Trillion-Dollar Surprise in the Inflation Reduction Act

The national debt surpassed a mind-boggling $35 trillion last week — approximately $100,000 for every man, woman, and child living in the United States. This means a family of four’s share of public debt, ~$400,000, is likely more than they owe on the mortgage of their house. Runaway government spending is no surprise. What is surprising is that one bill passed a couple years ago may end up costing trillions, with a “t,” more than the public was told. That bill is the Inflation Reduction Act (IRA).

Many have rightly called the IRA ‘the New Green Deal Lite.’ The massive bill was primarily geared towards funding green energy projects. Some of that funding took the form of direct subsidies for conservation and energy efficiency projects. The main goal was to divert hundreds of billions of dollars into green energy projects. The uncapped lion’s share of the cost comes in the form of special tax credits through the internal revenue code.

Initially the Congressional Budget Office scored the IRA in September 2022 as reducing federal deficits by more than $200 billion over a decade. The Center for a Responsible Federal Budget summarized the CBO’s score for the IRA as costing ~$391 billion in energy subsidies and tax credits as well as ~$108 billion in healthcare subsidies. The CBO estimated the IRA would raise ~$738 billion in revenue over a decade by increasing a variety of taxes and repealing various tax exemptions — a classic “spend now, pay later” scheme.

Yet the Joint Committee on Taxation put out revised estimates less than a year later (June 2023) that the clean energy tax credits, rather than costing roughly $270 billion, would cost closer to $663 billion. That’s about a $400 billion dollar increase in expected costs. And that’s likely too low.

Recent clarifications and definitions of the tax code, particularly Section 45Y about Clean Electricity Production Credits and Section 48E about Clean Electricity Investment Credits, combined with provisions of the Inflation Reduction Act, have opened the possibility that the IRA may eventually add trillions of dollars to the national debt due to future use of these tax credits. One estimate puts the total cost of these green energy credits closer to $3 trillion over their entire lifetime- which might be much longer than people realize. 

The first and most important problem is the open-ended timeline of the IRA. While 2032 was mentioned in the bill as a possible termination date, the bill also specified that the US hitting 25 percent or less of its 2022 emissions level was another possible endpoint. Here’s the catch, though. It is whichever date happens later!

Given historical trends, emissions won’t hit 25 percent of 2022 levels by 2032. The implausibility of reducing emissions to that level means we have no idea when these renewable energy tax credits will expire. There is a good chance they won’t hit that level before 2040, or even 2050. So the program could very well run more than twice as long as people thought. And every year, as governments heavily subsidize “green” energy, more and more tax credits will be claimed.

A second problem is all the distortions created by unequal treatment of renewable energy sources and fossil fuel sources. The new IRS rules define “zero-emissions” as no emissions created in the process of generating energy. Emissions from manufacturing and installing renewable energy assets don’t count. Nor do emissions created when servicing them. Nor do emissions created by other electricity generation to keep wind turbines turning when there is no wind or to provide supplemental energy to stabilize the intermittency of wind and solar energy.

When it comes to “net zero” policies, some emissions are more equal than others. Apparently it doesn’t matter that fossil fuels are used to min the minerals and raw materials used to create wind turbines, solar panels, and batteries. And the fuel burned by ships and trucks in the production process are also apparently benign.

But fossil fuels are treated differently. Fossil fuel energy assessment requires “life-cycle” analysis to calculate their emissions. That means in addition to the emissions created when generating electricity, all the emissions that didn’t count for wind and solar do count for fossil fuels: emissions generated in extracting, refining, and transporting, etc.

This heavy subsidizing of renewable energy installation has led to another problem in many electricity markets: near zero marginal cost electricity provision during the day. While the fixed costs of wind and solar are largely underwritten with subsidies and tax credits (and mandates when those incentives are insufficient), thermal fossil fuel energy generation has to compete with basically zero marginal price of solar and renewable energy throughout the day.

Another contributor to the cost and inefficiency of the IRA renewable energy credits is the “80-20” investment rule. If renewable energy facilities are upgraded or expanded to the extent that the new investment represents 80 percent or more of the current market value of a project, that project can apply for significant energy credits, even if it applied in the past. This rule will encourage aggressive depreciation of assets and premature, costly additions or upgrades.

Of course, the emissions created when installing new blades or solar panels, or the emissions created when transporting and storing discarded blades and solar panels, are “more equal” than other emissions. Even if one doesn’t think net-zero policy is the modern equivalent of tilting at windmills, the uneven playing field being created here should concern everyone who cares about justice and prosperity.

All this adds up to an open tab that renewable energy companies can use over and over again and that taxpayers will pick up. Unless Congress acts to close these loopholes, a $35 trillion debt will look quaint when the US surpasses $45 trillion or even $50 trillion in the 2030s. If this happens, the Inflation Reduction Act of 2022 will be a surprisingly large contributor.