Simulation ICs Welcome Cyclical Recovery, AI and Data Centers Provide Structural Incremental Growth

Overview

Morgan Stanley issued a report stating that the global analog semiconductor industry is approaching cyclical recovery after experiencing a 'L-shaped' bottom for over three years. Unlike previous cycles relying solely on macroeconomic rebound, this recovery benefits not only from channel inventory optimization and price stabilization but also from structural incremental demand brought by AI and data centers. The report believes this is not a simple replay of the shortage cycle of 2020-2023, but rather a selectivity tightening cycle where mature process supply, analog/MCU pricing, and AI-related power demand will provide the best investment opportunities.

Core views

Demand Side & Inventory Cycle: Inventory adjustments in industrial and automotive sectors have basically ended, with customer and distributor inventories at low levels, lead times beginning to extend by 4-8 weeks, supporting recent revenue growth through restocking needs. Meanwhile, demand is widening, no longer limited to single data center or power management, with order trends in industrial, aerospace, and defense also strengthening. Price & Supply Landscape: Price pressure on analog chips and MCUs is easing. As mature process wafer foundries prioritize high-profit AI server orders, 8-inch capacity supply tightens. TrendForce expects global 8-inch capacity to decrease year-over-year by 2.4% in 2026, with utilization rising to 85-90%, providing support for analog chip manufacturers to pass through costs and stabilize prices. Structural Empowerment from AI & Data Centers: AI infrastructure is reshaping the demand logic of analog chips. Taking NVIDIA's latest rack as an example, the evolution of 800V architecture and the introduction of Power Capacitor Systems (PCS) have significantly increased the semiconductor value of power supplies per rack. Additionally, silicon photonics, optical interconnects, and low-earth orbit satellite communication fields provide new growth engines for analog chip manufacturers. This shifts analog chips from purely automotive/industrial cyclic stocks to targets with AI content incremental attributes, potentially raising their mid-term earnings base and valuation multiples.

Analysis framework

The institution adopted a dual-mainline analysis framework of 'Cycle + Structure'. First, by tracking high-frequency data such as channel inventory (DOI), wafer foundry utilization, and spot prices, determine the inventory cycle position where the industry has reached the cycle bottom and opened restocking. Second, by bottom-up BOM (Bill of Materials) breakdown, quantify the increase in power semiconductor value per AI server rack, demonstrating the structural pull of AI on analog chips. Finally, combining business exposure and capacity layout of each company, classify targets into cycle recovery, AI/DC incremental, and high elasticity three investment camps.

Methodology notes

• Cycle & Prosperity FrameworkInventory cycle (Kitchin)

  Tracking industry inventory levels and channel restocking behavior

  Industry inventory levels (e.g., DOI) and channel restocking behavior are important leading indicators for judging turning points in strong cyclic industries like semiconductors. When channel inventory drops to low levels and lead times begin to lengthen, it usually means the industry is about to enter an active restocking phase, driving enterprise revenue and capacity utilization recovery.

• Industry/Industrial Analysis FrameworkVolume-price decomposition

  Volume-price relationship analysis in semiconductor recovery

  When analyzing semiconductor industry recovery, institutions focus not only on demand volume recovery but also closely track price stabilization. When wafer foundry capacity becomes tight causing cost increases, whether design companies can successfully pass costs downstream (i.e., price stabilization or increase) is key to judging if profit margins can be repaired.

• Industry/Industrial Analysis FrameworkUpstream-Midstream-Downstream Chain Transmission

  Impact of upstream capacity allocation on downstream pricing power

  The report analyzes capacity allocation and utilization changes of upstream wafer foundries (such as 8-inch mature processes) to deduce their impact on downstream analog chip/MCU pricing power. Upstream capacity tilting towards high-margin AI orders will indirectly improve the supply-demand pattern of traditional analog chips.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• ADI, NXP, Espressif, ALGM

  Beneficiaries of broad analog recovery, possessing flexible manufacturing matching capabilities and high-end analog/MCU exposure

• IFX (Infineon)

  Core beneficiaries of AI rack power and 800V architecture

  Strengths

  Deep layout in AI data center power supply and SiC fields

  Risks

  Automotive demand fluctuations affecting overall performance

• STM (STMicroelectronics)

  Beneficiary of Optical Interconnect, LEO Satellites, and Margin Recovery

  Strengths

  Leading silicon photonics and optical interconnect technology

• Renesas (Renesas)

  Beneficiary of Data Center Digital Power and Memory Interface

  Strengths

  Significantly expanded data center digital power capacity

• ON (Onsemi)

  High-beta power and 800V architecture target

  Strengths

  Greater earnings elasticity for power/SiC and 800V architecture recovery

Key data

• Global 8-inch Capacity and UtilizationCapacity down 2.4% YoY, utilization 85-90%TrendForce forecasts global 8-inch capacity decline in 2026, utilization significantly improving from 75-80% in 2025
• PMIC Lead Time35-40 WeeksPower management chip lead times extended significantly from 21-26 weeks
• NXP 2Q26 Revenue Guidance$3.45 BillionMedian guidance grew 8.5% QoQ, increased 18% YoY
• AI Rack Power Semiconductor BOM159-191 USD/kWRubin Ultra rack reached 159 USD/kW, Feynman rack reached 191 USD/kW

Impact & implications

The report believes that the addition of AI and data centers does not eliminate the periodicity of analog chips, but can improve earnings visibility at the cycle bottom and assign higher mid-term profit margins and valuation premiums to some selected targets. Investors should focus on companies that benefit from traditional industrial/automotive recovery while having clear product layouts in AI power or optical interconnect fields.

Risks

• Restocking rather than genuine terminal demand recovery (sell-through not catching up)
• Automotive demand fluctuations and industrial production recovery falling short of expectations
• Intensifying competition among Chinese local vendors affecting market share and pricing
• Delay in ramp-up of AI/Data Center related power and Silicon Photonics products, or excessive customer concentration
• Capacity expansion leading to oversupply, suppressing investment returns

What to watch

• Actual terminal sales (sell-through) data after channel restocking
• Latest disclosures and guidance from major analog chip manufacturers regarding Data Center/AI revenue exposure
• Changes in mature process wafer fabrication prices and analog chip manufacturer cost pass-through capabilities