CIGRE 2026 : construire les réseaux électriques pour une nouvelle ère de l'électrification

Le CIGRE (session 2026) appelle à un investissement mondial d’environ 600 Md€ par an dans les réseaux électriques, alors que l’AIE estime un besoin d’~600 Md$ d’ici 2030 (contre ~400 Md$ aujourd’hui, +50%). Le point de friction majeur reste l’allongement des délais d’exécution (≈1,5 an en Chine vs 3 ans en Inde, et souvent 8-10 ans en Europe/États-Unis). L’approche “One Grid” insiste sur une coordination de bout en bout (gestionnaires, fabricants/opérateurs, régulateurs) et sur la résolution d’une pénurie de compétences (besoin de +1,5 million de travailleurs d’ici 2030).

Analysis

This is a multi-year capacity bottleneck story, not a one-day macro headline. The incremental economic rent should accrue to the few businesses that control scarce grid hardware, engineering capacity, and field execution — especially where backlog visibility and pricing power are higher than in generic power-generation or software names. The market often overpays for “electrification” narratives tied to demand growth, while the real bottleneck is throughput; that makes the supply-constrained picks-and-shovels names the cleaner trade.

Second-order winners are the companies that sit closest to transmission, distribution, and interconnection queues: grid equipment, cable, switchgear, protection systems, and EPC firms. The hidden loser set is broader than utilities: renewable developers, hyperscalers, and industrial users can all see project IRRs compressed if connection timelines and labor availability remain the binding constraint. If copper, aluminum, and skilled labor inflation stay elevated, the near-term upside for OEMs may be offset by margin pressure unless they already have strong surcharge mechanisms and long-duration contracts.

The contrarian point is that consensus may be too optimistic about equity upside from higher grid capex. Regulators can socialize costs, which preserves volume but shifts economics toward ratepayers and fixed-income holders rather than utility equities. The best falsifier is not the conference rhetoric but order-to-bill, backlog duration, and margin conversion over the next 1-3 quarters; if those do not inflect, the theme becomes a valuation trap rather than a fundamental catalyst.

Key Decisions for Investors

• Long ETN / HUBB basket on any 3-5% pullback, 6-12 month horizon: best direct exposure to grid capex with higher margin resilience than utilities; risk/reward improves if backlog growth continues but valuation is the main risk.
• Long PWR versus short XLU as a relative-value expression for capex acceleration and regulatory lag, 3-6 months: PWR should benefit from execution scarcity while utilities face slower ROE transmission; thesis fails if rate cuts sharply improve utility multiples or if PWR margins compress.
• Watchlist long ABB or GEV on confirmation of order acceleration, not on the headline alone: wait for next quarter’s book-to-bill and backlog commentary before adding; this is a 1-3 quarter catalyst, not an immediate trade.
• Small tactical long FCX as a secondary beneficiary of grid intensity, 6-18 months: grid buildout supports structural copper demand, but size should be modest because the stock already has broad macro drivers; exit if copper inventories rise and cable order books do not follow through.
• Avoid chasing renewable developers until interconnection timelines and permitting improve: if the market starts pricing faster grid completion without evidence, fade the move via an underweight in ICLN/TAN rather than a single-name short.