What India should do to prep up its ambitious semiconductor plans
“We are on track to do in the next 10 years with these $10 billion what countries like China took 25-30 years and could not succeed,” Minister of State for Electronics and Information Technology Rajeev Chandrasekhar said earlier this week.
The minister was referring to India’s Production Linked Incentive (PLI) scheme for companies setting up fabs for semiconductors and displays ahead of the three-day SemiconIndia 2023 conference at Gandhinagar in Gujarat.
Statement, a sign of extreme bravery
For anyone who has tracked the global semiconductor industry, the statement will come as a sign of extreme bravery, if not wishful thinking. It may not be a coincidence that speaking at the SemiconIndia on Friday (July 28), Foxcon Chairman Young Liu observed that building an ecosystem for chips in India is “for the very brave”. However, as if to soften the blow, he added: “Where there is a will, there is a way.” Worth noting that Foxcon reneged on a JV with the Vedanta Group to set up a $20 billion semiconductor fab very recently.
Also Read: Audio & Story | India’s semi-plan for semiconductors
Deconstructing what the minister said, one can deduce three assumptions that the government is making. First is that 10 years is a good enough period to attain global leadership in semiconductors. Second, a $10 billion fund is big enough to set up a semiconductor ecosystem in the country. Three, China failed to succeed in its semiconductor objectives. Let us take up each of these, one by one.
Still a dream
India’s semiconductor ambitions are well over two decades old. It has had several false starts, with various rounds unleashing bouts of chip-frenzy. Policies have been announced at least three times, and several proposals made by companies.
The closest to setting up a fab was in 2007-08 when a semiconductor complex was seeded in Hyderabad and a second one in Gujarat, but the actual grounding of the projects never happened. Every big name in the business was bandied around as partners for these projects. India was not ready despite the intent.
As one of the people involved in the effort pointed out, choosing where you want to pitch yourself in the industry is of utmost importance. The semiconductor supply chain is a very complex mechanism with extreme dependencies on multiple players across continents.
Just consider this, making a single computer chip involves over 1,000 processes with several of them being conducted across several international borders before it reaches the end customer. It reportedly took TSMC, the world’s largest chipmaker, over 30 years to build a network of 2,500 top-tier suppliers and over 10,000 secondary suppliers.
No company in India could or can build such a network. Therefore, India needs to build its credentials one step at a time, instead of wanting to pole vault to the top with a stick that does not measure up to the task. There seems to be an apparent lack of clarity on where it wants to peg itself in the semiconductor supply chain and playing to its strength, which is semiconductor design.
Big boys, deep pockets
It is known that the global semiconductor industry is a turf that belongs to the big boys backed by governments throwing big money at them.
The EU endorsed on Thursday its own Chips Act, portending to mobilise over €43 billion ($48 billion) in public and private investments across the supply chain. Germany alone will be doling out €20bn ($22.15bn) in subsidies aimed. The US Chips Act promises $52.7 billion in grants to companies, American and otherwise, setting up or expanding operations in the country. Likewise, South Korea and Japan too have put together war chests that promise to be sizeable.
Also read: C295 a baby step; aircraft making in India has a long way to go
China has outdone all with its Integrated Circuit Industry Investment Fund (aka the Big Fund), starting off way back in 2014 by raising close to $50 billion in two phases. The overall funds available for semiconductor investments in China under various mechanisms have grown to nearly $150 billion.
Clearly, India’s $10 billion PLI scheme for semiconductors could be too little too late. Success is relative. China has set itself a target of producing 70 percent of its own chips by 2025, up from 16 percent in 2020. While it is anybody’s guess if these ambitious targets will be met, the Chinese semiconductor plans are by no means a failure.
China’s ambition
Before the chip-war with the US, China’s plans worked well with most global companies setting up capacities there. This enabled the country to develop its own capabilities apart from acquiring technologies needed to build legacy chips.
For instance, China has gained significant lead in the back end of the semiconductor value chain, with market shares in chip design, trailing-edge wafer capacity (legacy chips) and ATP (Assembly, Test and Packaging) that are larger than European companies.
To illustrate, 90 percent of the revenues of SMIC, China’s largest foundry, came from mature nodes of 40-250 nm in 2019. However, this is where the US wants to contain China.
China lags on the more advanced 5-14 nm nodes and technologies and materials for advanced chips like chemicals of high purity and precision lenses, mirrors, valves and tubes apart from access to sophisticated Electronic Design Automation tools and Deep or Extreme Ultraviolet Lithography tools. Even as it struggles with semiconductor IP, China seems to be getting there, particularly for 7-5 nm nodes and AI chips, albeit with make-do processes and technologies that may not be optimal.
US-China tensions
The US Chips Act, announced in August 2022, seeks to choke off access to all matured chip making technologies for China, whose success in mastering chip technologies is the very cause for fear in the West. At the same time, the Chips Act addresses the threat of a Chinese blockade of Taiwanese capacities, or even an attack on them, leading to “fab nearshoring” exercise.
Consolidation of global chip manufacturing capacities in the US and other Western allies, dubbed the Fab-4 (US, Japan, Taiwan, and South Korea apart from Europe), is currently on. Investments close to $200 billion have either already been grounded or are on the drawing boards. At least 40 new Fab facilities are said to be on the anvil across 16 US states.
Late starter
At a time when the global semiconductor supply chain is undergoing a geographical shift resulting in a no-holds-barred contest of wits, India is yet to get off the block.
Notwithstanding, shows and industry conclaves, business realities are what dictate investment decisions that too in a highly cyclical industry like semiconductors which is prone to steep troughs every few years. In stark contrast to the COVID spurt in demand, the semiconductor market has been going through a downturn since last year.
It will likely last for another year thanks to the macroeconomic outlook and semiconductor industry demand-capacity imbalance.
But what queers the pitch for the industry is the spate of new foundry investments planned purely for geopolitical reasons. While it remains to be seen how many of them will take off, if they do, then industry economics will be off balance.
Indian dreams
There is every likelihood of a glut in the market even if half of these investments fructify in the next five years, including for legacy nodes like the ones India is looking to enter.
The more important question is whether the industry will bite the bullet notwithstanding the positive statements and the low-level interest actually on ground. While the guessing game goes on, the Design Linked Incentive (DLI) seems more likely to succeed purely because it caters to India’s inherent strengths in the area.
Some 100 companies, primarily startups and micro, small and medium Enterprises (MSMEs), seek funding under the scheme. The scheme focuses on three components — chip design infrastructure support, product design linked incentives and deployment linked incentives.
That is what India should be doing – forget comparing itself to China and focus on what is doable and play to its own strengths.
(The writer is a Hyderabad-based journalist and has covered business and technology trends for the last three decades.)
(The Federal seeks to present views and opinions from all sides of the spectrum. The information, ideas or opinions in the articles are of the author and do not necessarily reflect the views of The Federal.)
