Editor's Note: The supply bottleneck of AI infrastructure is spreading from GPUs, memory, data centers, and power systems to even more foundational hardware components. Goldman Sachs and Morgan Stanley have now turned their attention to MLCCs—the multi-layer ceramic capacitors long seen as common passive components.
In AI servers, MLCCs are responsible for stabilizing current, filtering noise, and are a key component in ensuring chip high-speed operation. As Nvidia's next-generation rack architecture increases the usage of MLCCs per rack, their value is rapidly rising. Goldman Sachs estimates that the AI server MLCC market will grow over fourfold between 2025 and 2030, while industry capacity is increasing at a slightly higher than 10% annual rate, highlighting a supply-demand mismatch that is becoming a core variable in this cycle.
Moreso, the price cycle has already begun. Japanese industry leaders like Murata and TDK have initiated price hikes, and Japanese export data is starting to validate demand strength. For the capital markets, the logic around MLCCs is not complicated: demand is coming from AI servers and high-end automobiles, supply expansion is limited, and price increases can significantly enhance profit margins.
From chips to capacitors, the pricing power of the AI supply chain is trickling down to more granular and obscure links. Whether MLCCs will become the "next memory chip" still depends on whether AI server demand can continue to materialize; however, what is certain is that this once overlooked foundational component has now stepped into a new cycle of synchronized rising prices and demand.
The following is the original text:
The supply bottleneck in the artificial intelligence (AI) arms race is sequentially igniting opportunities across various hardware sectors. Following data centers, energy infrastructure, and storage chips becoming the focal points of capital attention, Wall Street giants Goldman Sachs and Morgan Stanley have simultaneously turned their focus to a long-underestimated foundational component: multi-layer ceramic capacitors (MLCCs). The two institutions predict that MLCCs will become the next key battleground for "synchronized rising prices and demand," and this AI-driven growth cycle may be the largest in history.
Goldman Sachs analyst Daiki Takayama pointed out in a report that the AI server MLCC market size is expected to surge from about ¥215 billion (approximately $1.4 billion) in the fiscal year 2025 to about ¥920 billion (approximately $5.8 billion) in the fiscal year 2030, more than quadrupling with a compound annual growth rate of 34%. Goldman Sachs stated that the current AI-driven MLCC cycle "will be the largest and longest-lasting in history, and we believe it is still in its early stages."
MLCC: The "Invisible Heart" Maintaining AI Server Operation
An MLCC (Multi-layer Ceramic Capacitor) can be understood as an extremely miniature, ultra-fast charging and discharging unit. Unlike regular batteries that store a large amount of energy and release it slowly, an MLCC stores minimal energy but can charge and discharge within extremely short milliseconds or even shorter. Its core function is to smooth power fluctuations and filter noise: absorbing sudden voltage spikes, rapidly replenishing current during voltage drops, providing stable power to sensitive chips, and blocking electrical interference that could disrupt digital signals.
The operational characteristics of AI servers make MLCCs indispensable. When AI models perform massive computations, the processor's power demand can spike within microseconds and then quickly drop to near zero after the computation is complete. The power system itself struggles to respond promptly to such intense fluctuations. MLCCs are typically installed directly near AI chips and release energy instantly when power spikes occur, preventing server crashes. Since AI chips like Nvidia GPUs need to process billions of tasks simultaneously, a top-tier AI server rack may require up to 600,000 MLCCs working together to maintain system stability.
Goldman Sachs analyst Nelson Armbrust further pointed out that MLCCs have become the third most expensive component in the AI server bill of materials (BOM), only after the GPU and memory. The current overall MLCC market size is around $15 billion, with the server-related market accounting for approximately $1.3 billion, expanding at an 80% compound annual growth rate. In contrast, demand growth in other application areas such as automotive and smartphones has significantly slowed down. Daiki Takayama expects that the cost of MLCCs in the AI server BOM will gradually increase from the current approximately 0.5% to about 1%.
Structural Supply-Demand Contradiction: Annual Capacity Growth Rate Only 10%, Difficult to Withstand a Fourfold Demand Surge
The key factor igniting market attention is the severe structural supply-demand imbalance the MLCC industry is facing. Goldman Sachs analyst Allen Chang clearly pointed out that the annual production capacity growth rate of the entire MLCC industry is only slightly above 10%. Furthermore, due to heavy reliance on internal production of equipment and materials by manufacturers, the expansion progress is limited by internal engineering resources, making it difficult to significantly accelerate. However, the demand impact from AI servers is not at the same level. Goldman Sachs projects that from the 2025 fiscal year to the 2030 fiscal year, the MLCC demand driven by AI servers will grow by approximately 4.3 times.
What is even more concerning to the market is that the strong demand for high voltage, high capacitance MLCCs driven by automotive electrification is still robust, with the per-vehicle MLCC usage continuing to increase. The two major demand pillars of AI servers and electric vehicles are jointly consuming the already limited additional production capacity. This has also led to customers actively seeking long-term supply agreements even as consumer electronics demand declines, to hedge against future shortage risks.
Signals of the current tight market have appeared on multiple fronts: delivery lead times for high-end MLCCs (high-capacity, high voltage specifications) have exceeded 20 weeks; low-capacity and consumer-grade MLCCs are affected by hoarding and duplicate orders, leading to a 20% to 40% price increase in spot and distribution channel prices; key raw material prices such as nickel and silver remain high, putting pressure on the cost of various products.
Price Hike Cycle Officially Begins: Japanese Giants First to Raise Prices, Official Data Confirms Trend
Price signals are rapidly strengthening. The price hike actions of Japan's two leading companies, Murata Manufacturing and Taiyo Yuden, mark the official start of the MLCC price hike cycle. Murata, starting from April 1 this year, will raise the prices of MLCC products in AI servers and high-end automotive applications by 15% to 35%. Taiyo Yuden has also notified customers that it will adjust prices for multiple product lines starting from May, involving MLCCs, inductors, RF devices, FBAR/SAW devices, and aluminum electrolytic capacitors, citing continuous increases in costs of precious metals and other raw materials.
Trade statistics data released by the Japanese Ministry of Finance on May 28 validated this price hike trend from a macro perspective. The data shows that in April, the average export price of MLCCs increased by 3% month-on-month and 16% year-on-year; export volume increased by 10% year-on-year; and export value surged by 28% year-on-year. Goldman Sachs believes that this data confirms the signals released in the recent financial reports of Japanese MLCC manufacturers: all companies have confirmed that order momentum remains strong.
Looking at the timeline of the entire AI supply chain, Goldman Sachs' analytical framework shows that the price hike of MLCCs significantly lags behind AI core components such as DRAM, NAND memory, ABF substrates, and copper-clad laminates (CCL). Therefore, Goldman Sachs predicts that among all AI components and materials, the price increase space for MLCCs is the longest, and the sustainability is the strongest. Goldman Sachs has revised its 2026 MLCC year-on-year price change forecast from approximately 0% to +5%, emphasizing that the actual future increase may far exceed this level.
Remarkable Profit Elasticity: 5% Price Increase Could Boost Operating Profit by Up to 37%
For investors, the profit elasticity resulting from an MLCC supply-demand mismatch is not to be underestimated. Daiki Takayama estimates that a mere 5% price increase could theoretically drive Murata's FY2027 operating profit up by around 13% and SolarEdge's operating profit by as much as 37%.
Goldman Sachs projects that Murata's FY2027 revenue will reach ¥10.5 trillion (approximately $66 billion), a 13% year-on-year increase, and SolarEdge's revenue will reach ¥286 billion (approximately $1.8 billion), also with a 13% year-on-year growth. Goldman Sachs maintains a "Buy" rating on Murata, SolarEdge, and TDK. Its constructed Asian MLCC theme stock portfolio has recently started to strengthen, but compared to other popular AI themes, there is still significant room for catch-up.
Morgan Stanley Breakdown of Nvidia's New Rack: Peripheral Component Importance Rises, MLCC Usage Surges 182%
Another heavyweight catalyst comes from Nvidia's next-generation Vera Rubin AI rack. Morgan Stanley, after dissecting Nvidia's latest VR200 rack, found that the significance of peripheral components in the latest BOM is rapidly increasing.
The value of MLCCs in a single rack has surged from approximately $1,530 in the previous GB300 era to around $4,320, a remarkable 182% increase. Although the absolute value of MLCCs still lags behind GPUs, memory, and PCBs, its growth rate in peripheral components is outstanding.
Morgan Stanley's channel checks further indicate a significant increase in MLCC usage on compute and switch boards, with compute board showing a more pronounced growth. Additionally, the newly introduced BlueField and ConnectX modules will further increase the total MLCC usage per rack. This partially explains why the current demand for high-end AI server MLCCs is so strong, prompting several ODM manufacturers to actively stock up in preparation for the mass production and delivery of the Rubin rack in the latter half of 2026.
Morgan Stanley's breakdown of the Nvidia Vera Rubin rack reveals the following key component value changes:
Market intelligence indicates that in the AI supercycle's infrastructure arms race, as supply bottlenecks rotate, it has led to one market winner after another. Goldman Sachs' latest assessment describes MLCCs as the "new memory chip"—marking the passive component subindustry that is currently at the inception of a cycle where both volume and price are rising simultaneously.
With the exponential impact of AI server and Nvidia Rubin rack demand, the high-end MLCC lead time has exceeded 20 weeks. The Japanese industry leader has initiated price hikes, official export data remains strong, all signals point to the same conclusion: this AI-driven MLCC supercycle is just getting started.
Welcome to join the official BlockBeats community:
Telegram Subscription Group: https://t.me/theblockbeats
Telegram Discussion Group: https://t.me/BlockBeats_App
Official Twitter Account: https://twitter.com/BlockBeatsAsia
