Anthropic Needs Gigawatts of Power It Doesn't Have. This Dividend-Paying Industrial Sells It.

NextEra Energy Resources reported 18% adjusted EPS growth and expanded its generation and storage backlog by 3.6 GW to more than 35 GW, positioning its renewable, gas, battery, and nuclear portfolio to supply AI data-center demand. The company secured up to $1.9B in DOE loans to restart its 615-MW Iowa nuclear plant by early 2029. Management targets roughly 10% dividend growth this year, 6% annual dividend growth through 2028, and at least 8% annual EPS growth through 2035; the shares' recent decline has lifted the dividend yield to about 3.2%.

Analysis

NEE's differentiated exposure is not simply incremental data-center load; it is the ability to bundle generation, storage and grid-facing development under long-dated contracts. That should make its backlog more financeable than merchant-power peers, but the equity upside depends on contract economics and rate-base/renewables development returns rather than headline gigawatts. AI demand improves NEE's negotiating leverage for firming and interconnection costs, potentially lifting returns on batteries and gas-backed renewable portfolios over the next 12-36 months.

The Iowa restart is strategically valuable as a firm-power option, but it should not drive near-term earnings: execution, relicensing, refurbishment scope and ultimate power-offtake pricing remain the key variables until the planned 2029 start. DOE financing reduces funding risk but does not eliminate construction-cost, outage-duration or regulatory risk. More immediately, NEE remains rate-sensitive: a renewed rise in long-end Treasury yields can compress its valuation despite improving operating fundamentals, while lower yields would disproportionately rerate its dividend-growth profile over the next 1-3 months.

The non-obvious competitive effect is that hyperscalers may increasingly prefer 24/7 clean-power contracts over stand-alone renewable PPAs. That favors integrated developers such as NEE and Constellation Energy (CEG), while pressuring pure-play solar developers whose projects lack storage, transmission access, or dispatchable backup. Consensus may over-credit nuclear scarcity: restarting an existing plant is less technically speculative than SMR development, but the value accrues only if a creditworthy offtaker pays a premium for carbon-free capacity rather than treating it as regulated-system supply.

Maintain a constructive but valuation-disciplined stance. The thesis is falsified by backlog conversion weakening, Energy Resources margin/earnings growth decelerating materially, a higher capex-to-cash-flow funding gap, or 10-year Treasury yields moving sustainably higher; for the nuclear leg, a material schedule or budget revision would negate the scarcity premium.

Key Decisions for Investors

• Accumulate NEE on rate-driven weakness rather than chase AI headlines; use a 6-12 month horizon and size against a 10-year Treasury hedge. Target a rerating if long rates fall and contracted backlog conversion supports high-single-digit EPS growth; exit/reduce on a sustained rise in long yields or weaker project conversion.
• Pair trade for 6-18 months: long NEE / short a basket of less-integrated renewable developers (TAN ETF proxy) where liquidity permits. The thesis is that firm-power, storage and customer-credit capabilities command better contract pricing; stop out if solar-project pricing or policy support materially improves relative returns.
• Use CEG as the cleaner near-term nuclear/firm-power comparator rather than treating NEE's Iowa restart as a 2026-28 earnings catalyst. Monitor disclosed offtake terms, refurbishment capex and regulatory milestones before assigning incremental value to the restart.
• Set an earnings watch item: require evidence that Energy Resources growth is translating into cash generation after development capex, not only adjusted EPS and backlog additions. A widening external-financing need would cap dividend-growth credibility and warrants reducing exposure.