HSBC Raises Intel Price Target to $200, Setting a New Wall Street High
Alina Collins
HSBC doubled its Intel price target from $100 to $200 — now the Street high — arguing that internal foundry capacity reallocation will drive two consecutive years of above-consensus server CPU shipment growth. This means → the market is repricing Intel's shift from chip company to foundry platform.
A target price doubled — where does HSBC's conviction come from?
Analyst Frank Lee raised Intel's target from $100 straight to $200, maintaining a buy rating. Intel shares rose in early trading on the news.
This means → HSBC believes the current price reflects only half of Intel's value. The other half sits in a foundry transformation the market has not fully priced.
The core driver is internal foundry capacity reallocation — Intel is freeing up wafer capacity previously reserved for its own chips to take external orders, lifting utilization and opening a new revenue stream.
Why should server CPU shipments beat expectations?
Lee raised his 2026 server CPU shipment growth forecast from 20% to 25% year-on-year, implying Data Center & AI (DCAI) revenue of $24.1 billion — 4% above consensus.
2027 is even more aggressive: growth lifted from 20% to 30%, on two legs — continued capacity release from reallocation, and the 18A process node (Intel's next-generation advanced manufacturing technology) ramping ahead of internal expectations.
In plain terms = even after consensus rose 23% since Q1 earnings, HSBC says the Street is still underestimating how fast Intel's capacity is coming online.
How is the foundry narrative actually improving?
TSMC's additional 3 nm capacity won't come online until the second half of 2027, forcing customers to look for new foundry partners.
Intel has already signed Terafab and Apple as customers and is in discussions with Google and Nvidia. This means → the foundry business is moving from "storytelling" to "order-signing," giving the narrative real contract backing.
This reflects a deeper shift: the global chip-manufacturing capacity bottleneck is pushing major customers — once TSMC-exclusive — toward Intel.
EMIB packaging — Intel's differentiation card?
EMIB — Embedded Multi-die Interconnect Bridge, a packaging technology that stitches multiple chips together — scales to 12× reticle area, versus a 3.3× maximum for TSMC's CoWoS-S.
In plain terms = the larger the reticle area, the more chips you can stitch together and the higher the combined performance. Intel can package at nearly 4× TSMC's scale — a tangible edge in the era of massive AI chips.
Meanwhile, TSMC's CoWoS-L capacity is largely locked up by Nvidia. Other customers wanting advanced packaging have to look elsewhere — and EMIB fills that gap.
What still needs to prove out?
Two critical checkpoints: whether customer engagements convert into stable foundry orders, and whether 18A production yields keep running ahead of internal targets.
This means → HSBC's $200 target is a best-case bet — foundry orders landing and yields hitting targets both have to come through. If either slips, the valuation wobbles.
Intel is at the "best-story" stage right now: customers in talks, the process node advancing, competitors capacity-constrained. Order conversion over the next 12 months will decide whether this Street-high target is foresight or overreach.
Content is for reference only, not financial advice.