According to a report published on 6th July 2026 by TrendForce, a Taiwanese market research firm, the rapid expansion of artificial intelligence infrastructure is creating severe supply-demand imbalances in the market for high-specification multilayer ceramic capacitors (MLCCs). Book-to-bill ratios at leading Japanese and South Korean MLCC suppliers have climbed to their highest levels since the pandemic, raising the prospect of an acute shortage in the second half of 2026.
AI infrastructure ignites MLCC demand
A book-to-bill ratio above 1.0 signals that incoming orders exceed shipments — a leading indicator of impending supply shortages. TrendForce reports that the ratios at Japanese and South Korean suppliers have now surpassed the peaks recorded during the Covid-19 pandemic, a period when an unprecedented combination of surging electronics demand and logistics disruption caused historic bottlenecks across global semiconductor and component supply chains. Exceeding even those levels has put the industry on high alert.
The primary driver of this demand surge is the rapid build-out of AI servers and high-performance computing (HPC) infrastructure. A single server based on NVIDIA's Blackwell-architecture GPU requires tens of times more MLCCs than a conventional server, owing to the concentration of compact, high-capacitance components in its power-supply and signal-processing circuits, which demand high-frequency, low-loss characteristics. As hyperscalers — Microsoft, Google, Amazon, Meta and others — compete aggressively to expand their data centres, this structural demand shows no sign of abating.
Supply-side vulnerability: why Japan and South Korea matter
The high-specification MLCC market is effectively an oligopoly, dominated by Japan's Murata and TDK alongside South Korea's Samsung Electro-Mechanics. Chinese manufacturers have rapidly gained share in the commodity end of the market, but the technical capabilities required for AI server applications — ultra-compact form factors (0402 and below), ultra-high capacitance (100μF and above), high voltage tolerance, and low equivalent series resistance (ESR) — remain largely the preserve of Japanese and South Korean producers. Even as Chinese rivals close the technology gap, industry consensus holds that stabilising mass-production quality will take considerable time.
The deeper problem is that supply cannot keep pace with demand. MLCC manufacturing is extraordinarily complex, encompassing hundreds of process steps from the synthesis of high-purity barium titanate (BaTiO₃) ceramic powder through multilayer stacking, sintering, and plating. Given that new production facilities require at least 18 to 24 months from investment commitment to operational output, any capacity expansion decided today cannot physically meet demand in the second half of 2026.
Samsung Electro-Mechanics: the scale and quality of the windfall
This market environment is expected to deliver powerful and multifaceted benefits to Samsung Electro-Mechanics. The company is the world's second-largest MLCC manufacturer after Murata, with the component accounting for roughly 40–45% of total revenues. Crucially, the high-specification product portfolio it has built through aggressive investment in recent years aligns precisely with the current AI demand cycle.
First, pricing power is shifting sharply in the company's favour. When book-to-bill ratios rise well above 1.0, suppliers gain control over price-setting. During the pandemic, Murata, TDK, and Samsung Electro-Mechanics all raised prices on premium MLCCs by 30–50% or more. In the current AI-driven cycle, demand for high-specification products is expected to be even more concentrated, suggesting that average selling price (ASP) increases could surpass those of the previous upcycle.
Second, the product mix is improving in ways that structurally lift margins. Samsung Electro-Mechanics earns far higher margins on large-format MLCCs for IT and industrial applications than on the small-format components used in smartphones. MLCCs destined for AI servers can carry unit prices five to ten times higher than their smartphone counterparts. As demand tilts towards AI servers, the company's operating-margin profile should improve structurally.
Third, there is an internal synergy with Samsung Electronics' own expansion in AI semiconductors and servers. As Samsung Electronics scales up its AI chip and server businesses, Samsung Electro-Mechanics stands to secure a steady volume of captive orders while simultaneously diversifying its external customer base among global hyperscalers.
Risks and a balanced view
Unqualified optimism, however, is not warranted. The most significant risk is a repeat of the inventory correction that followed the pandemic boom. When supply-chain participants over-stocked during the 2021–22 component frenzy, a severe destocking cycle followed in 2023. Should AI demand mature faster than expected, or should hyperscalers moderate their data-centre investment programmes, a similar pattern could recur.
Murata's own capacity expansion is another variable. The Japanese leader committed substantial capital expenditure to expanding high-specification MLCC output in 2024 and 2025. If that new supply hits the market simultaneously, some of the price gains now being accumulated could be surrendered.
Over the medium to longer term, the pace of the Chinese technological catch-up warrants scrutiny. Taiwan-based Yageo's acquisition of KEMET, as well as the investments of producers such as Chaozhou Three-Circle, indicate that manufacturers in the Chinese-speaking world are moving up the value chain. Sustained government support for domestic component sourcing could alter the competitive landscape within roughly five years.
Parallels with the pandemic shortage
Looking back at the 2020–21 MLCC shortage, the book-to-bill ratio again served as the key leading indicator, and a seller's market persisted for approximately 12 to 18 months thereafter. Murata and TDK each saw operating margins rise by five to eight percentage points year-on-year during that period, and Samsung Electro-Mechanics posted record profits in its MLCC division. The critical difference this time is that the demand base has shifted from smartphones and PCs to AI servers — a market characterised by higher unit prices and more attractive margin structures.
Also worth noting is TDK's strategic push to develop integrated solutions combining power-conversion components and MLCCs for AI servers. Samsung Electro-Mechanics is pursuing a similar transition, moving from being a commodity parts supplier to a solutions partner involved from the power-circuit design stage — a model that raises customer switching costs and enhances long-term revenue stability.
Outlook and implications
TrendForce's warning is not simply about a transient supply-demand mismatch. It illustrates the broader disruption that structural AI-driven demand is inflicting on electronic component supply chains. If shortages materialise in the second half of 2026, delivery delays for completed AI servers could follow, potentially disrupting the data-centre build-out timelines of the world's largest technology companies.
For Samsung Electro-Mechanics, this cycle represents a rare opportunity to accelerate both profitability recovery and the shift towards a premium product portfolio. Yet the full strategic benefit will only be realised if the company uses the upcycle not merely to maximise short-term earnings, but to lock in long-term supply agreements with AI server customers, invest in next-generation MLCC technologies — including high-temperature, high-capacitance products and integrated power inductors — and strengthen cost competitiveness through greater in-house control of key materials.
The global electronic components supply chain is being reshaped by AI, a demand engine unlike any that preceded it. Samsung Electro-Mechanics, alongside Murata, has the credentials to emerge as the primary beneficiary of this transformation. Whether it does so will ultimately depend on the speed and discipline with which it expands its technological leadership and manufacturing capacity.