BCE reports second quarter 2026 results

BCE Q2 2026 operating revenue rose 1.5% YoY to $6,176M and adjusted EBITDA edged up 1.0% to $2,702M, but GAAP net earnings fell 2.3% to $629M (net earnings attributable to common shareholders down 3.6% to $558M). Free cash flow declined 9.5% to $1,042M as capital expenditures jumped 41.5% to $1,080M to support Bell AI Fabric data centers and Ziply Fiber’s FTTP build-out, while AI-powered solutions revenues (Ateko + Bell Cyber) grew 29% YoY. BCE also reported 54,883 net FTTH activations (+14.5%) and improved postpaid churn to 1.02% (down 4 bps), and it reiterated/updated 2026 guidance amid lower expected free cash flow due to higher capex for the Saskatchewan AI data centre.

Analysis

This is a classic case where operating momentum is better than equity economics. The fiber/churn/streaming improvements support the case that Bell can defend share, but the market should care more about what the next dollar of growth costs: the company is effectively swapping low-teens revenue growth pockets for a heavier capital regime that drags on distributable cash and keeps valuation anchored to yield. In other words, the story is less "earnings beat" and more "capital intensity reset."

Second-order winners are the infrastructure vendors around the buildout, not BCE shareholders. Domestic network, power, cooling, and data-center supply chains should see cleaner incremental demand if the AI fabric program is real; Celestica is a plausible beneficiary, but only if procurement turns from branding into backlog. The loser is BCE’s multiple: regulated pricing friction plus legacy decline means investors will likely discount management’s AI narrative until FCF actually inflects, not just revenue.

Near term, the stock may get a relief bid because guidance was reaffirmed and wireless retention improved, but over 1-3 months the focus should shift to capex execution, financing cost, and whether AI/data-center spend is accretive or just dilutive. The contrarian miss is that "AI" here is not automatically a margin expansion story; if the projects do not produce contracted enterprise revenue quickly, the equity is carrying the cost of a long-duration bet while bondholders and suppliers capture more of the upside. Falsifiers: a clear 2026 free-cash-flow upgrade, a meaningful cut in capital intensity, or evidence that AI-powered enterprise revenue scales faster than D&A and interest expense.

Key Decisions for Investors

• Short BCE on strength over the next 1-3 months; thesis is that the market will fade the operational beat once it focuses on capital intensity and free-cash-flow pressure. Stop if BCE raises FY free-cash-flow guidance or materially lowers capex.
• Pair trade: long CLS / short BCE for 3-6 months. If Bell’s AI-fabric spend is real, Celestica should see cleaner second-order order flow while BCE absorbs the cash burden; risk is that Bell internalizes more of the spend than expected.
• Avoid chasing BCE as a dividend value trap unless the stock de-rates enough to compensate for capex risk; use it only as a carry name if management later proves FCF can stabilize without further leverage creep.
• Set an alert on BCE free-cash-flow midpoint and capital intensity at the next update. If free cash flow trends below the guide range or cap intensity stays near the current elevated level, expect multiple compression to continue.