Celestica: 2027 Could Unlock The Next Growth Phase

Celestica is expected to enter 2027 with 10 active 1.6T programs, continued 800G networking growth, and approaching production of AMD Helios and OpenAI racks. Management expects 2027 revenue growth to exceed 2026's 65%, supported by multibillion-dollar AI programs and long-term capacity commitments. The analyst's base case projects $31 billion in revenue and $16 in adjusted EPS, implying a $450 price target and roughly 24% upside.

Analysis

CLS is shifting from a cyclical electronics-manufacturing multiple toward an AI-infrastructure execution story, but the key underwriting issue is not top-line opportunity; it is whether program mix can sustain margins as hyperscalers exert purchasing power. Long-duration capacity commitments reduce utilization risk and should support working-capital financing, yet they can also leave CLS holding inventory, specialized equipment, and underutilized facilities if a customer platform is delayed. The market will likely reward verified backlog conversion and operating-margin resilience more than further program announcements.

AMD’s Helios ramp creates a second-order positive for CLS because rack-level integration raises content per accelerator deployment and makes switching manufacturing partners operationally costly. However, CLS remains exposed to a concentrated-customer and platform-cycle risk: any AMD share loss to NVDA, slower OpenAI cluster deployments, or a transition toward customer-designed systems could postpone revenue without necessarily eliminating committed capex. Over the next 1-3 months, the relevant catalyst is evidence of 1.6T production qualification and incremental capacity utilization; over 6-18 months, gross-margin trajectory and cash conversion will determine whether the stock deserves a durable premium to EMS peers.

Consensus may be extrapolating revenue growth too mechanically. At a $31B revenue scenario, each 100bp miss in operating margin is roughly $0.25-$0.30 of after-tax EPS sensitivity, while a lower valuation multiple can overwhelm an EPS beat after a strong run. The better asymmetry is to own CLS only on confirmation that AI mix is improving margins and free cash flow, rather than chase estimates based solely on announced programs.

Key Decisions for Investors

• Maintain a tactical long CLS only while it holds above the post-guidance support level; add after the next earnings release if AI-program revenue converts into both raised full-year guidance and stable-to-higher operating margin. Target a 15-25% upside over 3-6 months; exit on an AI-related guidance reduction, margin decline of more than 100bp, or material working-capital build.
• Use a 6-12 month pair trade: long CLS / short a broad EMS proxy such as FLEX, sized beta-neutral. CLS should outperform if rack-level AI integration remains supply constrained; close if its valuation premium expands without corroborating FCF conversion or if FLEX demonstrates comparable AI revenue acceleration.
• For AMD, treat CLS production ramp as a confirmation signal rather than a standalone long catalyst. Add AMD exposure only if management identifies Helios-related shipments as accretive to data-center revenue growth and gross margin; a delayed platform ramp or weaker accelerator demand would impair both AMD expectations and CLS utilization.
• Monitor quarterly inventory days, capex, customer prepayments, and cash conversion at CLS. A revenue beat funded by receivables or inventory growth is a watch-item, not validation of the bull case; this is the most likely early indicator that capacity commitments are transferring balance-sheet risk to CLS.