Key Takeaways
Apple's incoming CEO John Ternus inherits a company facing a structural inflection in its manufacturing model. Trump's tariff regime and rising memory costs have already begun eroding gross margins — the FY2025 10-K explicitly cites tariff costs as a headwind — and consensus analyst models carry zero credit for the capex and operational drag of reshoring production from China. The setup over the next 18-24 months favors a 200-300 basis point compression in products gross margin as Ternus is forced to accelerate US manufacturing investment, a break from the 40-year China playbook that has been Apple's margin engine. The call breaks if tariffs are rolled back materially or if Ternus delays reshoring beyond 2027.
Apple's gross margin has been the company's defining competitive advantage. For decades, the outsourced manufacturing model — with a significant majority of production concentrated in China, India, Japan, South Korea, Taiwan, and Vietnam — allowed Apple to compress costs while maintaining pricing power. That model is now under structural pressure from two forces: Trump administration tariffs that have already begun hitting the P&L, and rising memory costs that are squeezing component economics across the industry.
The tariff impact is no longer theoretical. Apple's most recent 10-Q (filed January 30, 2026, covering Q4 FY2025) explicitly notes that products gross margin percentage decreased "primarily due to a different mix of products and tariff costs." The FY2025 10-K, filed October 31, 2025, goes further, warning that "the imposition of new or increased tariffs and other trade restrictions, their overall magnitude and duration, and retaliatory actions in response" could have a "materially adverse impact" on gross margins. These are not forward-looking risk disclosures; they are backward-looking admissions that tariffs are already flowing through the income statement.
The Reshoring Trap
John Ternus's first major decision as CEO will be whether to accelerate reshoring of iPhone and Mac production to the United States. The political pressure is immense: Trump has explicitly called for US manufacturing of consumer electronics, and the tariff regime creates a financial incentive to move production domestically to avoid duties. The cost structure, however, is brutal. US manufacturing labor costs roughly 3-4x China's equivalent, and the US lacks the integrated supply chain density that makes China's Pearl River Delta the world's most efficient electronics manufacturing hub. Building that ecosystem in the US — fabs for memory, assembly lines, logistics networks — requires capex measured in the tens of billions, not millions. Apple's capital expenditures have shown significant volatility in recent years, ranging from $9.447 billion in FY2024 to $12.715 billion in FY2025; a serious reshoring program would likely require incremental capex of $2-5 billion per year for 3-5 years.
Consensus analyst models do not yet reflect this cost. Sell-side 2026-2027 EPS forecasts assume total gross margins remain near recent levels, with tariff headwinds offset by "cost savings" and "favorable mix." That assumption is increasingly untenable. If Ternus commits to reshoring even 15-20% of iPhone production to the US, gross margin compression of 200-300 basis points is the base case, not a tail risk.
Memory Costs Add Pressure
The second structural headwind is rising memory costs. NAND and DRAM prices have been climbing since late 2024 as AI capex has driven demand for high-bandwidth memory. Apple's iPhones and Macs are memory-intensive; a 10-15% increase in memory component costs flows directly to cost of revenue. Unlike tariffs, which Ternus might negotiate or mitigate through reshoring, memory cost inflation is exogenous and global. It will compress margins regardless of manufacturing location.
The combination of tariffs + memory inflation + reshoring capex creates a three-year margin headwind that the market has not yet priced. Apple's stock is trading at 28x forward earnings; that multiple assumes the company can maintain current gross margin levels. If Ternus signals a reshoring commitment in the next 12 months, and gross margins compress to 43-45%, the earnings multiple should compress as well.
The Falsification
The thesis breaks if: (1) Trump tariffs are rolled back or significantly delayed beyond 2026; (2) Ternus announces a "light reshoring" strategy focused on assembly only, not component manufacturing, keeping capex under $1 billion annually; or (3) memory costs fall sharply in 2026-2027, offsetting tariff and labor cost headwinds. None of these are the base case, but all are observable within 12 months.