Kioxia Holdings (285A): Goldman Sachs reiterates Buy on Kioxia as Kitakami efficiency initiatives support medium-term margin expansion

Overview

Goldman Sachs' Kitakami plant tour review argues that Kioxia's manufacturing design and automation initiatives should raise investment efficiency and support stronger profit margins over the medium term. The firm reiterates Buy, supported by its view that NAND supply-demand will remain tight and shareholder returns can strengthen.

Core views

Goldman Sachs says its 28 September tour of Kioxia's Kitakami plant did not add much information directly relevant to near-term earnings, but it improved the firm's understanding of initiatives intended to maximize investment efficiency. The central medium-term conclusion is that higher production efficiency at Kitakami should enable stronger profit margins. Goldman Sachs reiterates Buy, also citing an expectation that tight NAND supply-demand will persist for the time being and that shareholder returns can improve. Its 2Q FY3/27 operating-profit forecast of ¥1.98 trillion is above company guidance of ¥1.89 trillion, and it sees no particular risk factors to that guidance at present. Kitakami is one of Kioxia's main production bases alongside Yokkaichi. K1, a roughly 40,000-square-metre fabrication facility, began operations in 2020; K2, covering 31,000 square metres, commenced operations in September 2025. The site is dedicated to front-end wafer processes and currently does not undertake back-end processes. Kioxia has announced K3 on land south of K2, with the intention of making it at least as large as the existing buildings; operations are scheduled to begin in FY3/30, although no more specific timing was provided. The report highlights operational features that it views as efficiency enablers. Both K1 and K2 use automated production systems for operator tasks and transport systems intended to improve production efficiency. K2's seven-story structure includes two sub-fab floors beneath the cleanroom, versus one sub-fab floor in the five-story K1. As technology generations make processes more complex, chamber counts rise and more equipment requires substantial sub-fab space; the two-story design is intended to save space and simplify maintenance. Equipment is installed in just under half of K2, with further deliveries expected in stages through 1H FY3/28. Both facilities also use seismic isolation structures to reduce shaking from major earthquakes. At Kitakami, BiCS5 remains the main product, while Kioxia began a company-wide crossover to BiCS8 at the end of FY3/26. BiCS8 output at Kitakami is expected to increase in stages through FY3/28 after the Yokkaichi transition, and Kioxia also intends to pursue commercial BiCS10 production there. Management indicated that further production-efficiency improvement, including greater scale, will be needed to lift profitability. The two plants retain complementary roles: while production is not rigidly confined by product generation or site, the stated direction is for Kitakami to produce higher-capacity products and Yokkaichi to produce high-bandwidth and high-performance products. Goldman Sachs values Kioxia at a 12-month target price of ¥116,000 using the relationship between ROE and P/B and the average of its FY3/27-FY3/28 ROE estimates. The target implies FY3/28E P/E of 7.0x and P/B of 5.7x. The report lists a ¥53,800 share price as of the 28 September 2026 close, implying 115.6% upside to the target.

Analysis framework

The report combines observations from a plant tour with Kioxia's disclosed production plans and Goldman Sachs earnings estimates. It links facility automation, layout and capacity deployment to production efficiency and future margins, then supports its valuation with an ROE-to-P/B relationship using average FY3/27-FY3/28 ROE estimates.

Methodology notes

• Valuation methods

  ROE-P/B correlation valuation

  Goldman Sachs sets its 12-month target price by relating Kioxia's expected ROE to P/B, using the average of its FY3/27 and FY3/28 ROE estimates.

• Industry AnalysisUpstream-Midstream-Downstream Transmission

  Manufacturing efficiency and technology-transition analysis

  The report assesses how fab automation, space design, equipment installation and NAND-generation transitions could flow through to production efficiency and profit margins.

Asset mapping & comparison

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

• Kioxia Holdings (285A.T)

  Primary covered company; expected to benefit from improved Kitakami production efficiency and sustained tight NAND supply-demand.

  Strengths

  Automation, transport systems, expanded sub-fab capacity, phased BiCS8 output and planned BiCS10 production.

  Weaknesses

  Near-term earnings information from the plant tour was limited, and K2 remains only partly equipped.

  Comparison

  Kitakami is oriented toward higher-capacity products, while Yokkaichi is generally oriented toward high-bandwidth and high-performance products.

  Risks

  AI-investment slowdown, Chinese NAND competition, cost or utilization pressure on margins, yen appreciation, and weaker non-AI NAND demand.

Key data

• 12-month target price¥116,000Based on ROE-P/B correlation and average FY3/27-FY3/28 ROE estimates.
• Share price¥53,800Price as of 28 September 2026 close.
• Implied upside115.6%Upside from the reported share price to the target price.
• 2Q FY3/27 operating-profit forecast¥1.98 trillionGoldman Sachs forecast versus company guidance of ¥1.89 trillion.
• FY3/28E valuationP/E 7.0x; P/B 5.7xImplied by Goldman Sachs' ¥116,000 target price.
• K2 equipment installationJust under half of available spaceFurther equipment deliveries are expected in stages through 1H FY3/28.

Impact & implications

The report argues that Kitakami's automation, facility configuration and gradual capacity and technology deployment can improve Kioxia's investment efficiency and lift margins over time. It views continued NAND market tightness and potential stronger shareholder returns as additional support for its Buy stance.

Risks

• A slowdown in AI investment centered on hyperscalers could weaken demand.
• The rise of NAND players in China could intensify competition.
• Higher costs or changes in capacity utilization could reduce profit margins.
• A sharp appreciation of the yen against the US dollar could be adverse.
• A slowdown in NAND demand for non-AI applications could pressure the outlook.