To guarantee that subsidies by each nation do not stand in the way of India’s semiconductor goals, as stated by the widely publicised semiconductor policy published in December 2021, the US Department of Commerce and its Indian counterpart recently reached an agreement in March 2023. In order to strengthen industry-specific export control legislation in the semiconductor industry, the United States Department of State has also engaged with India. According to recent media reports, India has agreed to this.
They appear to be prerequisites in a protracted dance between officials on both sides, but can they convince a multinational corporation like Intel to invest in India and build a new, $10 billion 300mm wafer production facility there?
The premise of this question is based on the two-year-old courtship between Indian government ministers and Intel, who were observed courting the CEO of Intel in Davos. New Delhi is hoping that Intel Foundry Services, a business unit within Intel created after Intel acquired the U.S.-Israel-based Tower Jazz Speciality Foundry, will construct a greenfield plant, most likely in Dholera, Gujarat. In a world where Intel is still focused on establishing fabs within the United States, the realities on the ground do not support such an outcome. What direction does Indian semiconductor strategy need to take now?
The then-central government established the Semiconductor Laboratory (SCL) in Mohali in 1983 with the goal of establishing an electronics ecosystem at a time when Keltron, Uptron, and Webel were just starting up in a pre-liberalized India focused on consumer electronics. These expectations were crushed, however, by shocks like the 1991 consumer goods market opening and the SCL fire that started in 1989. The facility has, for the most part, remained an unfulfilled dream in its aim of developing a local semiconductor ecosystem, although receiving some support from the government exchequer to resurrect the factory to a 180 nm node pilot line to address the strategic demands of the country.SCL Mohali may be seen as a technological stack like to others like Aadhar, Aarogya Setu, and the Unified Payments Interface (UPI) that has a multiplier effect and encourages numerous integrated circuit design firms in India to think about building for India.
With the return of SCL to the Ministry of Electronics and Information Technology (MeITy) after a 15-year sojourn as a laboratory inside the Department of Space, as part of the new semiconductor strategy announcement in December 2021, the institutional basis for such a change in emphasis already exists. After such an announcement, however, no joint venture partner has been identified, leaving SCL employees in a precarious situation. At this time, MeITY appears to be concentrating on convincing Intel to establish a fab in India. This is implied by the phrasing of the request for proposals and the MeITY Ministers’ signals. The SCL upgrade for Intel, which predominantly uses a 22nm node and 300mm, cost about $10 billion.
Another strategy would be to use the SCL’s people and financial resources to expand on what already exists and accelerate the semiconductor mission by utilising recent technical advancements in a class of semiconductors that do not require sophisticated lithography technology. The “More than Moore” sector of >180 nm node involves wide bandgap (GaN, GaAs, Silicon Carbide), mixed signal analogue (BCD and SiGe), and silicon carbide for RF and power industries while utilising previously installed lithography technology at the SCL.In this case, a $50–$100 million investment could lead to the creation of Indian solutions for automotive electronics (EV traction inverters/on board chargers), PV inverters, 5G infrapower ampliers, and railway electronics (traction inverters), resulting in the creation of the Indian equivalents of Bosch, Siemens, ABB, Mitsubishi Electric, Thales, and ELTA.
The upgrade must, however, be supported by subsidies targeted at fabless design firms with proven designs (sales of more than $100 million annually) that are prepared to produce at the SCL in the 180nm+ node (and possibly transfer process intellectual patents if they have any). The incentives must be directed towards multinational design firms that produce goods for the Indian market, such as motor drives for BLDC fans or e-bike chargers. Regrettably, established international fabless design businesses are not given any such incentives under the current DLI/PLI programmes.
The India Semiconductor Mission’s recent initiatives to extend subsidies to small and medium-sized businesses worldwide in the upstream supply chain are applaudable since they will benefit an already-existing facility like the SCL. Yet unless combined with the above-described incentives and enhancements aimed at various player groups, this is insufficient on its own.
The stakes are enormous because, barring a change in direction on incentive objectives, India may once again entirely lose out on the semiconductor manufacturing bandwagon due to a lack of clarity and inertia. Lastly, rather of having a career scientist from the Department of Space, as is the case currently, the SCL requires a fulltime director with past “More than Moore” foundry expertise to carry out this strategy over the next five years. This is due to the complexity of the market that must be met.