Why doesn't India make its own multilayer ceramic capacitors (MLCCs)?

An electric vehicle contains 15,000 to 20,000 multilayer ceramic capacitors. An internal-combustion car needs 2,000 to 3,000. India's EV registrations grew from 95,198 units in 2018 to 1,670,736 in 2024 — and not one of those capacitors is confirmed to have been made in India.

The multilayer ceramic capacitor, or MLCC, is the most ubiquitous component in electronics. It stores and releases small amounts of charge, and it appears by the billions in smartphones, EVs, and defence and space systems. Its construction is deceptively simple to state and extraordinarily hard to execute: alternating layers of a ceramic dielectric — usually barium titanate, prized for its very high dielectric constant — are interleaved with metal electrode layers, then stacked and co-fired into a single monolithic chip.

The difficulty lies in the stacking and the sintering. Murata, the market leader, now builds an 800-layer part in a 0402 case and has pushed miniaturisation to the world's smallest 006003 size. Hundreds of ceramic and electrode layers must be laminated, cut, and terminated at micron tolerances and high yield. This demands precision capital equipment and, critically, mastery of the dielectric powder itself: most barium titanate is made captively by the capacitor manufacturers, not bought on the open market. The recipe is the moat.

That moat is why the industry is consolidated among five firms — Murata (Japan), Samsung Electro-Mechanics (South Korea), Yageo (Taiwan), Taiyo Yuden and TDK (both Japan) — commanding 69 to 79 per cent of the market. Murata alone holds over 40 per cent; Samsung Electro-Mechanics around 24 per cent in the first half of 2024. The market is tiered by technology: Japanese firms lead in automotive- and industrial-grade parts; Korean and Taiwanese firms hold a second tier with larger capacity but a technology gap; Chinese makers occupy a third tier of mid-to-low-end parts.

Even China illustrates how steep the climb is. Its manufacturers — Guangdong Fenghua, Chaozhou Three-Circle, Eyang Technology — reached about 10 per cent of global MLCC revenue by the second half of 2024, a rise of just four percentage points since 2019, despite China's vast manufacturing scale.

India stands at zero. No domestic company has been identified making MLCCs at commercial scale. India's own MLCC market, estimated at about $0.39 billion in 2025 and projected to reach $1.01 billion by 2031, is served entirely by foreign suppliers such as Darfon, Holy Stone, Jianghai, Kyocera AVX and Maruwa. This sits inside a larger dependence: 60 to 75 per cent of India's electronic components were imported in 2025, and the electronics import bill crossed $100 billion for the first time in FY2025-26, reaching $116.17 billion.

The gap persists because an MLCC plant is not one capability but a stack of them, and India possesses almost none. Barium titanate dielectric production, multilayer tape-casting and co-firing, and the precision stacking and termination equipment are all absent. Nickel base-metal electrode paste has no domestic source. High-purity precursors, Ni/Sn termination plating, AEC-Q200 qualification and reliability testing infrastructure are only emerging. Without the dielectric recipe and the equipment, the finished chip cannot follow.

What it would take is visible in the policy already in motion. The Union Cabinet approved the ₹22,919 crore ($2.7 billion) Electronics Component Manufacturing Scheme in 2025, explicitly targeting SMD passives — capacitors, inductors, resistors — with sales incentives and capital subsidies. But approved projects so far cover PCBs, lithium-ion cells, connectors and camera and display modules, not a capacitor plant.

Closing the gap means building the layers beneath the chip first: captive dielectric powder, the co-firing process, the stacking equipment, and the qualification path into automotive and industrial markets. Demand is not the constraint; the global market is projected to grow from about $15 billion in 2025 to $21.93 billion by 2030. The recipe is.