The Inflation Reduction Act's clean electricity tax credits were always going to be expensive. The question was whether they were expensive in a useful way — buying down the cost of carbon at a rate that made fiscal sense. That question just got harder to answer, because Congress has now added a different variable: a hard stop.
The reconciliation bill working through Congress — the "One Big Beautiful Bill Act" — imposes a July 5, 2026, construction start deadline on the Clean Electricity Production and Investment Tax Credits, the IRA's core incentives for wind and solar. Projects that haven't broken ground by then lose access to the credits entirely. That's not a phase-down. That's a cliff.
Before we get to what that means for deployment, it's worth revisiting where the IRA's cost-per-ton math actually stood — because I've covered this twice already and the picture keeps shifting.
The $83/Ton Number Deserves Scrutiny
The research memo for this issue cites an average carbon abatement cost of $83 per metric ton for the IRA's clean energy incentives. That figure is in the range of estimates that have circulated from independent analysts, and it's worth taking seriously — but also worth unpacking.
Eighty-three dollars per ton is not a single program's cost. It's an average across a portfolio of incentives with wildly different efficiency profiles. The production tax credit for utility-scale wind and solar in competitive wholesale markets probably clears that bar reasonably well — those technologies are now cost-competitive with gas in many regions, so the subsidy is buying acceleration, not existence. The same cannot be said for every credit in the IRA stack.
The Energy Communities bonus credit — which directed additional incentives toward projects sited in areas with historical fossil fuel employment — reportedly drove a 33% increase in renewable investment in qualifying areas. That's a real outcome. Whether it's a cost-effective one depends entirely on what those projects would have built anyway and at what timeline. A 33% investment increase in a region that was going to get wind farms in five years anyway is a different fiscal story than one that unlocked genuinely stranded capacity.
This is the IRA's persistent measurement problem: the credits are structured to incentivize deployment, not to optimize abatement cost per dollar. Those are related goals, but they're not the same goal.
High Rates Did What High Rates Do
The other variable that's been quietly undermining the IRA's deployment thesis is the interest rate environment. Clean energy projects are capital-intensive and long-lived — their economics are acutely sensitive to the cost of financing. When the IRA passed in 2022, the federal funds rate was near zero. It isn't anymore.
Higher financing costs compress project returns, which means the tax credit has to do more work to make projects viable than the original cost estimates assumed. A credit that penciled out at $70/ton in a low-rate environment might be running $100/ton or more in the current one — not because the technology got more expensive, but because the capital stack got heavier. The IRA's architects didn't model this scenario, or at least didn't price it in publicly.
This matters for the fiscal cost estimate. If projects require larger effective subsidies to clear their hurdle rates, the total cost of the program per ton of abatement rises even if deployment numbers look healthy on paper.
A Deadline Doesn't Fix the Math — It Just Ends the Experiment
The July 5 construction deadline is being framed in some quarters as fiscal discipline. That framing is mostly wrong. A deadline doesn't improve the cost-per-ton efficiency of the credits that were already issued or that get issued before the cutoff. It just truncates the program before the learning curve — the period when deployment volumes typically drive down both technology costs and administrative friction — has fully run.
What the deadline does do is create a rush. Developers who were planning 2027 or 2028 starts are now scrambling to establish "begin construction" status before the cutoff. That means front-loaded spending on site control, permitting deposits, and equipment procurement — activity that looks like deployment but may not translate to completed projects for years. The legal definition of "begin construction" is already subject to litigation, which adds another layer of uncertainty to any near-term deployment count.
The fiscal irony is real: a program that was criticized for being too expensive is being curtailed in a way that may produce a burst of low-quality spending right before the door closes, without the long-run deployment volume that would have improved the cost-per-ton average over time.
Watch for Treasury guidance on what qualifies as construction commencement under the new deadline — that's the document that will determine how much of the pipeline actually survives. If Treasury draws the line narrowly, the effective cutoff may arrive well before July 5 in practice.