India began its semiconductor journey in the 1970s. How advanced is indigenous manufacturing now?
With India’s recent objection to the American CHIPS Act at the World Trade Organisation, the focus is back on the economic, technological and geopolitical significance of the semiconductor industry.
The US law, ostensibly aimed at tackling the China “threat”, was criticised by the Indian government for restricting equal access to technology. And it wasn’t the first such episode hinting at the geopolitical vitality of semiconductors: the US had earlier cut off chip supplies to Russia and subsequently banned the shipment of Nvidia and AMD products to China.
Technological competition simmers as semiconductors increasingly become the backbone of the modern economy, with a core function in electronic devices and a substantial role in emerging technologies. But how is India, which began its semiconductor journey in the 1970s, placed in this power matrix? How advanced is indigenous manufacturing? And what are the government’s policies to help refine the country’s technology?
Before we explore such questions, let’s take a brief look at chips and their various applications.
Which semiconductor is better?
An iPhone 6 costing over Rs 60,000 in 2014 was 120 million times faster than the Rs 29 crore computer used for the Apollo moon landing of 1969. That’s the magic of semiconductors; doubling the number of components that can fit on a chip while also driving down costs every two years.
Semiconductors are materials that have electrical conductivity between that of a conductor (such as a copper wire) and an insulator (like a rubber glove). This unique property makes them ideal for use in electronic devices. They are typically made of silicon, although other materials like germanium and gallium arsenide can also be used.
Inside all electronic equipment, there is a board amassing a bunch of other components, which can be broadly classified as active and passive. The active components are inside that package and are integrated circuits, or ICs – the brains of the device and also known as chips. They are made from a thin slice or wafer of semiconductor material, usually silicon, that is cut into a small, flat piece. And the capabilities of a chip depend on how many components can be integrated into it.
Semiconductor technology is measured in generations, called process nodes, which are defined by the minimum feature size – or the smallest physical size possible for structures on a wafer – in nanometers. Popular process nodes include 40nm, 28nm, 16nm, 5nm, and 3nm. As the feature size decreases, the scope rises for the number of components as well as performance.
Why are they critical for India?
The focus on the semiconductor industry can be attributed to a recent global shortage which has impacted several industries, including automotive, consumer electronics, and telecommunications.
But there are several other factors to explain its significance beyond this demand-supply binary. Let’s look at what makes it critical for India.
Firstly, the country’s economic growth. India currently imports about 80 percent of its required semiconductors, and the failure to develop a domestic semiconductor industry could add $95 billion to the country’s import bill and widen its current account deficit.
Creating a robust semiconductor industry in India would offer significant job creation opportunities. The Indian IT industry today contributes $150 billion to the $1 trillion global IT market, and provides around 5.8 million high-quality jobs in the country – around one fifth of the world’s semiconductor design engineers are from India.
The global semiconductor market is projected to reach over $1 trillion by 2030, up from a market size of $590 billion in 2021, according to a McKinsey report. India’s semiconductor market, meanwhile, is forecast to grow from $15 billion in 2020 to $110 billion in 2030, putting it among the world’s fastest-growing semiconductor markets. To put these numbers in perspective, the Indian government’s estimated budget for the sector in 2023 is $550 billion.
Additionally, semiconductors have a key role in the emerging security state, particularly in military equipment, communication systems, and satellites. In contemporary warfare, the availability and sophistication of equipment often play a more crucial role than the sheer number or skill of military personnel.
For example, in an air battle, aircraft radar is used to track targets that are hundreds of kilometres away. (Yes, a pilot aiming at targets is almost passe). In a melee between two aircraft, the one whose radar can efficiently track more targets is likely to win – a strategic advantage made possible by semiconductors.
The emerging technologies in the field of artificial intelligence and quantum computing also heavily use semiconductors.
While AI systems rely on semiconductor-based components such as central processing units and graphics processing units to perform complex calculations, building functional quantum computers requires precise control over quantum states, which is only possible using semiconductor-based devices.
If India fails to develop advanced semiconductor design and manufacturing capability, the country may have to depend on semiconductor leaders to harness such technologies to their full potential.
How is Indian chip manufacturing positioned?
The electronics industry in India has seen remarkable growth in recent years, fueled by the government's focus on digitalisation, the Make in India initiative, and the rising demand for electronic equipment. The country is now the world's second-largest producer of mobile phones, positioned closely behind China, and a leading manufacturer of other electronic devices like LED televisions, laptops, and tablets.
But when it comes to semiconductor manufacturing, India lacks a robust manufacturing ecosystem and its semiconductor industry primarily focuses on the design and development of products, rather than their production.
Semiconductor design involves creating the blueprint or layout of a semiconductor chip, which includes specifying its functions and characteristics. Design engineers use specialised software tools and simulation techniques to develop the design and verify its performance. The primary goal of semiconductor design is to create a chip that meets the desired specifications and performs its intended functions efficiently.
On the other hand, semiconductor manufacturing involves the physical creation of the semiconductor chip based on the design specifications. This process typically involves several steps, including wafer fabrication, photolithography, etching, deposition, and packaging. Manufacturing engineers use specialised equipment and techniques to transform the design into a physical product. The primary goal of semiconductor manufacturing is to produce a high-quality product that meets the design specifications and is cost-effective for production in large volumes.
The process of packaging the semiconductor ICs is a complex and specialised task in the manufacturing chain, often handled by companies responsible for testing as well. These companies are commonly referred to as OSAT (Outsourced Semiconductor Assembly and Test) or ATMP (Assembly, Testing, Marking, and Packaging) houses.
ICs are manufactured by specialised companies known as semiconductor foundries, using a process called semiconductor fabrication. These foundries enable firms to focus on the design of their products while crafting chips to match their performance needs.
OSAT companies typically take wafers or bare dies (untested semiconductor chips) and package them into final products that can be used in electronic devices.
Designing semiconductors is equally challenging and important as manufacturing them. In fact, due to the complex nature of semiconductor manufacturing, a vast majority – approximately 95 percent – of semiconductor companies operate under the fabless model. This means that they focus exclusively on design, while outsourcing the manufacturing process to specialised fabs.
The rise of new players in the industry remains restricted as a result of the extremely capital-intensive nature of semiconductor manufacturing – which is closely tied to a specific process node – with each node needing additional capital expenditure.
The rapid rollout of new semiconductor technologies every two years, by established players, makes it even more difficult for new entrants to break into the industry. This is evidenced by the fact that TSMC, the top-ranked semiconductor manufacturer, generates more revenue than the combined revenue of the next 10 largest fabs. The other leading players in the foundry business include Samsung, GlobalFoundries, and UMC.
Further, as the global semiconductor industry becomes increasingly critical, countries are not only focusing on refining their own technologies but also restricting access to perceived adversaries. The US, for instance, had previously used the Foreign Direct Product Rule to ban Huawei’s exports to the country and cut off chip supply to Russia.
The CHIPS Act (Creating Helpful Incentives to Produce Semiconductors), a $52 billion semiconductor manufacturing incentive plan proposed by the US government, is another example. It seeks to provide significant financial incentives to chipmakers to encourage domestic production of semiconductors. It aims to reduce American dependence on semiconductor imports, particularly from China, and to ensure a stable and secure supply of semiconductors for national security purposes.
It includes a provision that restricts the export of chips (or any product in the semiconductor supply chain made using American technology) from being exported to China, reportedly, as part of efforts to protect the US intellectual property and national security. This provision has been a source of contention, with India and China asserting that it unfairly restricts access to critical technologies.
How are indigenous semiconductors faring?
India's semiconductor industry journey began in the 1970s with the establishment of Semiconductor Complex Ltd (SCL) in Mohali, Punjab. The state research institute produced silicon wafers, the building blocks of microchips, for strategic purposes. However, with technological advancements, SCL's current 180nm process node is not competitive for commercial applications, though it still caters to the needs of the defence and satellite industries.
Meanwhile, India’s Shakti microprocessor program, which began as an academic initiative in 2014, achieved a significant milestone by delivering the country’s first indigenous microprocessor in 2018. With manufacturing options in 180nm and 22nm, the processor is well-suited for industrial and internet of things applications, and marks an important step towards building a robust processor infrastructure in India.
Is the chip industry getting adequate policy push?
Electronics systems have been identified as one of the 25 key economic sectors for job creation and skill enhancement under the Make in India initiative.
The National Policy on Electronics 2019 aims to promote domestic manufacturing of electronics products, replacing NPE 2012. While the NPE 2019 covers the entire electronics value chain, from manufacturing to design and innovation, the NPE 2012 was limited to electronics manufacturing. The NPE 2019 emphasises the need for research and development (R&D) in electronics and aims to promote innovation and intellectual property creation in the sector.
The latest Semicon India Policy Framework – a policy document first released by the centre in 2007 – provides fiscal support of $10 billion to bolster the semiconductor supply chain. The framework also emphasises the modernisation and commercialisation of the SCL and has tasked the Ministry of Electronics and Information Technology with exploring the possibility of a joint venture with a commercial fab partner to upgrade the brownfield fab facility.
In February 2021, the India Semiconductor Mission was launched by the Ministry of Electronics and Information Technology and the Department of Science and Technology. It is implementing schemes to set up semiconductor and display fabs, offering higher incentives for more advanced process nodes. For process nodes at 28nm or lower, up to 50 percent of the project cost is covered by incentives. For process nodes between 28nm and 45nm, up to 40 percent of the amount is covered, and for nodes between 45nm and 65nm, up to 30 percent of the cost is covered.
Under this mission, compound semiconductors, silicon photonics, sensor (including MEMS) fabs, discrete semiconductor fabs, and semiconductor ATMP/OSAT units are eligible to receive 50 percent of the capital expenditure.
Then there is the design-linked incentive (DLI) scheme, which provides up to 50 percent of eligible expenditure as incentive to 100 domestic semiconductor design companies for Integrated Circuits. The scheme aims to boost the local design capabilities of ICs, the building blocks of all electronic devices, and to encourage innovation and entrepreneurship in the domestic sector.
Despite the Indian government’s impetus to the domestic chip industry, a large reservoir of resources, both human and material, remains untapped in the segment. While India has made leaps in design and development of chips, the industry needs the government’s concentrated efforts to boost the domestic production ecosystem. In view of the growing relevance of semiconductors in technology and its foreseeable compelling role in geopolitics, India’s continued dependency on imports for chips isn’t likely to bear well, and its journey of bridging this gap will be closely watched.
The writer is a semiconductor professional. He currently works for High Speed SerDes IPs at Synopsys Inc, in Mississauga, Canada.
The Science Desk is a collaborative effort between Newslaundry’s subscribers and its editorial team.