Subway closed over 700 US stores as franchise model faces strain

Subway closed a net 729 U.S. stores in 2025, bringing its domestic footprint below 19,000 locations, even as it opened 499 units and signed 93 franchise agreements. Net income rose to $688 million from $397 million last year, but total franchise revenue fell more than 6% to $767 million and about 800 stores were temporarily closed at year-end. The article points to ongoing pressure on the franchise model despite an expected ~100 new openings next year.

Analysis

This is less a Subway-specific story than a signal that the low-ticket franchise model is becoming less financeable at the margin. When unit-level economics deteriorate, the first-order effect is store shrinkage; the second-order effect is a tightening of credit and labor availability across the entire value sandwich/QSR cohort, because landlords and lenders start pricing in higher turnover and weaker resale value. That tends to favor operators with stronger brand equity, higher average unit volumes, and more disciplined franchisee balance sheets, while punishing concepts that rely on opening new boxes to offset maturing-store decay.

The important tell is that closures are now being “recycled” into reopenings, which implies the network is not structurally expanding so much as being re-underwritten. That matters for suppliers, equipment vendors, and local commercial real estate: refurbished-capex cycles usually generate less incremental demand than true net-new buildouts, so the upside to vendors is capped even if headline openings look stable. In other words, the ecosystem gets more cash-conservative, not more growthy, and that usually compresses multiple expectations across the segment.

The apparent disconnect between shrinking footprint and improved bottom-line profitability likely reflects a mix of fee economics, refranchising discipline, and lower corporate overhead rather than durable franchise health. The market may be underestimating how fragile that earnings quality is if closures continue for another 2-3 quarters: once franchisees see less confidence in the system, same-store traffic can roll over faster than reported unit counts suggest. The near-term catalyst set is a potential reset in 2026 guidance around openings and franchisee support; if management leans harder into incentives, margin quality could weaken again even as the store count stabilizes.

Contrarian angle: the overhang may be more about pruning bad locations than brand collapse, so the bearish read can be overstated if surviving stores are better monetized. But the key risk is that a smaller base reduces network effect and local convenience, making it harder to defend against regional sandwich chains and convenience stores that can steal share with faster service and better site density. Over a 6-12 month horizon, watch whether reopening rates remain above closures; if not, the model is entering a slow-burn contraction phase rather than a temporary cleanup.

Key Decisions for Investors

• Short SBUX as a thematic hedge only if the market is pricing franchise normalization; use a 3-6 month horizon and keep size modest because the name can look optically cheap on headline earnings while underlying unit economics deteriorate.
• Relative-value pair: long QSR-quality multi-unit operators with stronger franchisee health, short lower-productivity sandwich/franchise exposure; express via a basket or the weakest operator in your coverage universe over 6-9 months.
• Long commercial real estate service/maintenance vendors with exposure to remodel activity rather than new-build capex; the thesis is that reused/rehabbed units create work, but with lower volatility than expansion-driven suppliers.
• Buy downside protection on any restaurant-franchise lender or ABS exposure tied to sub-$500k unit-volume concepts; 6-12 month protection is attractive if closures continue and refinancing becomes more punitive.
• If management signals a stabilization in reopening/closure ratios, fade the bearish trade quickly: cover shorts on any evidence that 2026 is a cleanup year rather than the start of a multi-year contraction.