It took just 50 days. The ¥1.3 trillion ($8.2 billion) Denso bid for Rohm, one of the largest M&A proposals in modern Japanese industrial history, is being withdrawn by Denso itself, news broke on April 25. Rohm wouldn't sign on. Instead, Rohm is moving forward with the three-way alliance with Toshiba and Mitsubishi Electric. Japan's power semiconductor reshuffle has chosen the "national consortium" path, and Toyota Group's chip in-house strategy now needs a complete reset.

Related Articles This is the final chapter of a three-part series. For background, see:

Breaking: Denso Withdraws

On April 25, 2026, Yomiuri Online and Nikkei simultaneously reported that Denso, Toyota Group's auto parts giant, is preparing to withdraw its acquisition proposal for semiconductor maker Rohm.

According to reports, Denso said it will "actively explore partnership possibilities with third parties, not limited to Rohm" and will reset its semiconductor strategy. The roughly 50-day saga over the trillion-yen-plus deal has effectively ended without Rohm's endorsement.

The reason is straightforward. Per Nikkei's framing: "the two sides could not see eye-to-eye." Rohm reportedly had concerns about becoming overly reliant on automotive chips and damaging its existing relationships with other auto parts manufacturers. As an independent Kyoto-based maker, resistance to falling under Toyota Group's umbrella never disappeared.

Update (April 28, 2026): Denso formally announced the withdrawal at its FY2026 earnings briefing on April 28. President Shinnosuke Hayashi said the company could not obtain Rohm's endorsement and that continuing talks would not produce a scenario raising both companies' value, while adding that collaboration and personnel exchange with Rohm will continue.

Timeline: 50 Days from March 6 to April 25

Looking back, the speed of events stands out.

September 2024: Denso and Rohm announce a partnership in analog semiconductors.

May 2025: Both companies sign a basic agreement on strategic semiconductor partnership.

End of September 2025: Denso has built up a 4.98% stake in Rohm.

February 2026: Denso submits a tender-offer proposal to acquire all Rohm shares. Rohm sets up a special committee composed of independent outside directors to evaluate it.

March 6, 2026: Nikkei breaks the news. Rohm's stock hits the daily upper limit (+18% to ¥3,243). Denso's stock falls 3–5%.

March 24, 2026: Denso officially announces its acquisition proposal.

March 27, 2026: Rohm announces three-way merger talks with Toshiba Device & Storage and Mitsubishi Electric. Combined global share of about 11% would put the alliance second only to Germany's Infineon.

March 31, 2026: Denso unveils its mid-term plan "CORE 2030." CEO Hayashi declares "transformation into a semiconductor maker" while saying Denso "is not excluding broad possibilities" on the Rohm bid.

April 3, 2026: Rohm CEO Katsumi Higashi remarks "too many cooks spoil the broth" and "we want to be at the center", clearly signaling preference for the three-way alliance.

April 25, 2026: Denso's withdrawal becomes public. Rohm proceeds with the three-way alliance.

In barely a month and a half, Japan's power semiconductor map has been redrawn.

Why Rohm Chose the Three-Way Alliance

Rohm's ultimate refusal to fall under Denso reflects multiple converging factors.

First, management autonomy. As an independent Kyoto-based maker with nearly a century of history (including its predecessor entity), being absorbed by Aichi-based Toyota Group's core company was emotionally and strategically resisted. CEO Higashi has repeatedly invoked the failure of Elpida Memory in interviews, vowing "not to repeat the same mistake."

Second, sales channel concerns. Falling under Denso would effectively reorient Rohm's chips toward Toyota-Group automotive applications. But Rohm has strengths in consumer electronics, industrial equipment, and smartphones, becoming an "automotive-only" maker would mean a fundamental business model shift. The "impact on relationships with other auto parts makers" cited in Nikkei's reporting reflects this, selling to competing Tier 1s like Bosch, Continental, and ZF would become practically impossible under Denso ownership.

Third, capital ties with Toshiba. Rohm contributed about ¥300 billion ($1.9 billion) to the JIP-led consortium that took Toshiba private in 2023. This capital linkage made three-way integration easier, and made a Denso buyout much more complicated.

Fourth, METI's intent. Japan's Ministry of Economy, Trade and Industry has long pushed for domestic power semiconductor consolidation. Subsidies of ¥129.4 billion ($810 million) went to Rohm and Toshiba, and ¥22 billion ($138 million) to Mitsubishi Electric's Kumamoto 8-inch plant. With public money involved, "absorption by a single automaker" doesn't fit policy intent as well as a "national three-way consortium."

What "Reset" Means for Denso

The withdrawal hurts Denso, but it can also be read as a rational retreat.

A tender offer without target support amounts to a hostile takeover. While legally possible, being perceived as "raiding" an independent Kyoto-based maker carries long-term reputational risks for Denso's brand and relationships with other companies. Spending ¥1.3 trillion plus 2–3 years on integration may have looked less attractive than exploring alternatives.

Denso has held about 4.98% in Rohm since September 2024, and the existing business partnership can continue. Even without full acquisition, partial access to Rohm's SiC power semiconductor technology is preserved. The "below equity-method" practical benefits don't disappear.

What's notable is Denso's "CORE 2030" plan unveiled March 31. CEO Shinnosuke Hayashi explicitly stated Denso would become a "semiconductor maker," expanding into automotive, industrial equipment, and consumer electronics. Rohm was supposed to be the centerpiece of this grand plan. What now?

Candidates that emerge include Fuji Electric, Renesas Electronics, or partnerships with overseas players. Denso has historically collaborated with Fuji Electric on power semiconductors. The "third-party partnership" comment likely has these options in mind.

Three-Way Alliance: Can It Really Be #2 Globally?

If the Rohm-Toshiba-Mitsubishi Electric alliance materializes, combined global share would be around 11%. That trails Infineon (about 23–24%) but matches U.S. onsemi (about 11%), placing the consortium in the world's top tier.

But that's just a "sum of shares." Actual integration benefits require massive PMI (post-merger integration) work, technology complementarity, manufacturing line optimization, HR system unification, IT infrastructure integration. Large M&A integrations typically take 2–3 years.

The alliance's strength lies in genuinely complementary technology portfolios:

  • Rohm: Vertically integrated SiC power semiconductors (in-house from wafer to product), strong in discrete semiconductors
  • Toshiba Device & Storage: Silicon power semiconductors, 300mm wafer lines, IGBT modules
  • Mitsubishi Electric: High-voltage modules, automotive and industrial track record, new Fukuoka plant (full operations October 2026)

Past "Japan alliances" like Japan Display Inc. (JDI) and Elpida Memory failed because they combined "unwanted" units from each parent company. This time, each company is contributing its cash cow or future ace business, structurally a different proposition. So goes the optimistic case.

But global rivals are moving fast. Infineon's 200mm SiC megafab in Kulim, Malaysia is coming online. Onsemi is launching 8-inch SiC volume production in Roznov, Czechia. STMicro is building new fabs in Italy and through a Chinese JV. BYD and CRRC Times Electric are accelerating in-house production. The consortium's "agreement by summer" plus 2–3 years to actually integrate gives global rivals plenty of runway.

The Economic Security Lens

Power semiconductors sit at the core of Japan's economic security policy.

EV motor control, data center power management, renewable energy inverters, defense equipment power systems, all depend on efficient power conversion and control. Heavy reliance on overseas (especially Chinese) supply represents a dual risk: energy transition and national defense.

METI's 2024 revision of the "Semiconductor and Digital Industry Strategy" designated power semiconductors a priority area, with ¥129.4 billion in subsidies for Rohm and Toshiba's new SiC fabs and ¥22 billion for Mitsubishi Electric's Kumamoto 8-inch plant. The unstated policy goal: "consolidate Japanese power semiconductor makers into globally competitive scale."

Denso's solo acquisition didn't fit this intent perfectly. A maker optimized for Toyota Group means limited industry-wide spillover. The three-way alliance, where each company maintains corporate independence while integrating only the power semiconductor businesses, fits the policy template much better.

Positioning in the U.S.-China Chip War

The global semiconductor industry sits on the front line of U.S.-China rivalry. The U.S. has tightened export controls on advanced logic chips to China; China is accelerating domestic production through subsidies. Taiwan's TSMC is diversifying with investments in Arizona, Kumamoto, and Dresden. South Korea's Samsung is expanding in Texas.

Within this larger picture, Japanese power semiconductors fit into "U.S.-side supply chains." The U.S. Inflation Reduction Act (IRA) battery and EV subsidies, plus Europe's Critical Raw Materials Act (CRMA), all support demand for Japanese power chips.

If the three-way alliance grows into world No. 2, Japanese share recovery in U.S. and European EV and renewable energy markets has real upside. As an alternative to "Infineon dominance," Western automotive and energy companies seeking risk diversification may increasingly choose Japan.

Different Game from TSMC and Samsung

It's important to note that power semiconductors are a fundamentally different game from what TSMC and Samsung pursue.

TSMC focuses on 2nm and 3nm advanced logic chips, the "brains" of smartphones and AI servers. This is a race to ever-finer transistor miniaturization. Samsung competes on the same battlefield. Japan's Rapidus, the new venture in Chitose, Hokkaido, is also pursuing this advanced logic territory.

Power semiconductors compete on materials and packaging, not miniaturization. Wide-bandgap semiconductor materials like SiC and GaN, wafer scaling (150mm to 200mm), and module integration know-how, these are the competitive dimensions. TSMC is also entering power semiconductors, but logic chips remain its core.

In other words, Japan is fighting a two-front war: Rapidus for logic, the three-way alliance for power. The Denso withdrawal is strategically significant because, on the power semiconductor front, Japan chose the path of "semiconductor-maker alliance" over "absorption by an automaker."

Rohm Stock and the Special Committee

After surging to its daily upper limit (¥3,243) on March 6, Rohm's stock has held at elevated levels supported by three-way alliance expectations. Following formal withdrawal, disappointment over the lost TOB premium will mix with three-way integration optimism in market reactions.

Rohm's special committee, composed of independent outside directors, was tasked with independently evaluating the Denso proposal. If "Rohm could not endorse" the bid, the committee was likely heading toward effective rejection.

That said, some investors bought Rohm betting on TOB premium. Whether the special committee can justify "shareholder value maximization" by prioritizing long-term business value over short-term returns will become a focal point of governance debate.

Toyota Group Implications

Denso's withdrawal also affects Toyota Group's broader semiconductor strategy.

Toyota has historically relied on "long-term partnerships with chipmakers" rather than direct ownership. Denso's move on Rohm signaled a shift toward in-house production at the group level. With withdrawal, Toyota Group reverts to "external procurement plus selective in-house" architecture.

But Denso hasn't given up on becoming a "semiconductor maker." Of CORE 2030's ¥3.7 trillion R&D and ¥2.2 trillion capex over five years, a substantial share will likely target semiconductors. Deepened Fuji Electric collaboration, new Renesas relationships, or JV with overseas makers, options remain.

Notable: Denso signed a January 2026 joint development agreement with MediaTek (Taiwan) for automotive SoC (system-on-chip). SoC sits in the logic territory, distinct from Rohm's strengths. Denso has been pursuing "power semiconductor + SoC" parallel strategies, Rohm represented one wing. The SoC strategy continues even after the Rohm withdrawal.

Will the "Summer Direction" Materialize?

The three-way alliance has stated "direction by this summer." With Denso withdrawn, three-way talks should accelerate.

Open questions remain abundant:

  • Equity structure: JV format, or Rohm-centered merger?
  • Headquarters location: Kyoto (Rohm), Tokyo (Toshiba), or Fukuoka (Mitsubishi Electric's new plant)?
  • CEO: Which company supplies the chief executive?
  • Plant rationalization: How to coordinate Rohm's SiC fab, Toshiba's 300mm lines, and Mitsubishi Electric's Kumamoto 8-inch plant?
  • Brand strategy: Keep three brands, or launch a unified new one?

Past examples are cautionary. JDI and Elpida prized "equality" to the point of paralyzed decision-making and lost competitiveness. The three-way alliance must establish clear leadership to avoid that fate. CEO Higashi's "we want to be at the center" remark may have been a preemptive strike in the leadership contest.

The Real Threat: Chinese SiC Makers

A final point worth flagging: the rapid rise of Chinese SiC makers.

Rohm's ¥50 billion ($310 million) net loss in fiscal 2025, its first in 12 years, was driven not just by EV market slowdown but by Chinese SiC pricing pressure. BYD Semiconductor, CRRC Times Electric, TanKeBlue, and SICC (Shandong Tianyue) are leveraging government subsidies to push aggressive low-price strategies. Japan's JS Foundry, which had been targeting power semiconductor contract manufacturing, lost the price war and filed for bankruptcy in July 2025.

Even if the three-way alliance reaches world No. 2 in scale, Chinese pursuit will continue. Japanese makers retain a technology edge, but cost competitiveness is weak. METI's subsidies are essentially defensive, countering "Chinese deflation" in the chip space.

For the three-way alliance to truly be "globally competitive," scale alone isn't enough. Frontline investment in next-generation SiC (8-inch to 12-inch transition), GaN, gallium oxide, and diamond semiconductors, areas where China hasn't yet caught up, is decisive.

The "Real Face" of Japanese Industry Revealed in 50 Days

Over 50 days from March 6 to April 25, Japan's power semiconductor industry moved more than it had in years.

Denso's ¥1.3 trillion bid ended in short-term failure. But that proposal acted as the "external pressure" that finally pushed Rohm-Toshiba-Mitsubishi Electric three-way merger talks, long discussed but never executed, into formal motion. Denso's withdrawal isn't defeat: in some sense, "the role has been played."

What matters now is what comes next. Can the three-way alliance actually become globally competitive? Can Denso execute an alternative strategy successfully? How will Japan respond to the Chinese push? How long will Japanese government support continue under the economic security framework?

The dramatic 50-day chapter is closing. But the real battle for Japan's power semiconductor industry is just beginning.

Has your country seen automaker bids for chipmakers fall apart? Do national strategies pushing "industry consolidation" actually work where you live? In the U.S.-China chip war, where does your country's semiconductor industry stand? We'd love to hear your perspective.

References