1 Reason Eli Lilly Stock Is Still a Buy

Shares are down 15% YTD, yet Eli Lilly has meaningfully expanded gross and operating margins (as of Q4 2025) and now exceeds similarly sized peers, driven by rapid GLP‑1 sales. Management attributes gains to improved production efficiency and the company has invested ~$55 billion in manufacturing capacity since 2020 to drive future economies of scale. Lilly also built the pharmaceutical industry's largest supercomputer with Nvidia to deploy AI (management cites potential ~5% reductions in development time/cost), which could further support margins; however, lower realized prices and mounting GLP‑1 competition remain clear downside risks.

Analysis

Scale effects in drug manufacturing are not linear — once fixed costs and validation payloads are absorbed, each incremental SKU can add 200–400bps of incremental operating margin over a multi-year horizon via higher plant utilization, better procurement terms, and fewer batch failures. Expect that benefit to be realized unevenly: new fill/finish lines and automation yield steep unit-cost declines in the 12–36 month window after commissioning, then taper as incremental volume slows.

Second-order winners are not the obvious large-cap peers but the CDMOs, single‑use equipment and automation vendors, plus GPU vendors who supply the compute for accelerated discovery. If the firm outsources peak-volume campaigns, contract manufacturers will capture margin share but also create optionality for the originator to flex capacity without incremental capex. Conversely, entrenched incumbents in legacy wafer or CPU architectures face demand-share erosion as pharma workloads shift to accelerator-class GPUs.

Key risks are abrupt realized-price compression from aggressive payor contracting or rapid entrant commoditization — both can compress gross margin within a single quarter if rebates/managed care placement change. AI productivity gains are a multi-year, lumpy catalyst; treat near-term AI narratives as optionality rather than earnings drivers. The consensus tends to binary this story into ‘structural winner’ vs ‘pricing calamity’; a more realistic outcome is a multi-year, asymmetric recovery in free cash flow driven by scale + selective price concessions, where timing (12–36 months) matters most.

Key Decisions for Investors

• Long LLY 12–24 month call spread (buy 12–18 month ATM call, sell one higher strike to finance) — asymmetric 20–30% upside capture if scale and R&D productivity materialize, max loss = premium paid; target entry when implied vol > realized vol and ahead of next quarter’s manufacturing guidance.
• Long NVDA 6–12 month calls (or buy-dated call calendar) to express increased pharma AI compute demand — reward: leverage to accelerating GPU procurement cycles; risk: high IV — hedge by selling short-dated calls or pairing with a long-dated protective put.
• Pair trade: long LLY equity (small allocation, 12–24 months) funded by short INTC (or buy INTC puts) to express divergence between GPU-driven AI capex and legacy CPU exposure — expected payoff if accelerator adoption accelerates; risk if INTC reclaims meaningful data-center share or NVDA derates.
• Buy modest 6–12 month LLY put protection (cost = ~1–3% of position) to hedge downside from sudden payor-driven pricing shocks — preserves upside optionality while limiting single-quarter drawdowns if formulary/access dynamics shift materially.