Twenty years ago, Breakthrough helped launch the first effort to accelerate climate mitigation through large-scale federal investment in low carbon technology. At the time, the idea that public investment, not carbon regulations or pricing, would be the long pole in the decarbonization tent was controversial, indeed, "contrarian." Today it has become conventional wisdom among Democrats, progressives, and most of the center left, culminating in the passage of the Inflation Reduction Act.

But there was always a misunderstanding among many, both proponents and critics of this approach, about what "make clean energy cheap," a term that we coined, actually entailed. For a lot of proponents, it simply meant subsidizing clean energy, mostly wind and solar, basically forever. Critics saw it the same way and rightly pointed out that maintaining subsidies as clean tech began to scale would become incredibly expensive, inefficient, and—particularly with variable renewables—would distort electricity markets in ways that would likely raise overall costs even with substantial subsidies.

But the point of “making clean energy cheap” was not to do so via perpetual subsidy but to invest in innovation and early stage commercialization of clean technologies to “cut the Gordian Knot” that created the zero-sum tradeoff between the cost of energy and carbon emissions, as we wrote in one early policy review. So long as clean energy was substantially more expensive than fossil fuels, efforts to mitigate climate change would impose substantial economic costs on consumers and businesses that were, in the best case, challenging to sustain politically and, in the worst case, damaging economically.

Making clean energy cheap in real, unsubsidized terms remains the right approach today and, in our view, the only plausible way to assure either US energy security or significant global decarbonization in the long term. With the Trump Administration now in power and Republican majorities in Congress vowing to repeal the IRA, we published an op-ed in the Wall Street Journal last month arguing that the incoming Trump Administration should reform but not repeal the Inflation Reduction Act. We argued that with US electricity demand set to increase substantially alongside major expansion of US natural gas export capacity, making clean energy cheap will likely be necessary to assure that all energy in the United States remains cheap.

In the intensely polarized environment that has overtaken US climate and energy politics, this seemingly obvious point has become anathema to many partisans on both sides of the issue. So it is worth reviewing why a sustained commitment to US federal energy innovation, tied to real progress on the cost and productivity of clean energy technology rather than arbitrary and implausible emissions reduction targets, should remain a priority for both Democrats and Republicans in Washington.

Increasing the Cost of Fossil Fuels and the Iron Law of Climate Politics

Efforts to cut emissions by making fossil fuels more expensive have reliably foundered virtually everywhere whenever those policies result in, or simply happen to correspond with, a significant increase in energy prices. Well over a decade ago, Roger Pielke Jr. dubbed this phenomenon as the Iron Law of Climate Policy. In recent years, climate advocates have variously attempted to bargain with Pielke’s Iron Law by producing spectacular estimates of the social and economic cost of future warming, proposing to recycle proceeds from carbon taxes back to consumers, or hiding the “green premium” (the additional cost of clean energy) in subsidies that purport to offer energy consumers something for nothing. But thus far, the Iron Law remains undefeated.

The reelection of Donald Trump and the unpopularity of the IRA are but the latest examples of the Iron Law. It was unfortunate for Biden and IRA supporters that the passage of the IRA happened to coincide with post-COVID inflation and a big runup in energy and commodity prices. But the IRA also contributed to that runup, adding a half trillion dollars in public spending to an already overheating economy and offering huge public incentives for utilities and developers to deploy intermittent renewables even when penetrations of those sources were already high and hence the electricity they produced had increasingly low value to the grid.

The cost of the IRA to the public purse was also high. Initial estimates of the cost of the IRA by the Congressional Budget Office projected that the entire cost of the legislation for clean energy over the first ten years of its implementation would be around $390 billion. More recent estimates project the total cost of these programs to run closer to a trillion dollars, with the cost of wind and solar subsidies alone substantially exceeding the cost of the original estimates not only for the clean energy subsidies but for the entire cost of the package, inclusive of non-climate related spending.

The reason that the cost of the IRA is so high is because it proposes to subsidize mature, widely commercialized technologies. Solar now accounts for roughly 5% of US electricity generation and almost 25% in California. Wind accounts for over 10% of US generation and almost 30% in Texas. EVs will likely account for around 10% of US vehicle sales this year. With federal tax credits covering a third (or more) of the cost of wind and solar installations and $7500 for every EV that automakers sell, the aggregate costs of this approach to bringing down the unit costs of clean energy gets really expensive really fast. Renewables subsidies before the pandemic totaled less than $10 billion in federal expenditures annually. Current estimates suggest they will total three to six times as much on an annualized basis over the next decade.

Despite this significant escalation in the projected cost of the IRA, US emissions appear likely to significantly undershoot the original emissions reduction estimates that proponents of the IRA argued the package would achieve. After passage, Princeton’s Net Zero project estimated US emissions would fall 50% below 2005 levels by 2035. Last year, they softened this projection to 40% below 2005 levels. A 2024 report from Princeton, Rhodium, Energy Innovation LLC, and MIT found that deployment trends in low-carbon electricity were insufficient to achieve the goal of 40% reduction in emissions by 2030 established by the IRA. And as we found last year, taking these revisions into account, IRA investments would likely sustain the existing pace of US emissions reductions, not significantly accelerate them.

These downward projections come even as the estimated cost of the IRA programs has roughly doubled. So exactly as many skeptics suggested back in the mid and late 2000s, the attempt to use public clean energy subsidies as a reverse carbon tax is already proving to be an inefficient and expensive way to cut emissions.

Why Are Subsidies Needed if Wind and Solar Are Already So Cheap?

The idea that we need to subsidize wind and solar energy to make them cheap will strike many as odd, given that by many estimations, they are already incredibly cheap. Over the last decade, a wide range of prominent voices have insisted that wind and solar energy were the cheapest sources of new electricity generation in many places. As David Roberts wrote at Grist ten years ago, “a range of renewables are now competitive with fossil fuel power, without subsidies, and are only going to strengthen their position in coming years.” Chris Nelder shared this optimism as early as 2012, writing “unsubsidized solar generation could be cheaper than conventional grid power within a decade.” In 2013, Noah Smith predicted 2020 would be the year “when solar will actually start being cost-competitive with fossil fuels, without subsidies.” Influential industry analysts, such as the investment firm Lazard, also regularly declared solar and wind cost-competitive with or without subsidies.

What explains the shift from these celebrations a decade ago to the passage of the IRA in 2022?

The official answer to this question is that reducing emissions is a worthwhile public pursuit and the policies initially designed as temporary innovation subsidies were appropriately transformed into quasi-permanent emissions reduction subsidies. The less official answer, though, is that renewable energy advocates realized that absent very cheap storage and very substantial expansion of transmission capacities, continuing deployment of wind and solar wouldn’t be economically viable without direct subsidies in perpetuity.

Bret Christophers got a lot of attention for explicating the need for sustained subsidies to assure continued wind and solar deployment in his book The Price Is Wrong, published last year. But many energy analysts have been aware of the problem for over a decade. As declining unit costs of wind turbines and solar panels did indeed spur ballooning solar and wind penetration on electric grids, the value provided by that electricity declined. One of us (Alex) coauthored one of the early articles on the “value deflation” problem with Jesse Jenkins in 2015 writing that since geographically proximate wind turbines and solar panels tend to all generate electricity at the same time (when the wind is blowing and when the sun is shining), “the marginal value of variable renewable energy to the grid declines as the penetration rises.”

As Jenkins would go on to tell Brad Plumer in 2016, renewables’ grid parity with fossil fuels was “just the beginning of the race…solar cost reductions then have to continue to outpace their declining marginal value.” And as Zeke Hausfather wrote in a Breakthrough Institute report on solar value deflation in California in 2022, “some degree of future subsidies — either in the form of current tax incentives or future nominally technologically-neutral mechanisms like carbon pricing or a clean energy standard — will likely be needed to sustain cost-effective deployment of the high levels of solar…”

Up through the end of the first Trump administration, the prospects were thin for permanently extending policies which advocates had long insisted were temporary. Then came covid. Policymakers’ immediate response to the pandemic was rapid and enormous federal relief spending, and climate advocates leapt at the opportunity. Congress’s multi-year pandemic spending spree culminated with the Inflation Reduction Act in the summer of 2022, which not only extended but expanded clean energy subsidies far beyond what advocates could have dreamt of just a few years earlier.

And that’s how energy innovation subsidies that everyone agreed would eventually become unnecessary became emissions reduction subsidies scheduled to expire only when economy-wide US emissions are 75% below 2022 levels, which translates to almost 80% below 2005 levels, or 40 percentage points lower than the IRA’s 2030 emissions goal.

How to Right Size the IRA

Our advocacy for sunsetting clean energy subsidies is nothing new. From early in the last decade, most notably in a 2012 report titled Beyond Boom and Bust, coauthored with Jenkins, we advocated for doing so. Most recently, we did so in the early days of the Biden administration, in a report titled Saying the Quiet Part Out Loud, coauthored with Hausfather. As we noted in the Wall Street Journal, most IRA spending is slated to go to wind, solar, and electric vehicles:

Somewhere between half and three-quarters of projected Inflation Reduction Act spending over the next decade will be for wind, solar and electric-vehicle subsidies. These are all mature, cost-competitive technologies that don’t need further subsidization. Cutting their subsidies could amount to somewhere between $300 billion and $650 billion in savings.

Doing so need not bring about the death knell for any of these technologies. The solution to the value deflation problem is not to continue to subsidize deployment of variable renewable energy onto grids that don’t need it. It is to invest in and deploy transmission and energy storage technologies to increase the value of intermittent energy by making it available where and when there is actual demand for it. Similarly, the solution to range anxiety among mainstream automobile consumers is to invest in charging infrastructure and better batteries, not to continue to prop up the sector by handing out $7500 to predominantly high-income households to purchase or lease EVs.

There are substantial funds for batteries, charging, and transmission in the IRA and IIJA that we would strongly argue that the Trump administration and Congress sustain, much of it in the IIJA which, in contrast to the IRA, was passed with substantial bipartisan support.

Consistent with our long-standing advocacy, wind and solar tax credits will be folded into a technologically neutral clean energy tax credit in 2026 that applies to nuclear, carbon capture, and geothermal energy as well. This shift was long overdue and provides critical support for those technologies, as well as next generation renewable technology such as perovskite solar panels. The sunset of these credits is presently tied to emissions targets. By shifting phaseout criteria to trigger at a given threshold of grid penetration, Congress could sustain support for early stage clean energy generation technologies while eliminating the lion’s share of the projected cost of the IRA over the next decade.

Sunsetting federal support at 5% or 10% of total annual generation would allow for continued federal support for technologies as firms are attempting to commercialize them without committing the federal government to costly subsidies far into the future for mature technologies. Congress could even do so regionally, and not nationally, sunsetting support for qualifying technologies when they exceeded a set percentage of annual generation by NERC region. As recently as 2021, Senators Crapo and Whitehouse proposed this very model of subsidy reform, expanding the PTC and ITC to cover nuclear and geothermal in a way that “phases out credits as technologies mature, which provides an on-ramp for the most innovative technologies to get to market and then compete on their own.”

For EVs, the IRA and IIJA already provide substantial support for charging infrastructure and battery manufacturing. These latter investments are separate from consumer subsidies, have helped the US inshore manufacturing for battery technologies, and notably, have broad application beyond EVs. Current EV subsidies may help stimulate demand for domestically-sourced critical minerals. But Congress ought to promote battery metals security specifically rather than incidentally through the downstream EV credit. Phasing down generous consumer subsidies while sustaining support for minerals, battery manufacturing, and charging will nudge automakers to repurpose new assembly lines toward hybrid and other products that are better suited to the US market, to target export markets where EVs may be better suited presently, or to delay some EV production for a period of time while battery technologies improve to the point that they can meet the demands of US auto consumers.

The further requirement for this strategy to succeed in supporting US energy abundance is permitting, transmission, and interconnection reform. As noted above, wind and solar are not economically viable without subsidies not because solar panels and wind turbines aren’t cheap but because it has become increasingly difficult to site them in many places and to get the power they generate to the places where it is needed when it is needed.

Democratic leaders in the last Congress ultimately abandoned the opportunity to pass a legislative package inclusive of permitting and transmission reform, at the behest of an unholy alliance of environmental groups adamantly opposed to any meaningful reform of the National Environmental Policy Act and investor-owned utilities wary of competition that grid interconnection would bring.

That decision was myopic before the election and unconscionable after it. In the absence of forever subsidies for wind and solar, permitting reform and interstate transmission are essential if those industries are going to thrive. No one appears to have done any modeling directly comparing the emissions benefits of solar and wind with subsidies and without permitting reform and transmission versus with permitting reform and interstate transmission and without subsidies. But given that Jenkins’ modeling at Princeton’s Net Zero project, conducted shortly after passage of the IRA, found that 80% of the estimated total emissions reductions through 2035 vanished without major expansion of transmission, we suspect that transmission, interconnection, and cheap battery storage are much more critical to the future of wind and solar, and the emissions benefits that additional wind and solar deployment might bring, than are the IRA tax credits.

Today, a deal for permitting and transmission reform is likely the one post-IRA policy arena where Democrats still have some leverage, in that any significant reform of NEPA will require Democratic votes to overcome a filibuster in the Senate. This, together with an IRA reform package that cuts wind, solar, and EV subsidies but preserves incentives for early-stage innovation hold the possibility of not only saving key elements of the IRA but establishing a durable bipartisan consensus in Congress on energy innovation and regulatory reform.

Resisting the Siren Song of Resistance

Among architects and partisans of Biden era climate policy, there is an understandable temptation to simply resist changes to the IRA. A less ambitious climate agenda could slow emissions progress. Trump’s popularity and power may already be peaking. The preponderance of IRA investments in red states and districts will incentivize Republicans to protect them. Taken together, state-based action in blue states and resistance in key red states where IRA investments are already at work might hold IRA opponents at bay until the political winds shift again, when a Democratic Congress in 2026 and President in 2028 might ride into town to save the day.

But less than a month into the second Trump presidency, it is worth considering that other possibilities loom large. Trump’s declaration of a national energy emergency and sweeping executive orders suggest that betting on a bumbling second Trump administration unable to get out of its own way, manage its political coalition, or effectively press the case for its agenda is, at best, a risky proposition. Trump and his allies are much clearer this time around about what they want to do and how they intend to do it than they were in January of 2017.

Republicans in Congress will be under immense pressure to move quickly to eviscerate the IRA in its entirety. Moreover the most salient IRA investments for red districts and states—and for key Republican constituencies—are not wind, solar, or EV tax credits. The choice for climate and clean energy advocates is unlikely to be between preservation of the IRA in full versus partial repeal but rather between partial repeal and full repeal. That will require that Democrats and pragmatic climate advocates make hard choices, recognize that Trump and Congressional Republicans are going to take a pound of flesh out of the IRA, and think strategically about what it should be.

The same will be true around key regulatory reform issues. Lasting reforms will require statutory changes that can’t be effectively accomplished through budget reconciliation. Democrats in Congress have leverage - less leverage than they did last year but likely enough to demand reforms that are technologically neutral and that bring reasonable changes to transmission and interconnection rules along for the ride.

Bipartisanship can, of course, be a siren’s song of its own. But it is worth considering that insofar as it can be achieved, the advantages of durable policies achieved through normal legislative deliberation versus more sweeping policies achieved via reconciliation and executive action are potentially significant. Even without repeal, the IRA and various executive and regulatory actions undertaken by the Biden administration were on course to fall well short of the administration’s targets, even before the coming growth of AI driven load demand was accounted for. Given all of the technological, macroeconomic, and political uncertainties involved over not just the next four years but the next forty, sacrificing speed, scale, and ambition for energy policies that are built to last and have significant buy-in from both parties has always been the smarter and more likely path to clean energy abundance.

That is all the more the case with the Democratic coalition now in tatters. Perhaps a Democratic restoration over the next two or four years is in the cards. But it could just as easily go the other way, with Republicans holding on to one or both houses of Congress in 2026 and JD Vance succeeding Trump in 2028. Absent a Democratic trifecta in 2028, Biden’s climate and green industrial policies will be impossible to reconstruct should Trump and Republicans succeed in dismantling them in the current Congress. Even with a trifecta, it is highly uncertain whether Democrats would once again choose to spend political capital on a policy issue that few voters prioritize and that has demonstrably backfired on the party politically.

Ultimately, we don’t know if a return to the sustained bipartisan energy innovation consensus of the last decade or the construction of a new, abundance-focused consensus around regulatory reform for the next decade is possible. But Democrats and clean energy advocates would be well served to try.