Ericsson (ERIC) Q2 2026 Earnings Call Transcript

Ericsson reported Q2 net sales of SEK 52.7B (down 6% reported, -1% organic), while adjusted gross margin rose to 48.4% and adjusted EBITA margin slipped to 13.1%. Profitability was pressured by FX and lower volumes (net income SEK 4.1B vs SEK 4.6B YoY; EPS SEK 1.22 vs SEK 1.37), though Cloud Software & Services EBITA margin improved to 12.4% (from 9.6%) at a management-described new high. Management flagged input/component cost inflation (notably semiconductors) and guided Q3 Networks adjusted gross margin to 48%-50% (down slightly), alongside Q3 Networks sales growth expected above the seasonal average; it also returned SEK 8.2B to shareholders via repurchases and dividends.

Analysis

The near-term setup is more about cash conversion than headline profitability. The quarter showed that margin can be defended, but the inventory build means the business is effectively pre-funding Q3 shipments while component inflation is still moving through the pipeline; that is a classic setup for earnings quality to lag reported margins over the next 1-2 quarters. The immediate risk is that investors anchor on “stable margins” while free cash flow and working capital deteriorate before pricing actions fully offset the cost curve.

Competitive dynamics are nuanced: Ericsson’s pricing reset likely propagates industry-wide, but the benefit is asymmetric because only the highest-quality vendors can push through increases without sacrificing share. That argues for near-term share gains by low-cost or highly localized competitors in certain tenders, but the bigger second-order effect is slower buying cycles from operators if capex budgets are re-forecast around higher network gear prices. The AI/uplink narrative is real optionality, yet it is still a 6-18 month thesis; it does not justify paying up today unless order data starts to inflect.

The contrarian view is that consensus may be underestimating how much of the “AI-RAN” story is already in the stock while underestimating the drag from long-cycle contract renegotiations and rollout mix. In other words, the upside case is real but delayed, while the downside is mechanical and near-term: cash flow, inventory, and gross margin phasing. I would treat any post-earnings strength as a better opportunity to fade than to chase, unless Q3 commentary proves that price increases are already offsetting component inflation faster than expected.

Key Decisions for Investors

• Short ERIC into strength over the next 1-3 weeks, targeting a fade after the Q3 inventory-led shipment boost; risk/reward improves if the stock rallies on AI narrative without confirming FCF recovery. Falsify if Networks gross margin holds above 50% and FCF normalizes materially by Q3.
• Buy ERIC put spreads 2-4 months out rather than outright puts to express the view that margin pressure and working-capital drag show up with a lag; best entry is after any relief rally. Falsify if customer price renegotiations are visibly offsetting component inflation in Q3 guidance.
• Pair trade: long NVDA / short ERIC for a relative AI-infrastructure expression, on the thesis that the market is pricing Ericsson’s AI optionality more quickly than the monetization path. This works only if edge-compute enthusiasm remains high while RAN order growth stays flattish.
• Avoid adding to long ERIC until Q3 cash conversion and inventory unwind are visible; watch for FCF back above roughly SEK 2-3 billion and a clear decline in inventory days before re-underwriting the long case.
• If you want a lower-risk relative-value expression, long a basket of semiconductor beneficiaries versus ERIC on a 6-12 month horizon, because component suppliers capture the inflationary spend immediately while Ericsson’s pass-through remains contractually delayed.