A Response to Julie McNamara and the Union of Concerned Scientists
By Chris Hamilton  |  President, West Virginia Coal Association

Julie McNamara of the Union of Concerned Scientists published a column this week arguing that coal-fired power is expensive and unreliable, and that the Trump administration is doing ratepayers a disservice by supporting it. The argument is wrong on both counts. It was wrong before the Department of Energy announced $850 million in new coal investment on June 4th — including two brand-new coal-fired power plants, one in West Virginia and one in Alaska, the first new American coal plants since 2013. It is wrong now. And it will remain wrong for as long as the American economy keeps moving in the direction it is moving: more energy-intensive, more technology-dependent, and far more unforgiving of the kind of intermittent, weather-dependent power that McNamara would have us bet our future on.

Two claims. Two chances to get it right. McNamara missed both.

Before dismantling those claims one by one, it is worth pausing on something more fundamental. McNamara writes under the banner of the Union of Concerned Scientists — an organization whose name invites the public to trust that what follows is grounded in rigorous, dispassionate analysis. It is not. Her column contradicts the published findings of the North American Electric Reliability Corporation, the technical determinations of the Federal Energy Regulatory Commission, the capacity modeling of PJM Interconnection, and the load forecasting of the Energy Information Administration. These are the actual scientific and engineering bodies whose work shapes how the American grid operates. When your argument about electricity reliability runs directly counter to the conclusions of NERC, FERC, PJM, and the EIA simultaneously, you are not doing science. You are doing advocacy — and dressing it in a lab coat it has not earned.

Science follows evidence. It updates when data contradicts the hypothesis. It does not begin with the conclusion that coal must go and work backward to manufacture the supporting argument. McNamara's column is wrong on its face — not just in its details, not just at the margins, but in its foundational premises. The engineers and reliability analysts who actually keep the lights on in this country have reached conclusions that flatly contradict hers. That gap does not reflect a difference of opinion. It reflects the difference between analysis and advocacy.

Claim One: Coal Is Expensive. The Math Says Otherwise.

McNamara's cost argument depends on a carefully constructed illusion. She compares the cost of building new generation — implying coal is more expensive than renewables — without ever specifying what she is actually comparing. The relevant question for ratepayers is not what it costs to build a new plant of any kind. The relevant question is what it costs to operate the plants already in service, whose construction costs were paid off years or decades ago.

Existing coal plants carry operational costs of approximately $20 to $40 per megawatt-hour. That is the number McNamara needs to compete against. Wind and solar, once you add the costs that advocates routinely omit — transmission infrastructure to reach remote sites, battery storage or gas backup for the hours the wind is calm and the sky is overcast, capacity payments to keep dispatchable plants on standby — land at $80 to $120 per megawatt-hour or more in many regions. That gap does not close by pretending integration costs do not exist. It closes when someone is honest enough to put the full ledger on the table.

Lazard's Levelized Cost of Energy report, the standard industry reference McNamara's allies cite in virtually every column of this kind, compares new construction to new construction. Lazard explicitly states its analysis does not apply to retirement decisions for existing plants. Using it to argue that West Virginia ratepayers should pay to shut down and replace fully depreciated coal generation is an analytical error so basic that anyone who makes it either does not understand the tool they are using or has decided the conclusion before consulting the evidence.

And here is the point McNamara cannot answer: if coal were truly as expensive as she claims, no rational actor would fund new coal plants. The Trump administration just committed $850 million to do exactly that — including a 1.6-gigawatt plant in Grant County, West Virginia, and a 1.25-gigawatt plant in Anchorage, Alaska. You do not spend that kind of federal capital on a fuel that cannot compete on cost and reliability. The market knows it. The engineers know it. The Department of Energy knows it. McNamara is the last one in the room still pretending otherwise.

If coal were as expensive and unreliable as McNamara claims, the federal government would not have just committed $850 million to build two new coal plants. Someone is not being straight with the American public — and it is not the Department of Energy. 

Claim Two: Coal Is Unreliable. Ask NERC. Ask FERC. Ask PJM.

McNamara calls coal plants old, outdated, and unsuited for a modern grid. She does not cite the North American Electric Reliability Corporation, the independent federal body whose institutional purpose is to assess whether North American electric infrastructure can keep the lights on. NERC has spent three consecutive years issuing warnings — escalating in urgency each time — that accelerated retirements of dispatchable coal and gas generation are creating capacity shortfalls that threaten grid reliability across multiple regions. NERC does not issue those warnings because coal plants are the problem. It issues them because coal retirements are the problem.

PJM Interconnection, the grid operator for West Virginia and twelve other states serving 65 million people, has watched its capacity auction prices spike as dispatchable reserves tighten. The Federal Energy Regulatory Commission has intervened repeatedly to prevent premature retirements of coal and gas plants that grid operators identified as critical to reliability. The Department of Energy's own grid emergency orders have directed utilities to keep coal units operational past scheduled retirement dates. FERC, NERC, PJM — these are not coal industry advocacy organizations. They are the referees. And the referees keep throwing flags at the same play: retire dispatchable baseload too fast and the grid becomes dangerously fragile.

McNamara's reliability critique rests on the observation that coal plants require maintenance and that some are aging. This is true of every power plant ever built. The difference is that scheduled maintenance is a known variable — grid operators plan around it with days or weeks of notice. Renewable intermittency is an uncontrollable one. A wind farm that generates nothing for 72 hours during a winter high-pressure system cannot give the grid operator advance warning. A solar array blanketed by a January ice storm across five states does not call ahead. Coal can be planned around. Weather cannot.

Texas in February 2021 is the case study that permanently settles this argument for anyone willing to look at it honestly. ERCOT, the Texas grid, came within minutes of a complete statewide collapse. Extended cold disabled significant wind capacity, reduced solar output, and sent demand surging at the same moment. The resulting blackouts killed hundreds of Texans and caused an estimated $295 billion in economic damage — the costliest weather event in Texas history. That was not a failure of coal reliability. That was the direct, predictable, forewarned consequence of dismantling dispatchable generation without adequate replacement. Every grid engineer in the country watched it happen and knew exactly why.

Scheduled maintenance is a known variable. A wind farm that goes dark for three days during a winter high-pressure system is not. Grid operators can plan around the first. They cannot plan around the second.

The Demand Curve McNamara Refuses to Acknowledge

There is a third dimension to this debate that McNamara does not address at all, because addressing it would collapse her argument entirely. The American economy is undergoing a fundamental structural transformation — and it is moving in precisely the direction that makes coal's attributes more valuable, not less.

Data centers now consume approximately 4 percent of total U.S. electricity and that figure is rising steeply. The artificial intelligence buildout — the infrastructure arms race now underway among every major technology company in the world — requires power that is available 24 hours a day, 7 days a week, 365 days a year, without interruption, without weather dependence, without a single unplanned outage. A data center running a large language model does not have a backup mode for cloudy days. A semiconductor fabrication plant cannot pause production because the wind died. A steel mill reshoring from overseas cannot absorb a three-day blackout in its economic model.

Heavy industry is returning to the United States. Semiconductor manufacturing, battery production, steel, aluminum, chemicals — these are the industries the federal government has spent billions in incentives to attract back to American soil. Every one of them runs on large, continuous, reliable loads that intermittent generation physically cannot serve. Grid operators are not speculating about this demand growth. They are already revising their load forecasts upward at rates not seen in decades, driven by announced industrial projects and data center development that is either under construction or fully permitted.

The grid that McNamara wants — heavily weighted toward variable renewables, stripped of dispatchable baseload — is a grid designed for a 1990s electricity demand profile. The grid America actually needs is one capable of powering a $30 trillion economy that runs on computation, precision manufacturing, and continuous industrial processes. Coal, along with natural gas and nuclear, provides the foundation that grid requires. Wind and solar have a role to play at the margins. They cannot carry the load — not now, and not in any foreseeable future that grid planners, NERC, FERC, or PJM are actually modeling.

The Real Cost of Getting This Wrong

McNamara frames this debate as a choice between expensive, unreliable coal and affordable, modern renewables. The actual choice is starker than that. Nations and regions that have moved aggressively to dismantle dispatchable generation in favor of intermittent renewables have not achieved cheaper, cleaner electricity. They have achieved higher rates, recurring reliability crises, and an increasingly two-tiered energy economy in which industrial users absorb crippling costs and residential customers face bills that consume an ever-larger share of household income.

Energy poverty is not an abstraction. It is what happens when ideology overruns engineering — when advocates who will never miss a meal and will never see their electricity shut off make decisions that price working families out of reliable power. West Virginia families know something about bearing costs that others impose. They have watched their communities hollowed out by energy policies designed in think tanks and written by people who have never set foot in a coal county. They are not interested in another round of promises about a transition that arrives as job losses, plant closures, and electric bills they cannot pay.

Coal is cheap when you count all the costs. Coal is reliable when you understand what reliability actually requires. And the demand for exactly what coal provides — firm, dispatchable, around-the-clock power — is growing, not shrinking, as the American economy rewires itself for the next century. McNamara is arguing for a policy that would strand that economy on an intermittent foundation it cannot support. The people who would pay the price for that mistake are not the ones writing the columns. They are the ones reading them.