AI Server Demand Squeezes High-End Capacity, Global MLCC Faces Structural Capacity Reshaping
Claire Weston
Surging AI server builds are pushing high-end MLCC capacity to the limit — lead times exceed 20 weeks, spot prices up as much as 40% — and upstream equipment and materials suppliers stand to benefit first.
What are MLCCs, and why do they suddenly matter to AI?
MLCCs — multilayer ceramic capacitors, the tiny components on circuit boards that stabilize current — make up a global market already worth $15 billion.
This market used to ride on smartphones and consumer electronics. Now AI servers are absorbing capacity fast: the server segment hit $1.3 billion in 2025, with AI servers alone accounting for $600 million.
This means → an industry built around consumer gadgets is being pulled toward high-end compute infrastructure.
How severe is the supply–demand imbalance?
Applications like AI agents are driving CPU demand; general-server MLCC consumption is expected to accelerate by 30%–40%.
At the same time, smartphones and other legacy consumer segments face negative growth in 2026–2027.
In plain terms = the high end is desperately short of parts, while the low end can't move inventory — demand structure is flipping.
The result: high-capacitance, high-voltage MLCCs used in autos and servers now carry lead times beyond 20 weeks, and most suppliers have book-to-bill ratios above 1.
Where exactly are prices rising, and who is feeling the squeeze?
Nickel and silver — core raw materials — are pushing up costs across the MLCC supply chain.
In China's distribution channels, hoarding and double-ordering have driven prices of low-capacitance and consumer-grade parts up 20%–40%.
Contract prices for major OEM customers have not yet jumped sharply, but tight premium capacity is forcing top-tier makers like Murata and Taiyo Yuden to shift production entirely toward AI servers.
This reflects a bifurcated picture: blended average selling prices may stay flat, because consumer-side declines and AI-side premiums roughly cancel out.
Who ultimately captures the capex windfall?
Murata, Taiyo Yuden, and Samsung Electro-Mechanics — the first tier — are expanding capacity at full speed.
This means → the mid-to-low-end share they vacate opens room for second-tier and Chinese domestic makers.
But the real leverage sits further upstream: MLCC production-equipment and rare-materials suppliers are picking up the momentum baton.
Put simply = capacitor makers' stocks have already had their run; next in line, the companies selling the tools and the raw inputs will likely outperform the capacitor makers themselves.
Content is for reference only, not financial advice.