Nan Ya and Samsung Electro-Mechanics Push the CCL MLCC Price Cycle Higher

Nan Ya Plastics and Samsung Electro-Mechanics entered September with new pricing pressure across the CCL MLCC supply chain, despite uneven consumer electronics demand.

The increases are not isolated adjustments. Copper-clad laminate suppliers, passive-component manufacturers, and chipmakers have all moved prices during 2026. Several September changes now bring those pressures closer to device manufacturers.

Nan Ya reportedly raised prices for copper-clad laminates and prepregs from September 1. Samsung Electro-Mechanics is also raising fourth-quarter quotes for multilayer ceramic capacitors sold directly to original equipment and design manufacturers.

Qualcomm, Maxscend Microelectronics, Analog Devices, Taiyo Yuden, and Nationz Technologies have appeared in separate reports about current or planned adjustments. The original price increase overview, published September 1, grouped these developments into a new industrywide wave.

The common explanation is higher input costs. However, that answer misses the more important mechanism.

AI infrastructure is competing with mainstream electronics for advanced materials, production lines, and working capital. Suppliers are prioritizing components with demanding specifications and better returns. That shift can restrict ordinary products even when phone and PC demand remains uncertain.

This creates the central conflict. AI server investment supports higher component prices, while consumer electronics manufacturers still face buyers who resist higher device prices.

The next few months will show whether suppliers have achieved lasting pricing power or only triggered another inventory cycle.

The September increases reach three layers of electronics

This price wave matters because it touches chips, circuit-board materials, and passive components at the same time.

CCL, or copper-clad laminate, combines copper foil with a resin and glass-fiber insulating structure. PCB manufacturers process that material into the boards connecting chips, memory, power systems, and other components.

Prepreg is a related resin-coated reinforcement material used to bond layers inside a multilayer board. A price increase in either product raises the starting cost for board manufacturers before fabrication begins.

Nan Ya Plastics reportedly notified customers that selected CCL and prepreg prices would increase from September 1. Published accounts placed the adjustment for certain substrate and base materials around 20 percent, although product coverage can vary by contract.

That followed several earlier increases from large Japanese and Chinese suppliers. In April, Panasonic Industry announced revisions effective May 1. Its circuit-board materials notice listed a 30 percent increase for standard glass-epoxy CCL.

Panasonic also listed a 20 percent increase for standard prepreg. Its low-transmission-loss laminates rose 20 percent, while matching prepregs rose 15 percent.

Those figures matter because low-loss materials sit closer to high-speed networking and server applications. Signals moving through advanced switches and accelerators require materials that limit electrical loss and preserve signal integrity.

Resonac had already announced a 30 percent increase across its copper-clad laminates and prepregs for shipments beginning March 1. It cited constrained copper foil and glass cloth alongside labor and transportation costs.

The passive-component layer is moving as well. An MLCC, or multilayer ceramic capacitor, stores and releases small amounts of electrical energy throughout an electronic system.

A single device can contain many MLCCs for filtering, decoupling, voltage stabilization, and noise control. Their individual size hides their importance to the finished product.

Samsung Electro-Mechanics first raised certain distributor-facing MLCC prices for August shipments, according to industry reporting. It then moved toward higher fourth-quarter quotes for direct OEM and ODM customers.

TrendForce estimates that consumer-grade X5R products face average increases of 25 to 30 percent. AI-oriented X6S products reportedly face negotiated increases averaging 10 to 20 percent.

X5R and X6S identify temperature characteristics for ceramic capacitors. The labels help buyers choose components whose capacitance remains within specified limits across operating temperatures.

Chip pricing completes the picture. Qualcomm reportedly informed customers of double-digit increases for products shipped after September 1. Maxscend also announced September adjustments for certain radio-frequency chip products.

Nationz Technologies has confirmed another adjustment for selected products from October 1, according to supply-chain reporting. Analog-chip suppliers have also cited rising manufacturing and operating costs.

These notices do not prove that every electronic product will become proportionally more expensive. Component contracts, inventories, product designs, and supplier negotiations can delay or absorb part of the increase.

They do show that the pressure is no longer confined to one commodity. It now reaches several essential layers of a finished device.

Why the CCL MLCC squeeze is happening now

AI demand is not simply consuming more components; it is redirecting scarce production toward higher-specification products.

The conventional explanation begins with copper, glass cloth, and resin. Those materials form the physical foundation of a copper-clad laminate.

TrendForce estimates that copper foil represents roughly 42 percent of CCL costs. Resin contributes about 26 percent, while glass-fiber cloth contributes around 19 percent.

A rise in any of those inputs can narrow a laminate supplier’s margin. Simultaneous increases make repeated price revisions more likely.

Yet raw-material inflation alone does not explain the pattern across premium and mainstream products. The deeper issue is how AI systems change the mix of demand.

Advanced accelerators exchange enormous amounts of data with memory, networking chips, and other processors. Maintaining signal quality requires multilayer PCBs built with lower-loss laminates and specialized glass cloth.

These systems also need dense power-delivery networks. High-performance processors draw large and rapidly changing electrical loads, increasing the need for suitable capacitors near critical circuits.

Consequently, suppliers have strong reasons to direct equipment, engineering attention, and qualified materials toward AI products. Those orders often carry higher values than standard consumer components.

The result is a capacity squeeze with two dimensions. Premium products face direct demand growth, while mainstream products lose part of the capacity previously available to them.

TrendForce reported that Japanese and Korean MLCC manufacturers were shifting capacity from consumer-grade X5R products toward X6S and X7R parts for AI applications. Its MLCC market update described rising orders and tighter mainstream supply.

By late June, book-to-bill ratios reached 1.30 for Murata, 1.31 for Samsung Electro-Mechanics, and 1.25 for Taiyo Yuden. A ratio above one means orders exceeded recorded billings during the measured period.

TrendForce said those readings were the companies’ highest since the pandemic shortage. The overall MLCC industry ratio reached 1.04.

Shipments from the three suppliers also reached five-year monthly highs in June. Murata shipped 140 billion units, Samsung Electro-Mechanics shipped 98 billion, and Taiyo Yuden shipped 40 billion.

Those numbers support a real increase in activity. They do not reveal how much demand came from final consumption rather than inventory accumulation.

CCL has experienced a related shift. High-end AI substrates use premium glass cloth and low-loss materials that cannot be replaced immediately with ordinary alternatives.

Qualification adds another constraint. Server and networking customers test materials for electrical behavior, reliability, thermal performance, and manufacturing consistency before approving them.

A laminate producer cannot add qualified capacity simply by purchasing generic material. New formulations and production sites must satisfy customer requirements before entering critical designs.

TrendForce’s CCL supply analysis reported that Korean import prices reached $20,728 per metric ton in March. That was 74.5 percent above the reported year-earlier figure of $11,880.

The same report described a Korean PCB manufacturer placing advance orders worth 10 billion won with two Taiwanese suppliers. The order exceeded five times its normal monthly usage.

Advance buying protects production schedules, but it also amplifies apparent demand. Other customers see longer lead times and place their own defensive orders.

This is how a physical shortage can combine with an inventory cycle. AI creates the initial pull, while fear of shortages spreads pressure across the wider market.

AI infrastructure is pressuring consumer electronics

The primary contest is between AI infrastructure demand and consumer manufacturers that cannot easily pass higher costs to buyers.

Data-center operators are ordering systems designed around performance, power density, and deployment schedules. Component cost matters, but delays can be more expensive than paying a higher unit price.

A smartphone or basic laptop operates under different economics. Manufacturers compete within narrow retail price bands and must justify every increase to cost-conscious customers.

That distinction gives AI buyers greater influence over production allocation. It also explains why the supply chain can feel tight while traditional electronics demand remains mixed.

Samsung Electro-Mechanics offers the clearest example. TrendForce’s August research says the company raised quotes directly for OEM and ODM customers, not only distributors.

Direct pricing changes are more consequential than channel markups. They reach the companies planning production volumes and negotiating complete bills of materials.

The projected increase for consumer-grade X5R components exceeds the reported average range for AI-oriented X6S products. That initially appears backward.

However, the difference reflects capacity allocation. Higher-end AI products already command stronger pricing and receive investment priority.

Raising X5R prices can discourage excessive orders, recover margins, and push customers toward longer commitments. It can also compensate for capacity transferred away from consumer lines.

TrendForce characterized Samsung’s action as the start of an industry upcycle in its August MLCC review. The firm also identified Japanese suppliers’ response as a critical next signal.

The CCL market shows a similar split. AI servers need advanced laminates, but shortages in copper foil and glass cloth affect standard materials too.

When suppliers prioritize low-loss products, ordinary FR-4 buyers compete for the remaining inputs. FR-4 is a widely used flame-resistant glass-epoxy laminate for printed circuit boards.

General electronics companies therefore face pressure from both directions. Their standard materials cost more, while premium designs require even more expensive qualified inputs.

Chipmakers add another layer. Qualcomm’s reported adjustment would raise the processor cost for numerous Android manufacturers.

A chipset increase does not translate directly into an equal retail increase. The processor represents only one part of a phone’s total bill of materials.

Manufacturers can absorb part of the change, reduce margins, revise configurations, or negotiate other components. They can also emphasize older chips in more models.

Those responses all involve tradeoffs. Cutting memory, camera hardware, displays, or charging systems can weaken a product’s competitive position.

The timing is particularly difficult because AI features are increasing hardware requirements inside consumer devices. On-device models benefit from faster processors, more memory, and improved power management.

Manufacturers are being asked to add AI capabilities while several supporting components become more expensive. Marketing expectations and procurement realities are moving in opposite directions.

Automotive manufacturers face another version of the same problem. Vehicles require durable capacitors and chips qualified for long service lives and wide temperature ranges.

Changing a component can require lengthy validation. That makes short-term supplier substitution harder than it is for some consumer products.

Industrial buyers also value continuity over the lowest immediate cost. A missing capacitor or analog chip can stop production of a much more valuable system.

This gives component suppliers leverage, but only while customers believe availability is genuinely constrained. Persistent increases can eventually encourage redesigns, second sourcing, and slower ordering.

The pressure therefore moves downstream in stages. Procurement teams feel it first, product planners reconsider configurations next, and consumers see it last.

Higher quotes do not guarantee a durable shortage

The biggest uncertainty is whether end demand supports these prices after customers finish building defensive inventory.

The supply data point toward tightness. They do not establish that every reported increase will hold across every customer and specification.

Published price notices usually state headline adjustments. Actual transaction prices depend on order volumes, contract timing, delivery commitments, and customer relationships.

A buyer with a long-term agreement may experience a smaller immediate change. A distributor buying scarce spot inventory can face a much larger increase.

Product categories also differ sharply. A low-loss laminate qualified for a new accelerator platform does not share the same market as commodity board material.

The same distinction applies to MLCCs. Size, capacitance, voltage, temperature behavior, reliability requirements, and certification determine whether components are interchangeable.

A shortage in one high-capacitance specification cannot be generalized to every ceramic capacitor. Broad statements about “MLCC prices” can hide large differences within the category.

Inventory behavior adds further uncertainty. Rising book-to-bill ratios can reflect real consumption, stockpiling, or both.

The advance CCL order described by TrendForce illustrates the problem. Buying five times normal monthly usage sends a strong scarcity signal, but that material eventually enters inventory.

Once customers secure enough stock, order growth can slow abruptly. Suppliers then discover whether final demand supports the expanded pricing.

This pattern appeared during earlier electronics shortages. Long lead times encouraged duplicate orders across several suppliers.

When supply recovered, excess inventory produced cancellations and falling prices. The 2026 market does not have to repeat that cycle, but procurement behavior deserves scrutiny.

Traditional consumer demand remains another constraint. Inflation and higher financing costs can weaken purchases of phones, PCs, televisions, and appliances.

Suppliers cannot redirect every factory toward AI. They still need volume products to keep large production networks efficient.

If device sales disappoint, consumer-grade component demand can soften even while AI products remain tight. The market would then divide rather than rise uniformly.

Capacity expansion will also affect the outcome, although not immediately. CCL and glass-cloth producers are investing in new facilities and advanced products.

Some announced projects will not begin production for years. Others must pass customer qualification before they relieve pressure on premium materials.

China’s Guangyuan New Material has announced major investment in high-performance glass materials. Its first planned production phase is scheduled for 2028, according to T-glass expansion reporting.

That schedule shows why suppliers currently retain leverage. Demand can shift faster than specialized capacity can be constructed and approved.

However, high prices create their own correction mechanism. They encourage new entrants, material substitutions, design optimization, and more disciplined purchasing.

Customers can also qualify additional suppliers. That work takes time, but it reduces dependence on any single producer during the next design cycle.

Another risk lies in the AI investment cycle itself. Component demand assumes that cloud providers continue building accelerators, networking systems, and supporting power infrastructure.

A slower deployment schedule would reach upstream suppliers quickly. Materials ordered for planned systems can become excess inventory if server assembly targets fall.

The safest conclusion is narrower than the most bullish supply-chain narratives. Selected CCL and MLCC products are genuinely tight, and several suppliers have exercised pricing power.

It remains unclear whether that power extends across the full market. Reported list increases should not be mistaken for verified average transaction prices.

The price mechanism reaches products unevenly

Consumers will not see one universal electronics surcharge because manufacturers have several ways to absorb or redirect the pressure.

A finished device contains many cost layers. Semiconductors, memory, displays, batteries, circuit boards, passive components, enclosures, assembly, logistics, and software all contribute.

CCL and MLCC increases enter that system through different routes. Laminate changes affect the board itself, while capacitor changes affect numerous positions across that board.

For a simple device, the absolute increase from one capacitor can remain tiny. The effect becomes more meaningful when thousands of parts move together.

Complexity also matters. A server board with demanding signal and power requirements uses different materials from a low-cost household controller.

Premium AI systems are most exposed to constrained high-end inputs. Their buyers are also more likely to accept higher component costs to preserve deployment schedules.

Smartphone manufacturers have more flexibility. They can spread changes across product generations, adjust storage configurations, or reuse previous processors.

They can also negotiate lower prices elsewhere. Large brands routinely balance one supplier increase against savings from another category.

Smaller manufacturers have less bargaining power and lower purchasing volumes. They may encounter the price wave earlier through distributors.

Prototype builders and hardware startups face a related disadvantage. They rarely hold enough volume to negotiate priority allocation or customized agreements.

For those buyers, availability can matter more than the published increase. A component with a modest official revision can cost far more through a constrained spot channel.

Product redesign is another possible response. Engineers can substitute an approved component, change board layouts, or reduce the number of unique parts.

Each change creates engineering and validation costs. Redesigning around short-lived market conditions can also delay a product beyond the shortage itself.

Automotive and industrial programs have fewer quick options because qualification requirements are stricter. Manufacturers may accept higher prices rather than restart testing.

This uneven exposure explains why headlines about component inflation should not become simple predictions about retail prices. The transmission mechanism depends on the product.

A high-end Android phone using a costly chipset, advanced board, and large memory configuration faces several simultaneous pressures. Its manufacturer may adjust the launch price or product mix.

A mature industrial controller might use older components with stable contracts. It could experience little immediate impact unless a specific part becomes scarce.

AI servers occupy the other extreme. They combine premium processors, advanced memory, high-speed networking, sophisticated boards, and large passive-component counts.

Their demand is shaping allocation decisions far beyond the data center. The spillover reaches consumer components when suppliers move capacity toward higher-value specifications.

Procurement teams should therefore track exact part numbers rather than broad component categories. They should separate contractual prices from distributor quotes and reported list increases.

They should also distinguish delivery risk from cost risk. A higher component price can be manageable, while a missing qualified part can stop an entire launch.

For product leaders, the practical question is not whether all electronics will become more expensive. It is where their own design intersects with constrained materials and shared capacity.

Three signals will decide what happens next

Japanese MLCC pricing, fourth-quarter order conversion, and downstream product decisions will determine whether the cycle strengthens or fades.

The first signal is whether Murata and Taiyo Yuden broadly follow Samsung Electro-Mechanics in direct OEM and ODM negotiations.

Taiyo Yuden has already appeared in reports about selected increases. The larger question is whether Japanese suppliers adopt comparable terms across meaningful product volumes.

A broad response would strengthen the case for an industrywide MLCC upcycle. Continued restraint would suggest that Samsung’s pricing reflects its particular product and customer mix.

The second signal is fourth-quarter order conversion. Book-to-bill ratios and distributor activity should translate into sustained shipments rather than cancellations.

Lead times, capacity utilization, and inventory levels will matter more than announcement headlines. Rising shipments alongside low inventory would support the shortage argument.

Falling orders after a stocking rush would weaken it. That pattern would indicate that defensive purchasing exaggerated underlying consumption.

The third signal is how device and board manufacturers respond. New phone configurations, procurement comments, and PCB contract prices will reveal how much cost moves downstream.

Suppliers currently argue that higher raw-material, manufacturing, and logistics costs require price revisions. Their customers must decide whether to absorb, redesign, delay, or pass them onward.

No single response will dominate every market. AI servers, smartphones, vehicles, and industrial systems operate under different margins and qualification requirements.

The most important development is therefore the allocation contest behind the notices. AI hardware has become large enough to influence materials and components used across the electronics economy.

That does not make every increase permanent. It means ordinary device manufacturers can no longer treat AI demand as a separate supply chain.

The CCL MLCC cycle will remain credible while premium capacity stays constrained, mainstream inventories decline, and customer orders convert into shipments.

It will weaken if consumer demand stalls, duplicate orders unwind, or new qualified capacity arrives faster than expected.

For buyers, the next step is concrete. Review exposed specifications, compare contracted and spot quotes, and track lead times before committing to unnecessary inventory.

For investors and industry watchers, look beyond price letters. Watch Japanese supplier behavior, fourth-quarter shipment data, and the specifications appearing in new products.

Those signals will answer the real question: Is the ccl mlcc price wave a durable consequence of AI infrastructure, or another shortage amplified by fear?