To the surprise of no one—except the Congressional Budget Office, apparently—the cost of President Biden’s climate tax credits has doubled its original projections. The tax credits, which apply to electric vehicle sales and clean energy projects, are now projected to cost $428 billion through fiscal year 2033. 

Goldman Sachs warned last year that the Inflation Reduction Act (IRA) would likely cost three times its original $270 billion price tag, or $1.2 trillion. Many of the IRA’s provisions are uncapped, which means that the costs can grow with little restriction as more businesses and citizens utilize them.

The CBO has revised its projections in part because of the Energy Department’s proposed rule that would bring the calculated fuel economy of EVs into closer alignment with their actual efficiency. In effect, however, the rule would force automakers to manufacture even more EVs in order to comply with tightening emissions standards. Because manufacturers would be pushing more EVs onto consumers, more consumers would be eligible for the $7,500 EV tax credit under the IRA, driving up its costs. 

Increased tax credits to solar, wind, and battery manufacturers account for $204 billion of the new projected cost. Increased uptake of EV credits and foregone gas tax revenues account for $224 billion of the projected cost—since all of the new EVs on the road won’t be filling up at the pump and contributing to highway repair and maintenance funds. Worse: the true cost of the IRA is likely to be larger than the new prediction of $428 billion, as “the CBO only includes half of the effects of any proposed rule.” The other half of the rule’s effects won’t be accounted for until it is finalized.

Supporters claimed that the IRA would reduce the federal deficit by $300 billion. The CBO projects the deficit will grow from $1.6 trillion in fiscal year 2024 to $2.6 trillion in fiscal year 2034.
What reduction in carbon emissions is being bought with $428 billion in higher-than-expected costs? Reuters reports:

“Without the IRA bill, the Congressional Budget Office, the non-partisan research arm of the legislative branch, estimated when that [sic] U.S. greenhouse gas emissions would decrease by 24% to 35% by 2030 compared to 2005 levels. The same estimates said the bill could reduce emissions by 32% to 40% by 2030 compared to 2005 levels.” 

Note the overlap in the best-case scenario without IRA incentives (35%) with the worst-case scenario with IRA incentives (32%). While these figures may be outdated now, it is illustrative of an important point: no one knows how much the IRA will cost in the end, and no one knows for sure how the IRA will actually influence the problem it purports to solve. 

Forecasting climate, emissions, and human behavior is a tricky business. It would be wise for regulators to practice humility with their assumptions about what the government can, and can’t do, to “solve” climate change. The IRA may prove a costly lesson.