Micron slides after S. Korea unveils $518B chip plan

Micron Technology shares fell Monday, trading about 1% lower at $1,117.19 after dropping 6.7% on Friday, as investors reacted to a large South Korean investment plan that could expand memory-chip capacity.

South Korea’s industry minister announced that Samsung Electronics and SK Hynix will commit a combined 800 trillion won, roughly $518.6 billion, to develop new semiconductor manufacturing hubs in the country’s southwest region.

High-bandwidth memory, or HBM, is a key component in advanced artificial-intelligence systems. Micron, Samsung and SK Hynix are the main global suppliers of HBM. Market participants flagged the scale of the South Korean spending as a factor that could increase competition in the memory market over time.

Large fabrication plants take years to build and to reach full production. Micron’s own multibillion-dollar project in New York, announced in 2022 and valued at about $100 billion, is not expected to begin production until 2030, illustrating the long timelines involved.

Micron recently reported fiscal third-quarter revenue of $41.46 billion, compared with $9.3 billion a year earlier. Management forecast roughly $50 billion in revenue for the current quarter versus $11.3 billion a year earlier.

Following the earnings, several analysts raised forecasts for Micron. Barclays analyst Thomas O’Malley increased his price target to $2,000 and raised his fiscal 2027 earnings-per-share estimate to $166.74 from $106.77.

Micron disclosed that it has signed long-term supply agreements with 16 customers across data center, consumer and automotive markets. Most contracts run five years from 2026 through 2030, while automotive agreements generally span three years. The signed deals represent about 20% of Micron’s DRAM volume and roughly 33% of its NAND volume. Fourteen of the 16 contracts include cumulative minimum revenue commitments totaling about $100 billion.

Investors and analysts are weighing the long development timeline for new capacity against current supply constraints and ongoing demand for AI-related memory products.