What NERC’s 2026 Summer Reliability Assessment tells energy executives about where the power grid actually stands — and why the risks that remain are the ones that matter most.
On May 19, the North American Electric Reliability Corporation released its 2026 Summer Reliability Assessment — the annual report card on whether the U.S. and Canadian power grid can handle summer heat. The headline is reassuring: every assessment area in North America is expected to have adequate resources to meet normal summer peak demand. The number of regions carrying elevated risk dropped from six last year to three. On paper, this summer looks better than the last several.
Read the full report, however, and a more complicated picture emerges. The risks that defined last summer — inadequate reserve margins, insufficient generation capacity — have not been eliminated. They have shifted. The threat has moved from “do we have enough power at all” to “do we have the right kind of power at the right moment.” For energy executives, general counsel, and businesses whose operations depend on reliable electricity, that distinction matters enormously.
What Has Actually Improved
The improvement is real and significant. The North American grid added more than 58 gigawatts of new generation capacity since summer 2025 — including 16.4 gigawatts of solar, 14.7 gigawatts of battery storage, 6.7 gigawatts of natural gas, and 1.6 gigawatts of wind. FERC staff, reporting on the same day NERC released its assessment, confirmed that U.S. generating capacity increased by approximately 75 gigawatts this summer compared to a year ago, with about 26 gigawatts of that growth in Texas alone. Power plant retirements, meanwhile, slowed to about 8 gigawatts — well below the pace of additions.
In Texas specifically, ERCOT’s overall resource picture has improved materially. The EIA’s October 2025 Short-Term Energy Outlook projected ERCOT demand would rise 14 percent in the first nine months of 2026 compared to the same period in 2025 — the fastest growth of any U.S. grid operator — and while subsequent EIA forecasts revised that figure downward as large load interconnections came online more slowly than expected, new solar, wind, and battery capacity has kept pace with actual demand growth. NERC now considers the ERCOT area-wide outlook adequate for normal summer conditions, a meaningful improvement from 2025 when parts of Texas were flagged as elevated risk.
There is also an emergency backstop in place. The Department of Energy has used its authority under Section 202(c) of the Federal Power Act to keep multiple coal and gas plants running that would otherwise have retired or been mothballed — a practice the DOE has deployed more than 40 times since May 2025, stalling the retirement of at least 4.4 gigawatts of coal capacity as of April 2026. Those plants were not even counted among NERC’s anticipated resources — they represent additional buffer capacity for the critical spring-to-summer transition period.
Where the Risks Have Shifted
NERC’s director of reliability assessments was careful to frame the improved numbers accurately: “The improved conditions we’re seeing shouldn’t be interpreted as saying that overall reliability risk is declining.” That caveat reflects three specific concerns that run through the 2026 assessment.
The first is the timing problem. Heat events are arriving earlier in the season than they were a decade ago, and NERC’s analysis now identifies spring grid stress as a growing concern in certain regions — a category that barely appeared in utility planning frameworks ten years ago. The assumption that June is preparation time and July and August are the risk months no longer holds uniformly. Early summer heat can coincide with planned maintenance outages — work scheduled during what operators expected to be a lower-demand period — reducing available capacity precisely when it is needed most. The EIA’s 2026 summer energy outlook projects cooling degree days will run above the 10-year average across most of the country, with the heaviest burden in regions where grid capacity is already tightest.
The second concern is resource mix. Most of what has been added to the grid — solar, wind, and batteries — cannot be counted on with certainty at peak demand. Solar drops to zero after sunset. Wind is variable. Battery storage systems, while improving rapidly, are typically designed for four-hour discharge cycles and are not yet a substitute for dispatchable thermal generation during a multi-day heat event. NERC has been explicit that the grid still needs more firm, dispatchable capacity — generation that can be called on whenever it is needed, regardless of weather conditions. That gap is not closing as quickly as the capacity addition headlines suggest.
The third concern is large load unpredictability. NERC’s 2026 assessment specifically identifies the “unpredictability of large loads” as a new and growing reliability challenge. Data centers, hyperscale computing facilities, and large industrial operations are connecting to the grid faster than forecasting models can accurately track. Multiple assessment areas revised their load forecasts downward from mid-2025 projections specifically because large load interconnections are coming online more slowly than requested — but aggregate peak demand still increased by more than 11 gigawatts over 2025 projections. The result is forecasting uncertainty in both directions, which makes resource planning more difficult for grid operators.
The Three Regions Still at Elevated Risk
NERC designates three subregions as elevated risk for summer 2026: New England, the Canadian province of Saskatchewan, and the Pacific Northwest. Elevated risk in NERC’s framework means that resources are expected to be adequate under typical summer conditions, but could fall short during a worse-than-forecast heat wave, an unexpected loss of generation, or unusually low renewable output.
New England’s risk stems from declining firm import commitments from neighboring systems, leaving the region more dependent on non-firm supplies during high-demand periods. The Pacific Northwest faces drought-driven hydropower reductions — hydro has historically been the region’s reliability anchor, and below-normal snowpack is reducing its availability exactly when summer heat peaks. Both situations illustrate a broader point: regional grid reliability is increasingly dependent on interconnections and imports that can become unreliable precisely when they are most needed, because neighboring regions face the same heat events simultaneously.
In Texas, the overall picture has improved but one localized risk remains. NERC flags the far west Texas zone — the area served by the western edge of the ERCOT footprint — as subject to load disruption risk when solar and wind output is low and transmission constraints limit imports from the rest of the grid. That is a specific, known vulnerability in a state otherwise showing improvement.
What This Means for Energy Executives and Businesses
For energy companies operating generation, transmission, or distribution assets, the 2026 assessment is a planning document as much as a report card. The shift from capacity shortfall risk to timing and resource mix risk means that the questions worth asking have changed. The concern is less often “do we have enough generation” and more often “is our generation available at the right moment, and are our commercial agreements structured to reflect that?” Power purchase agreements and capacity contracts written when solar and battery storage were less prevalent carry different performance and availability assumptions than the grid now demands.
For industrial and commercial businesses that are large electricity users — manufacturers, data centers, hospitals, petrochemical facilities — the NERC assessment is a signal to review demand response enrollment, backup generation readiness, and interruptible service agreements before the first major heat event of the season. NERC’s own data show that the consequences of grid stress events have become more expensive as the economy has grown more electricity-dependent. The businesses that manage through summer grid stress best are those that have tested their contingency protocols before they need them.
For general counsel and legal teams, the assessment raises a more specific set of questions. Force majeure provisions in energy contracts apply equally to domestic grid reliability events as they do to international supply disruptions. A grid emergency that causes an operator to curtail, interrupt, or modify performance under a power supply agreement may or may not constitute a force majeure event depending on what the contract says and how the event is characterized. Texas SB 6’s remote disconnect provisions for large loads add another layer: a data center or large industrial customer that is curtailed by ERCOT during a firm load shed event needs to understand how that curtailment interacts with its commercial obligations to its own customers and counterparties.
The Broader Picture
The 2026 Summer Reliability Assessment reflects a grid in genuine transition. Record capacity additions, faster interconnection timelines, and improving storage technology are all real progress. But the grid is also absorbing demand growth faster than most forecasters anticipated, integrating a resource mix that performs differently than the thermal-dominated grid of twenty years ago, and facing weather patterns that are extending and intensifying the reliability stress season.
NERC’s director put the tension plainly: the grid needs more firm and dispatchable resources to maintain reliability and meet rising demand. The additions being made today — predominantly solar and storage — are necessary but not sufficient. That gap between what is being built and what the grid ultimately needs is the central energy infrastructure challenge of the next decade, and its resolution will shape electricity costs, project economics, and operational risk for every business that depends on the grid.
Two statistics from NERC’s broader work put the stakes in sharper focus. NERC’s January 2026 Long-Term Reliability Assessment projects U.S. summer peak load will increase by approximately 224 gigawatts between 2025 and 2035 — nearly 70 percent higher than the prior year’s forecast and the fastest acceleration in demand since NERC began tracking reliability data in 1995. At the same time, NERC’s 2026 Summer Reliability Assessment found that updated modeling of data center behavior during peak periods allowed ERCOT to reduce its summer demand forecast by 1.9 gigawatts, because data centers — when properly incentivized and equipped — can flex their loads rather than draw at maximum capacity during grid stress events. That flexibility is real, growing, and increasingly factored into how grid operators plan for summer. It is also a preview of how the relationship between large industrial power users and the grid is likely to evolve: less passive consumer, more active participant in reliability.
