Three Japanese firms have agreed on the financial structure for a power semiconductor consolidation, but negotiations remain deadlocked over whether the new entity should focus exclusively on power devices or retain analog and sensor businesses.
This is the scene unfolding in Japan's power semiconductor industry. Since signing a memorandum of understanding in March this year, Mitsubishi Electric, Toshiba, and Rohm have been negotiating the integration of their respective power semiconductor businesses. On October 2, Economy, Trade and Industry Minister Koizumi Akira stated at a press conference following a cabinet meeting that the ministry supports domestic collaboration and restructuring in the power semiconductor sector to strengthen the supply chain and enhance industrial competitiveness.
Layer One: The Money and Structure Are Already Agreed
The framework for the deal has emerged. The core plan involves establishing a holding company, with a single operating entity absorbing the power semiconductor divisions of all three firms. Rohm and Toshiba's analog semiconductor and sensor businesses would be spun off separately. The parties aim to capture approximately 11.3% of the global power device market, making the combined entity the world's second-largest power semiconductor manufacturer, trailing only Infineon.
For the Japanese industrial sector, the urgency of this consolidation stems from a stark contrast: Japanese companies have maintained a strong presence in power semiconductors, but their fragmented, individual efforts result in smaller business scales. In contrast, European manufacturers have long since completed their business splits and focus, securing the top two positions in automotive and industrial segments.
Layer Two: Stuck on Business Boundaries
The real point of contention is the scope of operations. Mitsubishi Electric advocates for the new entity to focus exclusively on power chips, while Rohm and Toshiba wish to retain their respective analog chip businesses. This disagreement has slowed progress at the working level, eventually drawing senior management into the negotiations. Beyond business boundaries, the equity split of the new company, the specific scope of divested businesses, and leadership control remain unresolved.
Rohm President Katsumi Azuma, who takes office in 2026, has previously stated that the difficulty and duration of the tripartite cooperation negotiations have exceeded expectations. Integrating production assets, allocating R&D resources, and coordinating sales systems are all challenging to implement.
1 In March 2026, Mitsubishi Electric, Toshiba, and Rohm signed a memorandum of understanding, officially launching negotiations to integrate their power semiconductor businesses, with the initial intent for Mitsubishi Electric to lead the formation of a joint venture.
2 The final agreement was originally targeted for "around summer." This timeline has passed, and negotiations remain deadlocked over the issue of business boundaries.
On October 2, 2026, Economy, Trade and Industry Minister Akazawa Ryozo expressed support for the restructuring of domestic power semiconductor operations at a press conference, explicitly framing the initiative within the policy framework of the 'Semiconductor and Digital Industry Strategy.'
The near-term target is to align with the November interim earnings announcement, but if adjustments to the integration structure continue to drag on, the timeline could slip to March of next year.
"Power semiconductors are critical to supporting a wide range of industries. From the perspective of strengthening the domestic supply chain and enhancing industrial competitiveness, we want to provide substantive support for collaboration and restructuring among domestic companies."
— Akazawa Ryozo, Minister of Economy, Trade and Industry, Japan, at the October 2, 2026 press conference
Layer Three: One More Approval Hurdle to Clear
Two other bottlenecks are also delaying overall progress. First, the companies themselves are stretched thin by their core operations—the surge in global semiconductor demand driven by the AI wave makes it difficult for all three parties to allocate sufficient personnel to handle integration matters. Second, the deal requires antitrust approval from Chinese regulators, a review process that inherently consumes additional time.
One point prone to misinterpretation needs clarification: Rohm and Toshiba have a separate government-backed production cooperation project with a subsidy cap of 129.4 billion yen. This is an existing cooperative arrangement between the two firms and is distinct from the current three-party integration of power semiconductor businesses; it cannot be counted as a policy consideration for the merger.
As horizontal integration faces obstacles, another path has already been attempted. The Rohm acquisition initiated by Denso with Toyota's backing ultimately failed, meaning the vertical integration route is currently blocked. Regarding horizontal mergers, past consolidation cases in the Japanese semiconductor industry have mostly been passive restructurings following operational distress, with few examples of companies proactively joining forces to compete for global market share.
Here is a testable judgment: even if the three parties ultimately sign an agreement, the new entity's substantive impact on the power semiconductor market will not be visible for at least three years, and the global pricing landscape for power semiconductors will not change in the short term. The reasoning is that the truly difficult part of integration happens after signing—each company's production lines, process platforms, and customer certifications must be realigned, and the certification cycle for automotive-grade products is inherently long. If, within 2027, this new entity appears on the list of key suppliers for automotive-grade silicon carbide (SiC) modules and exerts significant downward pressure on market prices, this judgment will be proven wrong.
If you were a Rohm executive, would you hand over the analog chip business for this merger? Power devices bring scale, while the analog business brings gross margins and customer stickiness—when you are reluctant to let go of either, collapsing the talks is actually the most convenient option.