The argument about energy in American politics is almost entirely about the wrong things.
The Pattern
Trace the major American economic expansions backward and you find the same structure each time: a new energy system that dramatically reduced the cost of doing everything else, which unlocked productivity growth across the entire economy, which produced an expansion that looked — in retrospect — almost inevitable.
The industrial expansion of the late 19th century was built on coal and steam. The post-war boom was built on oil and the infrastructure — highways, suburbs, supply chains — that cheap oil made possible. The information economy of the 1990s was built on electricity, specifically the cheap, reliable electricity that allowed data centers, server farms, and the internet's physical infrastructure to exist at scale.
Each transition produced enormous disruption. Each produced, on the other side of the disruption, a larger economy than the one before it. Each was politically contested at every stage, with the incumbent energy system's beneficiaries fighting the transition and the emerging system's advocates overselling what was immediately possible.
We are in the middle of one of those transitions now.
What's Actually Being Built
The clean energy buildout of the past five years is the largest capital investment in American energy infrastructure since the interstate highway system. The numbers are large enough that they are difficult to process without context.
The Inflation Reduction Act of 2022 committed approximately $370 billion in climate and energy investment — the largest single federal investment in energy infrastructure in American history. Since its passage, private capital has followed: clean energy investment in the United States reached $303 billion in 2023, according to BloombergNEF, up from $105 billion in 2019. Solar capacity additions set records in 2023 and again in 2024. Battery storage deployment is growing at a rate that would have seemed implausible five years ago.
The geography of this investment is striking. The manufacturing facilities being built to supply the energy transition — solar panels, batteries, wind components, heat pumps, electric vehicles — are concentrated heavily in states that voted for Donald Trump in 2024. Georgia, Texas, Kentucky, Michigan, South Carolina. The political debate about clean energy is national. The economic reality of clean energy manufacturing is emphatically not coastal.
This is not an accident of ideology. It is an accident of economics: land, labor, logistics, and state incentive structures have made the industrial interior the most attractive location for the factories the energy transition requires. The political contradiction — red states building green infrastructure — is real, persistent, and almost entirely absent from the political debate about energy.
"The political contradiction — red states building green infrastructure — is real, persistent, and almost entirely absent from the political debate about energy."
The AI Wildcard
There is a new variable in the energy equation that was not present three years ago and is not yet fully priced into the political debate: artificial intelligence.
AI training and inference is extraordinarily energy-intensive. A single large model training run consumes as much electricity as hundreds of thousands of homes use in a year. The deployment of AI at commercial scale — which is now underway — requires a physical infrastructure of data centers that must be powered continuously, at massive scale, with high reliability.
The hyperscalers — Microsoft, Google, Amazon, Meta — are in an arms race to build that infrastructure. They are signing power purchase agreements, investing in nuclear energy, and in some cases building dedicated generation capacity for their data centers. The electricity demand implications are large enough that grid operators are revising their long-term demand forecasts upward for the first time in decades.
This is an energy story disguised as a technology story. The country that can provide the most abundant, reliable, and affordable electricity will have a structural advantage in the AI economy — the same way the country with the most abundant, reliable, and affordable oil had a structural advantage in the 20th century industrial economy.
The United States currently has that advantage. Whether it maintains it depends on infrastructure decisions being made right now, most of which are being made by grid operators, utility regulators, and state permitting agencies rather than by Congress or the White House.
What the Political Debate Gets Wrong
The political argument about energy in America is conducted almost entirely in the vocabulary of the last transition rather than the current one. The debate is about coal versus solar, oil versus wind, fossil fuels versus renewables — a framing that made sense in 2010 and is increasingly inadequate to describe what's actually happening.
What's actually happening is more complicated and more interesting. Natural gas is playing a bridging role in the energy transition that neither side of the political debate wants to fully acknowledge. Nuclear energy — which produces zero carbon emissions and enormous amounts of reliable power — is being reconsidered seriously for the first time in forty years, driven not primarily by climate advocates but by technology companies that need reliable power for data centers. Offshore wind is struggling with supply chain and financing problems that have nothing to do with ideology. Onshore solar and battery storage are cheap enough that they're being built faster than the grid can absorb them in some regions.
The energy transition is happening. The political debate about whether it should happen is largely beside the point. What the political debate is actually determining — through permitting rules, grid interconnection policy, transmission investment, and tax incentives — is how fast it happens, who captures the economic value it creates, and whether the United States or someone else leads it.
Those are consequential questions. They deserve a more sophisticated debate than they're getting.
The Boom That's Coming
Economic expansions built on energy transitions share a common characteristic: they are not visible as expansions until they are already underway. The productivity gains that cheap oil enabled weren't obvious in 1950. The productivity gains that cheap electricity enabled for the information economy weren't obvious in 1985.
The productivity gains that cheap, abundant, clean energy will enable are not fully visible now. But the structure is becoming clear. Energy is an input cost for almost everything — manufacturing, transportation, agriculture, data processing, heating and cooling. When energy gets significantly cheaper and more reliable, the cost of everything that uses energy falls. When that happens at scale, across an entire economy, you get an expansion.
The question is not whether the energy transition will produce an economic expansion. The question is whether the United States will be the primary beneficiary of that expansion, or whether the capital, manufacturing, and innovation that drives it will migrate to places with better infrastructure, faster permitting, and more coherent industrial policy.
"The question is not whether the energy transition will produce an economic expansion. The question is whether the United States will be the primary beneficiary."
That question is being answered right now, in decisions that are receiving a fraction of the political attention they deserve.
Sources: BloombergNEF (clean energy investment data), Lawrence Berkeley National Laboratory (solar and storage deployment), IEA World Energy Outlook 2024, Energy Information Administration, Congressional Budget Office (IRA investment estimates), FERC grid interconnection queue data.
This is the second installment of "The Data," a series examining the numbers behind American political spending and governance. Next: "What America Would Look Like If We Actually Chose Innovation."