Rohm's SiC Business: A Look at the 158.4 Billion Yen Loss - Excess Inventory, Denso Partnership, and Merger Talks with Toshiba Mitsubishi-Electric

What a ¥158.4 Billion Deficit Means - The Structural Crisis in Rohm's SiC Business

For the fiscal year ending March 2025, Rohm reported a consolidated operating loss of ¥158.4 billion. While the figure itself is shocking, what is more serious is that a substantial portion of this loss stems from inventory write-downs and excess capacity within its Silicon Carbide (SiC) business. Investments meticulously built up for what was deemed "the definitive next-generation power semiconductor" have directly translated into losses at a cyclical downturn in demand.

Rohm is one of the earliest manufacturers to achieve mass production of SiC power semiconductors globally. The company established production lines for Schottky Barrier Diodes (SBDs) and MOSFETs in the early 2010s and has continued aggressive capital investment in anticipation of growing demand from Electric Vehicles (EVs). This strategy dramatically backfired with the slowdown in EV demand and a sharp drop in market prices from 2024 onwards.

EV Slowdown and SiC Inventory Glut - What Coincided at This Juncture

It is widely recognized within the industry that the overall SiC market entered a demand plateau in the latter half of 2023. Tesla's indication of a reduction in SiC module procurement, intensified price competition in the Chinese EV market, and a rollback of EV launch plans by European OEMs converged, leading to a simultaneous oversupply of expanded SiC production capacities.

For Rohm, the reason why this situation translated into such a significant figure is its position as an independent supplier. While companies like Infineon and STMicroelectronics have structures that can absorb demand fluctuations more easily through vertical integration and long-term contracts, Rohm has a higher proportion of sales to the open market with a broad lineup of discrete devices, bare dies, and modules. When the market shrinks, inventory tends to pile up at a faster rate.

Denso's TOB Withdrawal - The Axis of Industry Reorganization Has Shifted

What cannot be overlooked in the movements surrounding Rohm is the series of events concerning its capital relationship with Denso. In February 2026, Denso proposed a Tender Offer Bid (TOB) to acquire all of Rohm's shares. This garnered significant attention as a large-scale M&A aimed at strengthening the semiconductor business. However, a special committee composed of Rohm's outside directors concluded that it "did not reach a conclusion to approve the proposal." On April 28th, Denso formally withdrew its acquisition offer.

Denso's President Hayashi stated, "We were unable to envision a scenario that would lead to enhanced value for both companies even if we continued discussions." While the framework of capital integration through a TOB has disappeared, the two companies agreed to "continue personnel exchanges and co-creation activities," maintaining their collaborative relationship, particularly focused on analog semiconductors. However, utilization of Denso for "strengthening the financial base" and as an "anchor for demand" has, at least for the time being, not been realized.

Integration Negotiations with Toshiba and Mitsubishi Electric - Will It Become the Epicenter of Industry Reorganization?

Further advancing the discussion, there are reports of negotiations towards integrating with Toshiba Device & Storage and Mitsubishi Electric's power semiconductor businesses. If this becomes a reality, the structure of Japan's power semiconductor industry will fundamentally change.

The strengths in technology and products held by the three companies can be summarized as follows:

If the integration is realized, it could create scale merits not only through product line complementarity but also in areas such as standardization of manufacturing processes and inspection technologies, and improved procurement negotiation power. On the other hand, the costs of cultural and organizational integration, and customer concerns about the "risk of falling under the same umbrella as competitors," remain practical challenges.

In Europe, acquisition strategies by Infineon and alliances with ST have accelerated market consolidation. The fact that Japanese power semiconductor manufacturers are discussing such in-depth integration can be seen as a move mindful of this international competition. While the extent to which negotiations will proceed remains uncertain, "the mere existence of these negotiations" is already beginning to alter industry dynamics.

Short-Circuit Withstand Time and Device Competitiveness - Why Technology Matters

Regardless of whether integration negotiations progress or an independent path continues, Rohm's recovery of market valuation will be predicated on the technological competitiveness of its products. In this regard, what has become an important evaluation metric in the industry is the "short-circuit withstand time (SCWT)" of SiC MOSFETs.

Short-circuit withstand time refers to the duration a device can withstand a short-circuit condition before failing, serving as an indicator of the grace period before protective circuits activate. Due to their smaller die size and higher current density, SiC devices experience faster temperature rises compared to their silicon counterparts. For example, Microchip's 700V/1200V rated products state a typical value of 3µs under specific conditions in their datasheets. This is significantly shorter than the approximately 10µs of silicon IGBTs and directly impacts the design of protective circuits.

Short-circuit withstand time and on-resistance (Ron) are in a trade-off relationship. The lower the on-resistance and the reduced switching losses, the higher the saturation current during a short circuit, and the faster the thermal damage to the device tends to occur. Rohm has announced that its 4th generation SiC MOSFETs mitigate this trade-off with their unique device structure, and Mitsubishi Electric is addressing the same challenge by introducing a p-type protective layer in its trench-type devices. In the context of integration negotiations, these differing technological approaches directly translate into the practical question of "how to draw up the product roadmap after integration."

For protective circuit design, the DESAT (Desaturation) function is commonly employed. DESAT monitors the drain-source voltage (VDS) in the on-state and turns off the device when overcurrent is detected. Key parameters for design decisions include the DESAT trigger threshold (VDESAT), DESAT current (IDESAT), and short-circuit blanking time.

So, What Should We Watch For - Questions from Technology, Procurement, and Business Perspectives

Rohm's ¥158.4 billion loss should be interpreted not as an isolated financial anomaly but as evidence that the supply-demand cycle for the entire SiC industry has passed its initial peak.

Here are the key points to consider from both technological and procurement perspectives:

The view gaining traction within the industry is that the SiC market has not shrunk but is in a transitional phase where demand is shifting from EVs to industrial applications, data centers, and solar power. Rohm's current efforts in financial restructuring and negotiations for partnerships and integration are a process of seeking answers to the question of how to navigate this transitional period. The answers will likely emerge a bit further down the line.