Array
(
    [content] => 
    [params] => Array
        (
            [0] => /forum/index.php?threads/national-defense-authorization-act-ndaa-enactment-too-little-too-late.13492/
        )

    [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
)

National Defense Authorization Act (NDAA) enactment: Too little too late?

Fred Chen

Moderator

“Enactment of the NDAA will help propel U.S. economic growth and cement America’s position at the forefront of semiconductor innovation, which is critical to the game-changing technologies of today and the future,” said Bob Bruggeworth, President, CEO, and Director of Qorvo and SIA Chair. “The next step is for leaders in Washington to fully fund the NDAA’s domestic chip manufacturing incentives and research initiatives. Doing so will make America one of the world’s most attractive places to produce this foundational technology.”

The provisions authorized by the NDAA must still be funded through appropriations made by Congress. SIA and the Boston Consulting Group released a study in September finding robust federal incentives for domestic semiconductor manufacturing would reverse the decades-long trajectory of declining chip production in America and create as many as 19 major semiconductor manufacturing facilities, or fabs, and 70,000 high-paying jobs in the U.S. over the next 10 years. A separate SIA report released in June found federal funding for semiconductor research generates an outstanding return on investment for taxpayers, with each additional dollar invested adding $16.50 to U.S. gross domestic product.
 
I view these things very skeptically. Here’s why.
NDAA is a defense program, and GaAs fabs fit this description; IC for space, radiation hardened, or high speed; small scale production. It’s not what we think of as a modern fab. Qorvo, Skyworks, Sandia National labs, and others are part of this small semiconductor specialty.
70,000 jobs is just pie in the sky.
The key: NDAA is appropriations. That is 100% political pork-barrel politics. It means jobs in New Mexico and California, when money can be appropriated. Imagine “19 facilities” or “70,000 jobs” based on the next omnibus appropriations bill. Would you want to depend on that for your livelihood? Is that any way to run a business?
 
Too little maybe, but it's never too late. If we are late, then we should do even more to make up for the lateness. "pork-barrel"?? are you serious? Please open your eyes and look around what other countries or governments are doing.
 
Exactly. We need to do whatever we can to regain our leadership in microelectronics - a field that was started in the US and needs to be reinvigorated. Whether it be for consumer electronics or defense, we must be in control of our own destiny. Other countries heavily subsidize the semi industry, making it difficult for us to remain competitive. Investing in innovation is the best way to engage our best scientist and engineers to come up with the next breakthroughs.
 
I view these things very skeptically. Here’s why.
NDAA is a defense program, and GaAs fabs fit this description; IC for space, radiation hardened, or high speed; small scale production. It’s not what we think of as a modern fab. Qorvo, Skyworks, Sandia National labs, and others are part of this small semiconductor specialty.
70,000 jobs is just pie in the sky.
The key: NDAA is appropriations. That is 100% political pork-barrel politics. It means jobs in New Mexico and California, when money can be appropriated. Imagine “19 facilities” or “70,000 jobs” based on the next omnibus appropriations bill. Would you want to depend on that for your livelihood? Is that any way to run a business?

Indeed, the defense department is unlikely to fund processes and designs for better consumer smartphones and notebooks, which are the volume drivers. They would go for specialized electronics.
 
So, suppose you are a politician, what would you do for semiconductors, beyond NDAA?
- Start a “deep tech” venture capital fund, something like Norway‘s or Abu Dhabi’s sovereign wealth fund, with real venture capitalists, with a 50% domestic content mandate?
- Do a green new deal with solar and wind energy (which use a lot of electronics with a lot of ICs) with the requirement that the power electronics have 50% domestic content?
- Do a “brown new deal” so that rare earths can be refined in the USA, instead of China, despite the environmental impact? Solve the thorium waste issue with new nuclear reactors.

After you’ve done all these reasonable things, you would find it has no impact on Intel, or AMD, or other existing USA-based global chip companies. Which to me, would be fine, but I think it disappoints the people who want old businesses to become vibrant like we remember from the 80s or 90s. Which just isn’t how it works.
 
Last edited:
The history of the bill is here:https://en.wikipedia.org/wiki/Willi...efense_Authorization_Act_for_Fiscal_Year_2021

Although semiconductors are not mentioned there, another article discloses Sen. Schumer's effort to help Cree Semiconductor (specializes in RF) in New York with the provision: https://www.wktv.com/content/news/N...ry-good-news-for-Cree-facility-573331811.html

In the end, though, it's just an authorization, not binding. That last stipulation sounds ominous, but maybe Cree and Skywater might still be confident about it. Skywater already has some DoD support: https://www.skywatertechnology.com/...tion-to-support-dod-investment-of-up-to-170m/

My understanding is that TSMC's 5nm Arizona deal has both state and federal support, but maybe this was already secured in 2020, so this is something else? Even in that case, it still can be considered too late.
 
Last edited:
So, suppose you are a politician, what would you do for semiconductors, beyond NDAA?
- Start a “deep tech” venture capital fund, something like Norway‘s or Abu Dhabi’s sovereign wealth fund, with real venture capitalists, with a 50% domestic content mandate?
- Do a green new deal with solar and wind energy (which use a lot of electronics with a lot of ICs) with the requirement that the power electronics have 50% domestic content?
- Do a “brown new deal” so that rare earths can be refined in the USA, instead of China, despite the environmental impact? Solve the thorium waste issue with new nuclear reactors.

After you’ve done all these reasonable things, you would find it has no impact on Intel, or AMD, or other existing USA-based global chip companies. Which to me, would be fine, but I think it disappoints the people who want old businesses to become vibrant like we remember from the 80s or 90s. Which just isn’t how it works.

This is the only recent example I've seen: https://www.governor.ny.gov/news/go...onductor-acquire-and-preserve-globalfoundries No federal help, though. And of course, this scale is too small. We're expecting the US government to effectively buy GlobalFoundries back and put it back on 7nm and smaller.
 
This is the only recent example I've seen: https://www.governor.ny.gov/news/go...onductor-acquire-and-preserve-globalfoundries No federal help, though. And of course, this scale is too small. We're expecting the US government to effectively buy GlobalFoundries back and put it back on 7nm and smaller.

Can't wait to see the look on Cuomo's face when he learns how that gang of ex-IBM cynics have been working with McK to rip the copper out of the walls at GF for years now.

If it's anything like COVID, the outcome won't matter because it's about the book deal to detail how he saved America with his leadership, and the NYT constituency will slop it up like hogs.
 
@coldsolder215 No need for heated rhetoric here.

Malta is the ONLY green field 300 mm factory built in last years (not an expansion). 19 new factories is pure fantasy, no matter what SIA or DoD says.

ASMC 2021 is organizing a panel discussion in May that will cover "the geography of manufacturing, why things are built where". See SEMI.org for details. Complicated topic.

Semi manufacturing is in the same boat as all US-based manufacturing: training/education, healthcare costs, NIMBO, competition from GAFA for talent, infrastructure (power, water).

We will keep the factories we have (albeit under a different name in case of Intel sites), get a new factory in the form of TSMC-AZ and a few expansions in TX and NY but that will be it, DoD or not.
 
Back
Top