Coal’s decline did not stop, but it did slow

U.S. coal-fired power plant retirements fell to their lowest annual level in 15 years in 2025, according to the supplied Energy Information Administration material. Only 2.6 gigawatts of coal-generating capacity retired during the year, spread across four power plants. That was the smallest annual retirement total since 2010 and a marked slowdown from the much larger reductions seen only a few years earlier.

The number is significant because it cuts against the expectation that coal’s share of the U.S. power system would continue shrinking at the faster pace seen earlier in the decade. Instead, 2025 became a year defined by postponements and reversals. Operators delayed 4.8 gigawatts of planned retirements into future years, canceled retirement plans for two coal plants totaling 1.1 gigawatts, and also canceled planned 2027 retirements for another 1.2 gigawatts. A separate facility expected to close in 2026 pushed its retirement back to 2029.

That combination did not revive coal’s long-term growth prospects, but it did show that the path downward is more contingent than inevitable. Grid reliability concerns, replacement timing, and policy intervention all played a role in keeping older coal units available longer than previously expected.

The plants that actually retired

The supplied text identifies four retirement events in 2025. Indian River Unit 4 in Delaware, rated at 410 megawatts, retired in February. Cholla Units 1 and 3 in Arizona, totaling 383 megawatts, retired in March. The largest retirement came late in the year, when Units 1 and 2 at the Intermountain Power Project in Utah, totaling 1,800 megawatts, retired in October and November. Prairie Creek Unit 1 in Iowa, a much smaller 15-megawatt unit, retired in December.

Even within that limited group, the replacement story matters. The Intermountain retirements were partly offset by a new natural-gas combined-cycle facility rated at 1,017 megawatts that opened at the same site in late 2025. That illustrates an important reality of the energy transition. Coal exits do not automatically translate into renewable replacement. In many cases, the near-term substitute is gas, especially where grid operators and utilities are prioritizing dispatchable capacity.

So while the annual retirement total was low, the composition of replacement capacity is equally consequential. The system is changing, but not always in a straight line from coal to zero-carbon power.

Emergency federal orders changed the timeline

A major reason more coal did not retire in 2025 was federal intervention. The U.S. Department of Energy issued emergency orders under Section 202(c) of the Federal Power Act directing several coal plants to remain temporarily available in the name of grid reliability. The supplied EIA text names plants in Michigan, Washington, Indiana, and Colorado among those affected.

Most of those delayed retirements shifted from planned 2025 closure dates into early or mid-2026, based on what operators reported to EIA. In practical terms, the emergency orders bought time for aging units that otherwise appeared headed out of service. That does not guarantee long-term operation, but it shows how reliability concerns can interrupt decarbonization timelines.

This matters because retirement schedules are often treated as settled milestones in energy-transition narratives. The 2025 data suggest they are better understood as targets subject to revision when grid conditions tighten or replacement resources lag. As electricity demand becomes harder to forecast and capacity needs become more acute, policymakers may intervene more frequently to preserve plants that had been slated to close.

Coal’s recent retirement trend has softened

The supplied data also put the 2025 figure in context. Coal retirements have generally fallen since 2022, when operators retired 13.7 gigawatts, equal to about 6.5% of the fleet operating at the end of 2021. By contrast, the 2.6-gigawatt total in 2025 represented only 1.5% of year-end 2024 coal capacity.

That slowdown does not erase the long-term decline of the coal fleet, but it does suggest that the easiest retirements may already have happened. What remains may be more entangled with regional reliability, local politics, or replacement-resource constraints. The closer the system gets to the hard cases, the more likely schedules are to slip.

There is also a broader strategic lesson here. The transition away from coal is influenced not only by economics and emissions policy but also by the sequencing of infrastructure. If new generation, transmission, storage, or gas capacity is not ready in time, old plants can win a temporary reprieve. That creates a system in which announced retirements are real signals, but not final guarantees.

What the 2025 numbers really show

The headline figure, the lowest retirement total in 15 years, could be misread as a coal comeback. The supplied information does not support that conclusion. What it supports is something narrower but still important: coal’s decline can stall when grid operators and federal authorities decide reliability requires more time.

That nuance matters for energy planning. Policymakers, utilities, and investors cannot assume that retirement pipelines will unfold exactly on schedule. They also cannot assume that every coal retirement will be matched by renewable generation. In at least some cases, gas remains the bridge technology that arrives first.

The 2025 experience therefore serves as a warning against overly linear transition models. Coal is still moving out of the U.S. system, but the pace is now being shaped by more active intervention and more visible tradeoffs.

The year ahead

If the postponed closures identified by EIA move ahead in 2026, last year’s unusually low retirement total may prove temporary. But if additional reliability concerns emerge, further delays are possible. Either way, the 2025 data mark a clear shift in tone. The central question is no longer whether coal is retiring. It is how quickly, and under what conditions the federal government is willing to slow that process.

This article is based on reporting by CleanTechnica. Read the original article.

Originally published on cleantechnica.com