The New Import Substitution Map: India’s Next Domestic Manufacturing Plays
$112 billion. That was India’s trade deficit with China in 2025-26, a record. Electronics imports alone crossed $116 billion. And in roughly 636 product lines, China now supplies at least 80% of what India buys from abroad.
Those numbers are a warning. They are also a shopping list. Every large, concentrated import bill is a market that a domestic manufacturer could take, if the economics, technology and policy line up. This insight maps where they are lining up now, where they could in three to five years, and where the prize is real but the path is still long.
The Map in 30 Seconds
- Ready now: electronics components and PCBs, solar cells, defence. Policy is live, capacity is being built and demand is already local.
- Building: semiconductor packaging, lithium-ion cells, pharmaceutical ingredients. Large prizes, early plants running, but scale and cost parity are years away.
- Long shots: rare earth magnets and high-end medical devices. Strategically vital, but India is starting close to zero.
Why It Is Back, and Why This Time Is Different
Import substitution has a bad name in India. The old version relied on high tariffs and licences and protected inefficiency for decades. The new version uses different tools:
- Pay for output, not for existence. Production-linked incentives reward units actually made and sold.
- Mandate quality domestic supply where it matters. Solar’s approved-manufacturer list and defence’s positive indigenisation lists create guaranteed local demand.
- Link it to exports. Defence exports hit a record ₹38,424 crore in 2025-26, up 63% in a year (Ministry of Defence). The goal is to compete globally, not just to replace imports.
Our read: the easy layer is done, the hard layers are next. India substituted the final assembly of phones and electronics, but dependence on the inputs has deepened at the same time. The number of electronics product lines where China supplies at least 80% of India’s imports rose from 44 in 2018-19 to 71 in 2025-26 (Koan Advisory and Institute of Chinese Studies). In pharma, China’s share of India’s API and intermediate imports rose from 68.2% to 73.7% between FY21 and FY25, even as incentive schemes drew more investment than targeted.
Each layer further upstream has smaller volumes, harder chemistry and thinner margins for error. That is where the next manufacturing plays sit, and it is why they need patience as much as capital.
The trigger was supply risk, not just cost. When China restricted rare earth magnet shipments in 2025, Indian automakers scrambled for supply (Reuters). That episode moved import substitution from an industrial-policy idea to a board-level risk question.
The Map
We plotted eight industries on two questions: how big is the import prize, and how ready is India to take it?
Rubyworks assessment, September 2026. Positions are judgement-based, drawn from import data, approved capacity and scheme progress cited in this note.
How to read it: the top-right is where capital can be deployed with the least execution risk today. The top-left holds the largest prizes, but they need time, technology partners and sustained policy. The bottom-left is strategically important but still at pilot scale.
Ready Now: The Scale Plays
1. Electronics Components and PCBs
- The prize: electronics imports reached $116 billion in 2025-26 (Commerce Ministry). China’s dominance sits in motors, cables, switching equipment and other core parts, not finished goods.
- What is happening: the Electronics Component Manufacturing Scheme pipeline covers 75 projects with ₹61,671 crore of proposed investment. Once running, approved projects are expected to meet about half of India’s PCB demand.
- The catch: components are high-volume and thin-margin, and Chinese suppliers have a decade of scale. Design depth in India is still limited.
- Our call: the most investable near-term play, because the customers (phone, auto and appliance assemblers) are already in India. Multilayer PCBs, connectors and display modules look strongest.
2. Solar Cells
- The prize: India localised solar modules first but kept importing the cells inside them.
- What is happening: from June 1, 2026, approved solar projects must use domestically made cells. About 30 GW of cell capacity was on the approved list by May 2026.
- The catch: the next layers, wafers and polysilicon, are still imported. Mandated demand also invites overbuilding once the initial rush ends.
- Our call: cells are a play now. Wafers and polysilicon are the next, larger and riskier opportunity, and whoever integrates backward first will set the cost curve.
3. Defence
- The prize: defence production reached a record ₹1.78 lakh crore in 2025-26, up 15.6% in a year, against a government target of ₹3 lakh crore by 2029 (Ministry of Defence).
- What is happening: ten positive indigenisation lists now name 5,521 items for domestic sourcing, and the private sector’s share of production has reached a record 24%.
- The catch: orders are large but lumpy, cycles are long and state-owned firms still dominate production.
- Our call: the demand is the most secure on the map. The underappreciated opportunity is in sub-systems and component suppliers to the big platforms, where the indigenisation lists bite first.
Building: Big Prizes, Early Plants
4. Semiconductors
- Where it stands: twelve units approved with over ₹1.64 lakh crore of investment; Micron, CG Semi and Kaynes are in commercial production, all in assembly and testing. The first silicon fab is scheduled for 2028. China still supplied 48.9% of India’s semiconductor imports in 2025-26.
- The catch: packaging is the easier half; front-end fabrication needs process know-how India does not yet have at scale.
- Our call: the near-term play is packaging, testing and the equipment and materials suppliers that cluster around them, not the fab itself.
5. Lithium-Ion Cells
- Where it stands: imports from China more than doubled since 2021-22 to $3.9 billion, and China’s share reached 83.6%. Only 1.4 GWh of the 50 GWh targeted under the battery cell incentive scheme was commissioned by October 2025 (IEEFA).
- The catch: the scheme’s slow start shows how hard cell chemistry, equipment and mineral supply are to localise at once.
- Our call: cells for electronics and energy storage will localise before EV-grade cells. Recycling, as a domestic source of lithium and cobalt, is an underrated adjacent play.
6. Pharmaceutical Ingredients (APIs and KSMs)
- Where it stands: 39 bulk-drug projects covering 28 APIs are commissioned, including Penicillin G and clavulanic acid, and 191 of 726 targeted APIs, starting materials and intermediates are now made in India for the first time. Yet China’s share of these imports still rose to 73.7% in FY25.
- The catch: China’s fermentation scale sets prices India cannot yet match for high-volume antibiotics.
- Our call: the winners will be molecules where buyers pay a premium for supply security, and makers who pair domestic capacity with export contract manufacturing.
Long Shots: Strategic, But Starting From Near Zero
7. Rare Earth Magnets
- Where it stands: a ₹7,280 crore scheme targets 6,000 tonnes a year of capacity and drew 20 bids. India holds the world’s third-largest rare earth reserves, yet imports almost all its magnets, and China controls about 90% of processing.
- Our call: small tonnage but outsized strategic value. Automakers and defence buyers may pay a premium for secure supply, which could make the economics work sooner than volumes suggest.
8. High-End Medical Devices
- Where it stands: India’s device market is 70% to 80% import-dependent by value, a figure that has barely moved in a decade, with imaging and implants the most import-heavy. Consumables are already competitive.
- Our call: substitution will climb from consumables into mid-tech devices first. High-end imaging will need global partners, not stand-alone Indian plants.
The Traps: Where Substitution Fails
- Assembly without depth. Making the final product locally can raise imports of the parts inside it. Electronics exports and imports both grew strongly in 2025-26, which shows how easily assembly and dependence rise together.
- The mandate hangover. Guaranteed domestic demand draws a wave of capacity. When the wave peaks, prices fall and weaker players get squeezed. Solar cells are the category to watch.
- One layer’s protection is the next layer’s cost. Duties and mandates that help cell makers raise costs for module makers and power developers. Every substitution decision moves margin along the chain.
- Incentive-dependent economics. If a plant only makes money while incentives last, it is a subsidy, not a business. Investors should test unit economics with incentives stripped out.
- Hidden upstream imports. Equipment, specialty chemicals and minerals for chips, batteries and magnets are still largely imported. A plant can be domestic while its inputs are not.
What to Watch
- Commissioning, not approvals. The number of component, cell and chip projects actually producing, against the number approved.
- The China-concentration count. Whether the 71 electronics lines with 80%-plus China dependence start to fall. That is the cleanest test of real substitution.
- The first domestic magnets. India aims to start producing rare earth magnets by the end of 2026. On-time output would move magnets up the map fast.
- Solar’s next layer. Announcements of wafer and polysilicon capacity will signal who is integrating backward.
- Battery cell ramp-up. Whether commissioned cell capacity moves meaningfully beyond pilot scale in the next 12 months.
- API data for FY26. A falling China share would be the first proof that the bulk-drug scheme is changing the mix, not just adding capacity.
The Bottom Line
India’s first import substitution wave replaced finished goods. The second has to replace what goes inside them, and that is harder, slower and more selective. The near-term money is in electronics components, solar cells and defence supply chains, where demand is local and policy is working. The larger prizes in chips, battery cells and pharmaceutical ingredients will reward investors who can wait three to five years and who judge projects on unit economics, not incentive announcements. The one number to track is simple: how many product lines still depend on a single foreign supplier.
