor.

 
VCT, you state "engineer salary for TSMC and Intel are very similar"--> Can you state the role and the salary?

FWIW I believe Morris Chang, in the Brookings Institution interview, when he stated "the same product, the Oregon cost is about 50% more than the Taiwan cost" (32:20)

Earlier in that interview, which is a series "Vying for Talent", he mentions that from the 1950s to the mid-1970s the US had manufacturing talent, in his estimation, in semiconductor manufacturing. After about 1980 though the top people went into higher paid fields like finance and consulting, judging by where the top schools send their graduates.

I have something to add to Mr. Chang's analysis of TSMC Oregon costs. I worked at Qimonda up until 2008, when the fabs shut down in Richmond. For a year prior to that we had McKinsey onsite improving our processes, using Lean Six Sigma, Factory Physics, and other modern techniques. We reduced our cycle time but our costs remained too high to compete with Samsung in DRAM and so had to shut down.

Qimonda Richmond (RIP), TSMC Oregon, Samsung Austin all had/have this same problem, where we need to pay more to attract better people, but in Korea and Taiwan they don't have to, they get their talent without the finance and consulting syphon, which drives up cost.

So salaries are pretty key, and they need to be higher to compete with consulting and finance, but that makes cost higher too, which is already too high. It's a catch 22.
The salaries cost was not a big part of semiconductor manufacturing. The major cost is the equipment and the cost is highly related to the production yield rate.

""the same product, the Oregon cost is about 50% more than the Taiwan""

In Taiwan, after work, people get a phone call and get back to the fab to solve the problem immediately. Those little stuff accumulate into big yield difference and cost difference.


Also the other rumors is that TSMC employees have higher interest to go to TSMC Japan instead of TSMC Arizona.
 
Last edited:
In Taiwan, after work, people get a phone call and get back to the fab to solve the problem immediately. Those little stuff accumulate into big yield difference and cost difference.
I know people who work in Intel's manufacturing group who work on problem-solving. They're on call 24/7, depending on their role. The problems get worked immediately. Even in the product design groups, I know people who have been called back from sabbaticals. Intel has a lot of problems, but I've never noticed that a lack of sense of urgency was one of them.
 
Back
Top