The Cabinet just cleared the biggest tech-manufacturing push in India's history — but the distance between assembling a phone and designing its brain is measured in decades, not rupees.
Here is a number that should stop you mid-scroll: ₹1.9 lakh crore. That is roughly the size of Kerala's annual GDP — and the Union Cabinet committed it, in a single Wednesday sitting, to making India a serious player in the global technology manufacturing stack. One scheme will try to take India's phone factories up the value chain. The other will try to put India on the semiconductor map of the world. Both rest on the same uncomfortable insight: India's assembly-line success story is not enough, and the country needs to start owning the intellectual content of the devices it makes, not just the labour that puts them together.
To understand why the government felt the need to spend this much, you first need to appreciate what PLI 1.0 for mobile phones — formally the Production Linked Incentive scheme for Large Scale Electronics Manufacturing, or PLI-LSEM — actually delivered.
The headline numbers are genuinely impressive. Total exports under the scheme crossed ₹6.2 lakh crore by February 2026, which was 27% more than the target of ₹4.87 lakh crore. India is now the world's second-largest mobile phone manufacturer by volume, with 99.2% of mobile phones used in the country manufactured domestically. And the cherry on top: smartphones became India's largest exported product category in 2025, overtaking traditional export items such as diesel fuel and cut diamonds.
Production and exports have scaled dramatically. Mobile production in India roughly doubled from ₹2.14 lakh crore in FY2019-20 to ₹5.5 lakh crore in FY2024-25, while smartphone exports grew from ₹27,000 crore to over ₹2 lakh crore in the same period — a roughly eight-fold increase.
That sounds like a roaring success. So why replace it?
Because there is a caveat buried inside these numbers, and it is a significant one. Domestic value addition — the share of the final product's value that is actually created in India — is estimated at around 18–20%. That means for every ₹100 worth of phone that ships out of India, roughly ₹80 worth of components — displays, chips, camera sensors, passive components — was imported, assembled here, and re-exported. India is, in the bluntest terms, a sophisticated screwdriver factory.
The job creation story is not spotless either. Job creation fell slightly short, with 1.85 lakh jobs generated against the 2 lakh target.
And the Apple problem — more on this in a moment — concentrates risk in a way that makes policymakers nervous.
The new Mobile Phone Manufacturing Scheme (MPMS) succeeds the PLI for Large Scale Electronics Manufacturing, whose tenure ended on March 31, 2026. The successor scheme has three changes that mark a genuine departure from the old model.
First, the incentive structure is tiered, not flat. Under PLI-LSEM, the incentive was essentially a percentage of incremental production. The new scheme creates layers. A baseline incentive of 2.25%–5% on eligible sales is the floor, with an additional incentive of up to 1.5% linked to domestic sourcing of key components and sub-assemblies. And — this is the most significant addition — an extra 3% incentive on eligible sales for design and R&D of the product. In plain language: you get paid more for thinking, not just for assembling.
Second, it explicitly aims to build Indian brands. The official release states that the MPMS "aims at building Indian brands to achieve technological sovereignty, capture large economic value and create Indian patents in design and R&D." India has never produced a globally competitive handset brand, and this is the first policy that formally acknowledges the gap — and puts money behind closing it.
Third, it runs for five years, FY2027 to FY2031, giving manufacturers a long enough horizon to justify deeper investment decisions. The government expects the scheme to generate ₹39 lakh crore worth of mobile phone production and create around 60,000 direct jobs.
Before we get to the bigger chip play, there is a structural risk hiding inside the phone success story.
In 2025, the iPhone became India's single largest export item, with roughly $23 billion worth of devices shipped from Indian factories. Smartphones overtook all other categories to become India's top export for the first time, reaching $30.13 billion across all brands — with Apple accounting for about 76% of that total.
Apple's scale is a gift — it has put India on the global supply chain map and turbocharged export numbers. But a country's top export category being dominated to this degree by one foreign brand's manufacturing decisions is a fragile position. The PLI programme ties government subsidies to export performance, which helped manufacturers offset real structural disadvantages: logistics costs in India remain higher than in China, and the local supplier ecosystem, while growing, is still thinner. Apple is believed to be absorbing some of these disadvantages in exchange for supply chain flexibility and reduced geopolitical exposure. If that calculus changes — say, a US-China trade truce reduces the urgency of diversification — India's export figures could shift dramatically.
Nearly 97% of Apple's Indian exports in the March-to-May quarter went to the US market, up from around 50% a year earlier. That concentration within a concentration — one brand, one destination — is a risk profile worth watching.
The MPMS's push for Indian brands and local design is partly a hedge against this single-point-of-failure risk. Whether a credible Indian handset brand can emerge in a market where Chinese brands own the mid-range and Apple owns the premium is a separate, harder question. But at least the policy now acknowledges the gap.
Now for the bigger number and the more complex ambition.
The Union Cabinet cleared ₹1.27 lakh crore for the second edition of the India Semiconductor Mission. The government expects the new scheme to attract investments of around ₹4 lakh crore and lead to semiconductor production worth ₹2 lakh crore during the scheme period.
This represents a substantial escalation from the ₹76,000 crore allocated under the first phase — nearly 68% larger, and covering a far more ambitious scope.
What ISM 1.0 actually built: Under Semicon 1.0, the government approved 12 manufacturing units involving cumulative investments of more than ₹1.64 lakh crore. These include one silicon fab, one silicon carbide fab, one integrated gallium nitride Micro LED display fab, and nine packaging units catering to sectors such as consumer electronics, automobiles, telecommunications, aerospace and power electronics. Of the approved projects, Micron, Kaynes and CG Semi have already commenced commercial production, while another unit is expected to begin operations in 2026.
That is real infrastructure in the ground — but it is heavily skewed toward the back end of the chip-making process (packaging and testing), not the high-value front end (design and fabrication).
The six pillars of Semicon 2.0: The first pillar focuses on strengthening India's chip design capabilities, building on the work of 105 startups already developing semiconductor chips — supporting the development of intellectual property, chip designs and systems for both strategic and commercial applications.
The second pillar addresses a critical gap by incentivising companies involved in manufacturing and research of semiconductor equipment, materials, chemicals and gases essential for chip production. This is new and significant — ISM 1.0 did not explicitly incentivise the suppliers of chip-making machinery. Without domestic equipment suppliers, every Indian fab will remain perpetually dependent on imports for its most expensive inputs.
Manufacturing expansion forms the third pillar, attracting additional fabrication units including silicon fabs, compound semiconductor facilities, discrete component fabs and display manufacturing plants, with the world's first major fabrication plant scheduled to be commissioned in 2028. The fourth pillar targets advanced Assembly, Testing, Marking and Packaging (ATMP) capabilities, while the fifth emphasises R&D to help India progress beyond current 28–110nm technology nodes through collaboration with leading institutions globally.
Talent development constitutes the sixth pillar, with 315 universities already training students using advanced Electronic Design Automation (EDA) tools — the software used to design chips — and around 68,000 students trained so far.
The scheme has also attracted a notable roster of global players. Applied Materials and AMD have each announced investments of $400 million, while Microchip Technology has committed $300 million. Lam Research has announced an investment of $1.1 billion, and KLA has committed $400 million. ASML, IBL Electron and Merck have signed MoUs with the Tata Group to support semiconductor ecosystem development in India.
The phrase "chip sovereignty" gets thrown around, but it deserves unpacking. A semiconductor chip is not a single product — it is the output of a 500-step manufacturing process involving raw silicon, ultra-pure chemicals, equipment that costs hundreds of crores per machine, and precision engineering that took Taiwan and South Korea decades to master.
India's honest position right now: the country already hosts nearly 20% of the global semiconductor design workforce, giving it a genuine comparative advantage in chip architecture, verification, and IP development. India has the brains. What it lacks is the hands — the fabrication infrastructure — and the materials and machines to run it.
Semiconductor design accounts for a disproportionately large share of value and margin in the chip stack. The Semicon 2.0 framework builds on ISM 1.0, which supported 24 semiconductor design projects and provided 105 startups with access to industry-standard design tools for applications spanning artificial intelligence, telecommunications, drones, satellite communications and IoT devices. Leading with chip design as Pillar 1 is economically rational: it's where India already has engineers, and where the margin per engineer-hour is highest.
With the first fab scheduled to be commissioned in 2028, the world is showing greater confidence in India's semiconductor strategy — but India is still years away from making chips domestically at scale. "Chip sovereignty," then, is less a destination and more a direction of travel. The 2026 decision is about widening that road.
A piece on this much government money would be incomplete without asking where the bodies are buried.
The equipment trap. The dependence on imports for upstream equipment and materials is the biggest structural weakness. Semicon 2.0's Pillar 2 addresses this — but building a domestic equipment industry from scratch while simultaneously building fabs is like learning to bake bread while also trying to build a wheat farm and a flour mill. The sequencing problem is real.
The talent gap at the shop floor. While India has a strong base of chip designers, it lacks enough process engineers, metrology specialists, yield engineers and cleanroom technicians needed for manufacturing. Universities producing design graduates is not the same as having enough people who can run a cleanroom at globally competitive yield rates. The 315 universities training students in EDA tools are building the design pipeline; the manufacturing pipeline needs a separate, harder push.
The node gap. India's semiconductor journey has started with 28nm–110nm as the node, and the focus will now be on developing more advanced nodes in collaboration with leading R&D centres. A 28nm chip is what goes into your car's control systems, your smart meter, your industrial equipment — genuinely useful, large market. But it is not the cutting-edge chip that powers an AI data centre. The world's most advanced fabs are now at 3nm and below, a capability concentrated in TSMC (Taiwan) and Samsung (South Korea). India will not be competing there anytime soon, and should not pretend otherwise.
The cost disadvantage. Manufacturing in India faces structural cost disadvantages compared to global leaders, particularly in assembly and electronic components. Logistics costs in India remain higher than in China, and the local supplier ecosystem, while growing, is still thinner. Incentive schemes can bridge this gap — PLI 1.0 effectively did exactly that for phones. But incentives end; competitiveness needs to become structural before the schemes do.
The strategy is credible. The execution is the exam.
Step back from the individual numbers and what emerges is a coherent, if ambitious, industrial logic. India has already proved it can move from making negligible quantities of phones to exporting ₹2 lakh crore worth of them in a single year. That took less than a decade. The approach for chips is similar: target selective strengths — packaging, mature nodes, chip design — rather than attempting to replicate the entire global semiconductor value chain overnight.
The geopolitical tailwind is real. Strategic partnerships have been established with major semiconductor economies, including the United States (2023 and Pax Silica in 2026), the European Union (2023), Japan (2023), Singapore (2024), the Netherlands (2025) and Germany (2026). In a world where every major economy is trying to reduce its dependence on Taiwan for chips and China for assembly, India is the largest credible alternative on the board.
The government targets a semiconductor market size of $100–110 billion by 2030 — roughly 10% of global semiconductor demand — and expects India to achieve the capability to design and manufacture chips required for nearly 70–75% of domestic applications by 2029. These are ambitious targets, but they are at least internally consistent with the pillars being funded.
The irony worth sitting with: India's first electronics PLI was about getting global companies to assemble here. MPMS 2.0 and Semicon 2.0 are about getting India to own what goes inside. The shift from screwdriver to IP is not a five-year journey — Taiwan started this in the 1970s. But every complex thing begins with a decision, and the decision to try has now been made, at ₹1.9 lakh crore of intent.
The question is no longer whether India wants chip sovereignty. It is whether the factories, the engineers, the chemicals, and the institutional patience will all show up on time.