Aug 4, 2026 earnings call — Japan silicon photonics capacity expansion and records
They guided third-quarter midrange revenue to be $520 million, representing an annualized revenue run rate of above $2 billion.
Revenue $0.5B (est $0.5B) · EPS $0.88 (est $0.76) · Next-quarter revenue outlook $0.5B
Key quotes
"3 weeks ago, we announced a dual-track 300-millimeter capacity strategic expansion in Japan for our silicon photonics, silicon germanium and advanced optical packaging capabilities, having gained the support of the Government of Japan through the Ministry of Economics, Trade and Industry, METI."
"Under the updated model, we target $3.6 billion in annual revenue, which is $760 million higher than the February model target, reflecting 27% revenue increase."
"We announced customer contracts representing approximately $1.3 billion of silicon photonics revenue for 2027 with even higher growth for 2028."
Analyst Q&A 5
Japan capacity, bookings, and growth outlook
Q (Cody Grant Acree · Benchmark StoneX): He asked if the SiPho bookings number for 2027 (the $1.3 billion) was updated and whether current visibility lets them sustain or accelerate growth. He also asked about the scale and timing of the Japan Track 1 and Track 2 projects and whether the 4x includes Fab 6.
A: Management said they did not update the $1.3 billion bookings number and clarified the Q4 start rate equals 3x the Q2 shipment (680 times 3 as capacity). They expect to reach the 2028 financial model run rate (stated as $3.6 billion in the script) and said Track 1 assumes 85% utilization. Track 2 is planned to quadruple 300-millimeter capacity, budgets are approved, target tool install and function by Q4 2028, and the 4x includes Fab 6.
"I stated in the script that the Q4 start rate, which would be fully realized in Q2 '27 revenue was 3x higher than the Q2 shipment." — Russell Ellwanger (CEO)
NPO impact and manufacturing footprint
Q (Mehdi Hosseini · SIG): He asked if NPO products increase content per transceiver or create synergy with stacked PICs, and whether U.S. customers will be mainly served from Japan or U.S. sites. He also asked when RF mobile revenue would stabilize as they move to 300-millimeter.
A: Management said selling NPO does not increase content per transceiver in the way asked; NPO would often be sold instead of pluggable and sometimes both, while true PIC stacking for next-generation would be positive and centers on advanced packaging (die-to-wafer and wafer-to-wafer bonding). They said Phase 1 focuses on SiPho expansion and SiGe growth is more in Phase 2; they are increasing capacity in Newport Beach and San Antonio while Japan gives geopolitical neutrality and high-quality workforce. On RF mobile, they would not call it "stabilize" but expect Fab 10 to be very full by Q2–Q3 of next year with the largest 300-millimeter RFSOI manufacturing they've done.
"To the extent I understand your question, I would say no." — Russell Ellwanger (CEO)
200mm vs 300mm mix, NPO, and packaging role
Q (Richard Shannon · Craig-Hallum Capital Group): He asked how much of capacity after Track 2 will be 200-millimeter versus 300-millimeter, how meaningful NPO will be to SiPho revenue in 2027, and how much of SiPho revenue will be packaging-related long term.
A: Management said by mid '27–mid '28 most growth in SiGe and especially SiPho will be on 300-millimeter. They intend to add at least 20,000–25,000 wafer per month silicon photonics capacity (which could go higher) and said the predominant portion of the 4x growth will be silicon photonics, then silicon germanium. On NPO, they said it will be significant — not single digits but in the tens of percent of shipments, especially in the second half of 2027. For packaging, they said packaging will be an enabler rather than a separated packaging revenue line; they expect several tens of millions of revenue next year from integrated laser (chip-to-wafer bonding) and will bring that capability in-house.
"As a minimum, we would intend to be adding 20,000, 25,000 wafer per month silicon photonics capacity, that can go much, much higher than that." — Russell Ellwanger (CEO)
Industry capacity additions and differentiation
Q (Cody Grant Acree · Benchmark StoneX): He asked how to view industry supply/demand given peers adding capacity over the next 12–24 months and what metric best shows Tower's leadership.
A: Management said supply is increasing but they hold strong customer qualifications and contracts through '28 and focus on keeping full or majority share with lead customers. They emphasized speed and multi-generation programs, often with exclusivity that gives lead customers a head start. For differentiation, they pointed to technical figures of merit — especially insertion loss — and leadership in next-generation development.
"Insertion loss is probably one of the biggest things that one could look for." — Russell Ellwanger (CEO)
Supply chain and biggest model risk
Q (Lisa Thompson · Zacks Investment): She asked if they have any material shortages, mentioning indium phosphide, and which assumptions create the most variability in their business model (expenses, timing, or prices).
A: Management said big picture they have no broad supply shortages; they were under a crunch for indium phosphide but addressed it with a contract with IQE and multiple steps to increase substrates. Oren said the model relies on several key assumptions: selling price per wafer, cost and timing of installation and qualification, and 85% fab utilization — and that selling price per wafer usually has the biggest range of outcome because price flows directly to margin.
"Usually per wafer because if you have more or less quantity, so you have more or less variable costs associated with that." — Oren Shirazi (CFO)