AES vs. DUK: Which Is Better Positioned for Rising Power Demand?

Rising electricity demand from AI-powered hyperscale data centers is accelerating the clean-energy transition and benefiting utilities positioned with renewable generation and long-term PPAs; AES reports 2.2 GW signed/awarded (1.6 GW with data centers) and a backlog of 11.1 GW under signed PPAs, with ~4.2 GW of data-center PPAs in operation and 8.2 GW signed in total. Zacks projects AES 2026 EPS growth of +8.44% (long-term 11.17%) versus Duke’s +6.15% (long-term 6.87%); AES shows higher ROE (18.83% vs Duke 9.98%), higher dividend yield (4.96% vs 3.57%), lower forward P/E (6x vs 17.73x) but higher debt-to-capital (78.58% vs 61.97%). Duke has filed to revise North Carolina rates to fund grid upgrades to serve large data-center loads; overall the piece favors AES as the better current pick based on growth, valuation and returns metrics.

Analysis

Market structure: Hyperscale data-center operators, IPPs and renewables developers (AES-style) are primary beneficiaries as PPAs shift long-duration demand to project owners; merchant fossil generators and small municipal utilities without scale are losers as baseload shifts to contracted renewables. AES’s 11.1 GW PPA backlog and 8.2 GW data-center agreements imply greater pricing power for long-term contracted cash flows, while regulated utilities like DUK earn through rate-base recovery but face political/regulatory timing risks. Net effect is higher capex across the sector and upward pressure on wholesale power and grid-renewal capex for 3–7 years, tightening availability of shovel-ready capacity in key nodes.

Risk assessment: Key tail risks include adverse rate-case outcomes or new state-level restrictions on cost recovery (regulatory, 5–30% downside shock to utility equity), counterparty PPA defaults during tech cycles, and a 200–300 bps sustained Fed rate shock that raises AES’s refinancing costs given 78.6% debt-to-capital. Near-term (days–weeks) risks are regulatory filings and earnings beats/misses; medium-term (3–12 months) are construction execution and PPA commissioning; long-term (3–5 years) center on commodity/battery cost declines that could compress merchant returns. Hidden dependencies: AES relies on hyperscaler demand concentration (4.2 GW operating data-center PPAs); loss of a single large counterparty could disproportionally dent near-term cash flow.

Trade implications: Direct: consider a modest long in AES (capitalizing on P/E 6x, 4.96% yield) sized 2–4% NAV with a 12-month horizon and 20% stop; offset credit risk by avoiding unsecured AES bonds. Relative value: pair trade long AES / short DUK (1–2% NAV each) to capture AES’s ~11% long-term EPS growth vs DUK ~6.9% while hedging macro rates. Options: buy a 9–12 month AES call spread to cap premium (cost-limited upside) and buy protective puts on AES if leverage-sensitive rates rise >150 bps; buy DUK 6–9 month puts if NC rate-case headlines are unexpectedly negative.

Contrarian angles: Consensus favors AES but may underprice execution and refinancing risk from high leverage; a 200–300 bps hike in borrowing costs could erase valuation gap versus DUK quickly. Conversely, DUK’s lower leverage (62% debt-to-capital) and regulated earnings make it a safe-haven if political/regulatory outcomes favor accelerated recovery — DUK could outperform on a conservative rehabbing of allowed returns. Historical parallels: past renewable build waves (2015–2019) produced temporary oversupply and margin pressure 12–24 months post-capex; monitor utilization and merchant price signals to avoid frothy entry.

Key Decisions for Investors

• Establish a 2–4% NAV long position in AES (ticker AES) within 30 days to capture P/E 6x rerating and 4.96% yield; add on pullbacks ≥10% or if backlog increases by ≥1 GW q/q; set a hard stop at −20% and a 12-month target total return of 15–25%.
• Initiate a 1–2% NAV pair trade: long AES vs short Duke Energy (DUK) for 6–12 months to express the EPS growth differential (AES ~11% vs DUK ~6.9%); rebalance if interest rates move >150 bps or NC rate-case decision favours DUK (close short leg).
• Buy a cost-limited 9–12 month AES call spread (size 0.5–1% NAV) to participate in upside while capping premium; simultaneously purchase AES downside protection (puts) if 10-year Treasury yield rises >150 bps from current levels.
• Overweight utility IG credit allocation to Duke Energy senior bonds (up to +3% of fixed-income sleeve) versus AES corporate bonds due to AES’s 78.6% debt-to-capital; only increase equity exposure to DUK after NC rate-case resolution (expect decision window 4–9 months).