top of page

📢 STT Premium is now open

Every free article stays free — nothing moves behind a paywall. Premium adds the weekly deep-dive, full earnings breakdowns and all long-form reports. Note: premium articles are published in Traditional Chinese.

Free keeps you current. Premium lets you judge.

NT$199/month · NT$2,000/year (about NT$167/month)

👉 Subscribe: vocus.cc/salon/simpletechtrend

A-Share Laser-Chip Stocks Priced the Wrong Year: Laser Capacity Is Booked to 2029, and InP Supply, Not the FCC List, Decides 2027 — STT Weekly W41 2026

3 days ago
2 min read

Free Section

On October 5 we wrote about US curbs on Chinese optical modules and the 3.2T threshold (in Chinese). One week later, three new facts have landed.

First, a Morgan Stanley note from October 1 modeling an FCC 3.2T scenario hit the market after China's holiday. On October 8, A-share laser-chip names sold off hard: Yuanjie and Everbright Photonics fell by the 20% daily limit, Dongshan Precision fell by its 10% limit, and the Wind optical-chip index dropped 8.18% in a day.

Second, Taiwan's September revenue came out. LandMark Optoelectronics booked NT$540M, up 190.2% YoY, while EZconn, Elaser and Luxnet all grew close to 100% YoY.

Third, Lumentum's CEO said in Tokyo on October 9 that some capacity is already committed into early 2029, that around 70% of demand for some products cannot be met next year, and that adding new laser capacity takes about three years.

In the same week, A-shares sold on the policy clock while Taiwan's revenue ran on the shortage clock. STT's view: through 2027, laser capacity, not the FCC list, decides laser-chip makers' revenue, and the two clocks meet in 2029.

▲ Figure: The shortage clock vs. the policy clock on one timeline; they converge in 2028–2029. Sources: Lumentum CEO 10/9 (Bloomberg TV via Finviz), Morgan Stanley 10/1 (ABMedia, 21jingji), Marvell Investor Day 10/6, SemiAnalysis 9/28; compiled by STT

What the Paid Section Covers

  • Gap breakdown: how Lumentum's gap widened from 30% to 70%, why its booking horizon moved from 2028 to early 2029, and what a three-year build cycle means for 2027

  • Power-tier breakdown: the year 70–100 mW, 100–300 mW and 300 mW-plus lasers each flip to oversupply

  • Timeline breakdown: the policy clock and the shortage clock on one axis, and why 2029 is where they meet

  • What A-shares should really fear: Chinese CW laser makers are expanding 2.5–3x, so 2027 low-power laser pricing is the nearer risk

  • Taiwan supply chain, link by link: September revenue and roles for LandMark, TrueLight, Luxnet, Elaser, EZconn and three GaAs names, narrowed to two real bottlenecks

  • Testimony from the other side: what the sold-off Chinese laser-chip makers say themselves

  • 8 counterarguments and 6 signals to watch (including 3 verifiable STT calls)

──────────────────────────────

That's the summary of this piece.

STT's full analysis (the gap, power-tier and timeline breakdowns, the Taiwan supply chain table, counterarguments and verifiable calls) is in the premium section, with 6 original charts.

👉 Read the full analysis on vocus (in Chinese): https://vocus.cc/article/6acc6496fd89780001de6659


──────────────────────────────

Comments

Rated 0 out of 5 stars.
No ratings yet

Add a rating
bottom of page