Eutelsat shares fall as Q3 revenue slips on weaker video business

Eutelsat reported third-quarter revenue of €293 million, down 2.3% year on year, as a 5.6% decline in reported operating-vertical revenue was driven by a 15.6% drop in Video revenue to €128.0 million amid sanctions on Russian channels and contract terminations. Connectivity revenue rose to €155.7 million, with LEO revenue up 65.0% like-for-like to €62.2 million, and the company reaffirmed fiscal 2025-26 targets including ~50% LEO growth, EBITDA margin slightly below last year, and gross capex of about €900 million. The company also completed a €1.50 billion senior notes offering and expects net debt-to-EBITDA of around 2.7x by fiscal year-end.

Analysis

The market is likely underestimating how much of this headline is about China policy optionality rather than a one-day semiconductor tape reaction. If Huang’s presence on the visit lowers near-term escalation risk, the first beneficiaries are not just Nvidia but the whole US AI hardware stack: HBM suppliers, advanced packaging, and the foundry ecosystem should see a lower probability of abrupt order deferrals or licensing delays over the next 1-2 quarters. The key second-order effect is that “restricted but not frozen” China demand can still support premium mix for compute vendors even if unit growth remains capped.

The bigger signal is that geopolitical noise can now move NVDA more through regulatory path dependency than fundamentals. Any thaw in dialogue raises the value of China-specific product skus, while any disappointment after the visit would likely hit sentiment fastest in the most China-levered semis, not the broader index. That asymmetry argues for trading the policy headline with options rather than outright cash exposure, because the underlying earnings sensitivity is modest versus the multiple sensitivity to export-control headlines.

Contrarian take: the move may be too small if the market assumes this is merely symbolic. CEO-level engagement on a China trip can be read as a softening of the probability distribution around export enforcement, which matters more than a single revenue quarter. Conversely, if investors extrapolate a broad reopening, that’s too aggressive—structural US restrictions remain the binding constraint, so any rerating should be tactical and time-limited rather than a thesis break.

Key Decisions for Investors

• Buy NVDA weekly or 1-2 month call spreads on any intraday pullback if the market fades the headline; use this to express upside from policy de-escalation while capping premium outlay. Risk/reward favors convexity because the catalyst is binary but the downside is limited to paid premium.
• Go long a basket of NVDA supply-chain beneficiaries (e.g., ASML, AMAT, TSM) against a short of a China-sensitive semi laggard for 4-8 weeks; the trade captures lower export-control risk without paying full multiple for NVDA itself.
• If NVDA opens stronger and implied vol stays elevated, sell out-of-the-money put spreads 30-45 days out to monetize the headline premium; thesis only works if you believe policy news improves sentiment faster than fundamentals.
• Avoid chasing common-stock longs in NVDA after an initial gap-up greater than ~3%; instead, wait for the next policy headline or earnings revision cycle. The near-term upside is more about multiple expansion than estimate changes.
• Monitor Chinese ADR semis and networking names over the next 1-2 sessions; if they do not confirm the move, fade the broad read-through and keep exposure concentrated in the most liquid AI names.