Ideas

The Chip Conundrum: Filtering What Is Relevant For #FabInIndia

Arun Mampazhy

Apr 08, 2021, 05:44 PM | Updated 05:44 PM IST


A semiconductor plant in India.  (Representative image)
A semiconductor plant in India.  (Representative image)
  • A gentle advice for the government would be: “Do not tire out a good proposal, in the wait for a perfect one”.
  • Ever since my previous article was published in Swarajya (30 January 2021), news items related to the following topics have been doing the rounds in the Indian media, and I will cite mostly the corresponding Swarajya reports.

    1) Chip shortages and their impact, especially on the automotive industry (17 February).

    2) Intel, having fallen behind TSMC and Samsung in terms of being able to successfully manufacture the sub-10nm chips, around mid-March announced a $20 billion plan. More than the money, what is important is that it is opening up its fabs for external customers, that is to say, it is going back to what a real integrated device manufacturer (IDM) should have done all along.

    3) Pressure had been building up on US President Joe Biden since February to incentivise new fabs in US, and by early April there is news of a $50 billion push being approved.

    4) Amidst all this, the TSMC report was interesting. Its $12 billion Arizona plant was approved in November 2020 itself, and European Union too has been wooing TSMC for a while now. But post the Intel $20 billion announcement, there seems to have been a “$100 billion expansion in 3 years” plan in the news.

    However, there are reports of the TSMC chairman calling this unrealistic and non-profitable. That seems quite logical because it is near impossible that the semiconductor world is suddenly facing such a shortage that it needs $120 billion investment in a few years to cope with. And, mind you, companies like Samsung, and the Chinese government too have plans running into tens or even hundreds of billions of dollars to expand chip manufacturing.

    There is a shortage, but not to the extent warranting such big money to be invested in such a short time. It looks like it is more of geopolitics that is driving some of these decisions –– which is fine as long as it is aligned with market realities.

    So what is India doing? Union Ministry of Electronics and Information Technology (MEITY) put out a call for proposals on 15 December 2020, initially with a 31 January deadline. But the deadline has since been extended thrice –– to 28 February, 31 March and the latest is 30 April.

    The latest extension came right after a purportedly “leaked” information that India may offer $1 billion or more for every approved proposal as cash incentive ‘picked up’ by Reuters and then publicised all over the media. Sometimes, ‘leaks’ get more publicity than ‘straight talk’, so be it.

    Why is the government of India extending the deadline again and again? In the absence of a clear answer from MEITY, one will have to assume. So let us consider both the good assumption and the not-so-good one.

    Many have been appreciative of the fact that unlike in the past where the government of India came up with a policy for supporting fabs and told the stakeholders "take it or leave it", this time it has been in a consultative mode. The promise is that it will go through the demands of the ones who propose and do its best to incentivise.

    If that were to be assumed as the mood –– that this time around India is desperate to have some commercial #FabInIndia pick up –– that is positive.

    On 20 March, there was news that “in order to facilitate investments and promote manufacturing in technology-intensive sectors including semiconductors, the government of India has decided to constitute an empowered committee for manufacturing in high technology areas”.

    Right after that, there was news that MEITY is trying to reach out to the semiconductor giants –– the names listed include TSMC, UMC , Intel, Micron etc.

    Unless I missed it, three names from the top-10 foundries are missing –– that of Samsung, GlobalFoundries and Tower semiconductors. This is leading to speculation that they are already part of proposals that are at MEITY's desk, but there are not many ways of confirming that unless MEITY speaks out or one of the companies do.

    While there is no such official confirmation from MEITY, at least one name did come out in reports.

    A consortium of investors led by Abu Dhabi-based fund Next Orbit Ventures (NOV) has reportedly expressed keen interest in setting up in India.

    It was also reported that NOV declined to comment. However, to be honest, what they are trying to do is available publicly, if only one cares to look for it. A 14 December 2017 news report from their website says:

    Next Orbit Ventures, a newly launched Abu Dhabi-based fund established under the Abu Dhabi Global Markets (ADGM) jurisdiction, has announced the launch of $2 billion fund for building semiconductors and electronics fab ecosystem in India. Next Orbit Ventures intends to raise $1.5 billion from the Gulf region as feeder fund, the remaining $500 million has already been secured by Next Orbit Ventures Fund-II from a consortium of investors involving both the Indian government and UHNIs.

    The fund has already signed agreements with technology providers UMC (Taiwan), AMD (US), TowerJazz (US) and Centrotherm PV (Germany) as technology licensees for its projects. Ajay Jalan, founder and managing partner of the fund said discussions are on with Andhra Pradesh and Gujarat to secure land for the project.

    The project aims at building the required infrastructure to cater to India’s fast-growing market for electronic goods and components.

    Their big, wholesome plan, from their webpage is as follows:

    Indications are that the group already has submitted a good, solid, strong proposal for the Analog and Mixed Signal (AMS) fab and the chosen location is most likely the Dholera smart city in Gujarat. It is not clear whether their efforts to go ahead with a proposal for digital fab have also been successful.

    Looking at the technology partners listed above, the likely partner for a digital fab will be UMC. It is a very practical and wise decision because as the world woos TSMC, which itself has started giving indications that it cannot put its leg into too many things, getting the ‘second in Taiwan’ and ‘fourth in the world’ is a much more practical path for India.

    From UMC's perspective too, it makes sense to expand outside an already water-strained Taiwan and a never-know-what-will-happen ‘mainland’ China.

    However, indications are that it may take a bit more reaching out and convincing from India's side for them to press the button –– not just the ‘$1 billion’ kind, but some political and market kind.

    Will the proposal deadline be extended beyond 30 April 2021 also? It almost appears that the government of India has some ‘target’ in mind –– it may be waiting for some heavy weight proposals. There is nothing wrong in it. However, my gentle advice would be, “Do not tire out a good proposal, in the wait for a perfect one”.

    Arun Mampazhy has a BTech from IITM and MS from University of Maryland in semiconductor fabrication and over a decade of industry experience. His dreams of seeing a commercial fab takeoff in India has changed from black and white to colour over two decades. He can be reached via email nanoarun(at)gmail(dot)com or @nano_arun on twitter. Views expressed are personal.


    Get Swarajya in your inbox.


    Magazine


    image
    States