Envisioning a Semiconductor Aatmanirbhar Bharat

Today, investing in one semiconductor manufacturing unit, known as a ‘fabrication plant’ (fab), requires nearly a billion US dollars. While countries like China and South Korea possess pre-established technical ecosystems to attract these monies, this is not the case with India. Despite the fact that semiconductor consumption in India was worth USD $21 billion in 2019 (growing at the rate of 15.1 percent)—with research and development (R&D) generating about USD $2.5 billion in revenue, an Indian semiconductor fabrication plant that mass produces the Monolithic Integrated Circuit remains elusive. 

Recently, these facts seem to have gathered the government’s attention owing to which on 15th December, 2021, the Government of India undertook the Semicon India program with the key objectives of: setting up greenfield semiconductor and display fabs, developing R&D, and advancing design capabilities. Supplementing this, the Government of India, last year, also announced three schemes, namely; i. Performance Linked Incentive (PLI), ii. Promotion of Electronic Components and Semiconductors (SPECS), and iii. Electronics Manufacturing Clusters 2.0 (EMC2). The total funding under these schemes amounts to USD $10 billion that will be provided over a period of six years, complemented with special incentives worth USD $30 billion. Cumulatively this is expected to ramp up investments to USD $22.5 billion, a number remarked as satisfactory by various scholars.

These steps have fashioned a benign image of the Indian government towards chipmakers. This has resulted in the government receiving proposals worth USD $20.5 billion from five companies. The applications have been received for setting up 28 nm to 65 nm fabs with a projected investment of USD $13.6 billion. They have sought support from the Centre to the tune of USD $5.6 billion. Of these, Next Orbit Ventures (Abu Dhabi) has planned to invest USD $100 million in a semiconductor fab project based in Gujarat; and INVECAS (California) has planned to invest between USD $15-20 million for the setup of design centres in Bengaluru and Hyderabad. In addition, in an attempt to refurbish and reinstate the rather inefficient Semi-conductor Laboratory (SCL) brownfield fab facility (Mohali), the government is now exploring possibilities for a joint venture with a commercial fab partner to modernise it.

Though the schemes are ambitious and imperative, they possess multiple inadequacies showing a clear paucity of evidence-based policy formulation. They fail to capture the reality of the fab market ecosystem in India. For instance, COVID-19 has severely affected many small semiconductor start-ups which are in their incubation stage, setting them behind, technologically, by three to four months. It has also restricted their capacity to fund for access to tools and Electronic Design Automation (EDA). These schemes hold no provision for fiscally aiding these incubation centres. Even though a new initiative – Semiconductor Fabless Accelerator Lab (SFAL) has been instituted to offer fundraising support, business connections and company support services to prospect startups, the negligible financial support for prototype silicon product development and the scanty number of active projects  point to a low level of commitment resulting in policy deficit. 

The selective monetary provisions under PLI, only incentivize companies that already possess Assembly, Testing, Marking and Packaging (ATMP) units in India. In effect, this discourages new entrants. Similarly, the current Rs 3,285 crores outlay for SPECS is also fraught with financial ambiguity as there is no upper limits or categorization on how much Capital expenditure (Capex) reimbursement a “minimum 1 crore investment” can apply for versus a “minimum 1000 crore investment”. Another, concerning instance, is the exclusion of interest of advanced wafer technology foundries from the SPEC annexures.

By far, one of  the most pertinent issues has been one concerning the requisite infrastructure i.e.,  semiconductor grade water, high quality power and availability of specialised chemicals and gases. The Taiwan Semiconductor Manufacturing Company (TSMC) stands out as a peculiar example of  a policy concern to be kept in mind, particularly in terms of water. Greenpeace estimates that TSMC alone uses 4.8% of Taiwan’s electricity annually. This is expected to rise to 7.2% this year as production at new plants begins in TSMC’s home country. What this means for India is a greater refinement in the associated policy field of water policy of the nation.  

All these issues collectively demonstrate a reserved-selective incentive model instead of a hybrid incentive model—which is essentially what the Indian semiconductor industry needs right now. Therefore, if the Indian government wants a favourable output, it must allow room for special incentives. With the background of the existing policy framework, the following might be suggested to give the techno-sector a required boost.

  • Devise and apply low tax rates for fabs.
  • Offer long term sovereign guarantees to both design and manufacturing units.
  • Impose duty on the components imported by ATMPs, to push for domestic production
  • Refinement of the water policy, so that chances for poor ad-hoc policy decisions can be minimised. 
  • Specify Capex reimbursement brackets and broaden the conditions.
  • List of semiconductor wafers (raw materials) should be made flexible, allowing room for the latest technologies.
  • Chip design studios (start-ups) must be incentivized to link with commercial manufacturing units. 
  • Greater insistence on developing More-than-Moore (MtM) technologies as they have lower starting project cost and approximately 45% share in the semiconductor industry.
  • Diversify investment in fabrication plants of different wafer fabs.
  • Encourage the industry to process raw materials for the fab and ATMP.

A fiscally sound policy to aid Semiconductor manufacturing can bring multiple gains. It will help avert the Covid-induced unemployment crisis and provide impetus to industrial growth, lifting people out of unsustainable farm employment. Focus should be concentrated on the concept of ‘Design in India’ by strapping diverse, functional allied sectors such as Med Technovation, EV and Agri Tech with the program. Also, steps to liaise initiatives by Centre and State governments might result in the development of high-tech clusters and faster substantive gains for the industry. Overall, the ambition of the government of setting up at least two greenfield semiconductor fabs and two display fabs seems like a rational starting point.

  1.  nm stands for Nanometer. nm is a unit of measurement for length in a metrics system just like meters, centimeters, etc. It is used to express dimensions on the atomic scale. In technical terms, it is referred to as “process node” and “technology node“.
  2. In simple terms, Moore’s law states that our computers would become more powerful and less costly. MtM refers to technologies that use digital and non-digital functions on the same chip.

About Anmol R Singh and Sirjan Kaur

Anmol R Singh is pursuing his MA in Public Policy and Governance from Centre for Federal Studies, Jamia Hamdard, New Delhi.

Sirjan Kaur works as a researcher at Trayas, an independent regulatory research and policy advisory organisation.

Disclaimer

The views expressed in this article are the author's own.

Leave a Comment

Your email address will not be published. Required fields are marked *

Share via
Copy link
Powered by Social Snap