Samsung Considers Splitting Leadership of Memory and Mobile Divisions
nashnova research
Samsung Electronics may announce as early as mid-November whether to appoint separate heads for its memory and mobile businesses — one unit now delivers over 99% of operating profit while the other is losing money, making the case for a split hard to defer.
Who runs what today — and why split now?
Two executives each hold dual roles: DS division vice chairman Jun Young-hyun doubles as memory chief; DX division president Roh Tae-moon directly oversees the mobile (MX) business.
This means → Samsung's two largest units have no dedicated leader. Strategic attention is diluted across portfolios.
The trigger is extreme financial divergence between the two businesses — one setting records, the other bleeding cash. Keeping one person over both is increasingly hard to justify.
How strong is memory? How weak is mobile?
The DS division posted KRW 14.71 trillion in Q3 2026 revenue and KRW 10.65 trillion in operating profit — over 99% of Samsung's total. Both figures are all-time records.
This means → virtually all of Samsung's profit comes from one line. The drivers: rising HBM4 (high-bandwidth memory chips built for AI servers) shipments and server SSD demand.
The DX division recorded an operating loss of KRW 0.8 trillion; mobile and network combined lost KRW 0.6 trillion. In plain terms = rising memory prices boost the semiconductor arm's earnings but simultaneously inflate component costs for Samsung's own phones — the left hand earns what the right hand gives back.
How likely is a leadership change on the mobile side?
MX chief operating officer Choi Won-joon is the leading candidate to replace Roh Tae-moon. He previously worked at Qualcomm and was appointed MX COO in April 2025.
If Roh steps back from running MX directly, he would focus on the broader DX division — smartphones, TVs, and home appliances.
This means → the signal for a mobile leadership change is stronger than for memory. A loss-making unit faces less resistance to a handover, and there is already a ready successor inside.
Why is the memory side in less of a hurry?
According to iNews24, Samsung may keep Jun as memory head through 2027 to preserve continuity with key customers.
In plain terms = semiconductor suppliers and their clients go through product development, qualification, and supply agreements — cycles that take years. Swapping leadership mid-process risks disrupting deals already in the pipeline.
Candidates on the memory side include DS chief technology officer Song Jai-hyuk (led flash development and semiconductor research) and memory development executive Hwang Sang-joon (heads DRAM technology).
What would a split mean for Samsung's foundry business?
If Jun is relieved of his memory duties, he gains bandwidth to focus on Samsung's foundry and system-LSI operations.
The foundry unit already showed Q3 improvement: higher sales of HBM base dies and advanced-node chips, plus new 2 nm orders from hyperscale cloud customers. Samsung's Taylor, Texas fab is expected to begin production in 2027.
This reflects Samsung's strategic priority: memory is already running; foundry is the gap that must be closed — but progress depends on whether leadership actually frees up the capacity to drive it.
Why didn't Samsung split last time — and will this round be different?
In its November 2025 executive reshuffle, Samsung chose to keep the dual-role structure intact.
This means → splitting leadership is not a new idea; it was shelved once before. What has changed: memory's profit share has pushed toward 100%, mobile losses have deepened, and foundry orders are recovering — three variables now pushing simultaneously, making the case harder to dismiss than a year ago.
The appointment announcement, expected as early as mid-November, will be the clearest signal yet of whether Samsung can shift from "one person, multiple hats" to dedicated leadership for each pillar.
市场有风险，内容仅供研究参考，不构成投资建议。