Published September 11, 2026
By NZero
The United States is on track to consume more electricity than ever before, and the increase is arriving faster than many regional grids can comfortably absorb. According to the U.S. Energy Information Administration’s (EIA) September 2026 Short-Term Energy Outlook, total U.S. electricity sales are projected to reach 4,135 billion kilowatthours (BkWh) in 2026, up nearly 2% from 2025, and climb further to 4,211 BkWh in 2027, both record levels. This is not simply background noise in the energy markets. It points to a period where electricity costs, grid reliability, and interconnection timelines are likely to carry more operational weight than they have in years.
The EIA is explicit about where this growth is coming from: data center development and increased manufacturing activity, concentrated in the commercial and industrial sectors. Commercial electricity sales are forecast to rise 3.3% in 2026 and 2.7% in 2027, and that single sector accounts for 63% of total U.S. sales growth in 2026 and 56% in 2027. Industrial demand is also rising, up 1.6% in 2026 and 2.6% in 2027. In practical terms, this means the facilities competing for grid capacity and interconnection slots increasingly include hyperscale data centers and expanding manufacturing plants, the same category of large, power-intensive load that many industrial and commercial operators fall into. Reporting from Utility Dive on the same EIA data underscores the scale of this shift, noting that commercial-sector growth alone is now the single largest driver of the nation’s electricity demand curve. For any organization planning a facility expansion, a new production line, or additional server capacity, this growing competition for grid headroom is a scheduling and cost variable that needs to be modeled explicitly, not assumed away. The growth is also not evenly spread across the country: the West South Central region, anchored by Texas, is projected to account for roughly 20% of nationwide electricity sales growth in 2026 and as much as 40% in 2027, making it one of the tightest markets for new large-load connections over the next two years.
On the supply side, the EIA projects solar generation growing 21% in 2026 and 18% in 2027, and wind generation rising 7% and 5% over the same two years. Natural gas generation is increasing more modestly, up 2% in 2026 and 1% in 2027, while coal generation is falling 8% in 2026 and 6% in 2027 as coal production drops from 528 million short tons in 2025 to 497 million short tons in 2027. Solar capacity is expanding quickly in absolute terms, from 151 gigawatts in 2025 to an expected 222 gigawatts in 2027, but that new capacity is heavily concentrated in specific markets. ERCOT (the Texas grid) is expected to add 18 BkWh of solar generation in 2026 and 20 BkWh in 2027, while MISO adds 13 BkWh and 11 BkWh respectively over the same period. The PJM Interconnection, covering much of the Mid-Atlantic and parts of the Midwest, is projected to account for nearly 45% of total U.S. generation growth. For facilities located outside these fast-growing supply regions, the pace of new generation coming online may not match local demand growth as closely, which has direct implications for both electricity pricing and supply reliability. Coal’s declining role adds another layer to this picture: even as coal exports may rise compared with 2025, domestic power-sector coal demand keeps falling, particularly in the Northwest and MISO regions, meaning some facilities in coal-dependent markets will be relying on a generation source that is being actively phased down at the same time overall demand is climbing.
This mismatch between demand growth and localized generation capacity is not theoretical. Texas, currently the region with the strongest overall growth trajectory, has paused new data center grid interconnection requests to conduct a regulatory review, even as demand in the region continues to climb. That is a concrete example of a fast-growing market where new large loads can no longer assume automatic, fast approval to connect to the grid. Organizations evaluating new sites, capacity expansions, or major equipment additions should treat interconnection timelines as a planning variable with real uncertainty attached, not a formality. At the same time, residential electricity prices are rising alongside demand, from an average of 17.3 cents per kilowatthour in 2025 to a projected 18.2 cents in 2026 and 18.6 cents in 2027, a trend that tends to move commercial and industrial rates in the same direction, even where contract structures differ. This combination of rising base prices, tightening interconnection capacity in high-growth regions, and an increasingly competitive queue for new grid connections translates into a wider range of possible cost and timeline outcomes for any project that depends on securing new or expanded electricity service.
The next two years of U.S. electricity data point to a straightforward operational reality: demand is climbing to record levels, the generation mix is shifting toward solar and wind at real scale, and both of these trends are landing unevenly across regions. For those who plan around electricity availability and cost, this is a signal to treat energy pricing and grid interconnection timelines as active planning inputs rather than fixed assumptions. Organizations that track their facility-level electricity consumption, monitor regional grid and interconnection conditions, and build flexibility into procurement and capital planning will be better positioned to manage both the cost volatility and the capacity risk that this period of record electricity growth is bringing with it.
