The US Department of the Treasury has proposed long-awaited guidance on the implementation of the 45V tax credit for clean hydrogen production, with implications for the US hydrogen industry, emissions, and clean electricity.
In a thoughtful holiday present to energy policy analysts everywhere, the US Department of the Treasury released its proposed guidance on the 45V tax credit for clean hydrogen production on the Friday before Christmas. Earlier leaks to the media that this guidance would accord with the so-called “three pillars” approach mostly were accurate. Instead of rehashing prior analysis, I’d like to focus on the novel aspects of the guidance. I will give a brief overview of what’s in the proposed guidance with respect to the treatment of electricity and discuss the reasons why I think the Treasury Department should allow state-level policies to play a larger role in compliance with the guidance. In addition, I will highlight an anti-abuse provision in the guidance that may help reduce the subsidization of an extremely counterproductive method of producing hydrogen.
I have discussed the 45V tax credit in prior blog posts, so I won’t go too deep into the details here. The most important aspect of the tax credit is that the value of the credit is based on the life-cycle greenhouse gas emissions of hydrogen production and not just on the emissions that a hydrogen facility produces on-site. This link to life-cycle emissions means that the value of the tax credit depends on the emissions from the production of electricity that facilities use in the production of hydrogen, along with any upstream leakage of natural gas if a facility uses natural gas to produce hydrogen.
How Electricity Can Qualify as Clean under 45V
The largest controversy surrounding the 45V tax credit always has been the treatment of life-cycle emissions from electricity production. In particular, debates have centered on how hydrogen producers can use energy attribute certificates (EACs) to claim to be using clean electricity, rather than electricity from the grid which includes generation from fossil fuels. EACs, sold by clean electricity producers and purchased by hydrogen producers (and others), represent the “attributes” of clean power. Importantly, these credits do not represent the actual consumption of clean power.
As hydrogen producers ramp up electricity consumption, some analysts have voiced the concern that EACs are just pieces of paper that have no impact on reducing emissions or incentivizing clean energy generation. This concern has led to proposals that, for EACs to comply with 45V, credits must fulfill three requirements, which many proposals call the “three pillars.” In this discussion, I will use the Treasury terminology of incrementality, deliverability, and hourly matching.
Incrementality. EACs must come from clean generation that goes online less than 36 months before the hydrogen facility in question begins producing hydrogen. Note that the Treasury Department usefully has chosen the term “incrementality” over “additionality,” a term that alternative proposals for 45V have favored, but which already has a different meaning in other contexts.
Deliverability. EACs must be associated with clean electricity generation that comes from the same region as the hydrogen production facility that buys those EACs. The regions in the proposed guidance are derived from the regions that are defined in the National Transmission Needs Study, which the US Department of Energy (DOE) published in October 2023.
Hourly matching. EACs must be associated with clean electricity that is generated in the same hour in which a facility consumes that electricity to produce hydrogen.
The motivation for these requirements is the hope that, when hydrogen producers purchase EACs that follow these three requirements, emissions will be lower than in a scenario in which the EACs are unrestricted. I have written previously about the extent to which this expectation for lower emissions is true. In Treasury’s guidance, the implementation of hourly matching is delayed until 2028 to allow for the development of tracking systems for these EACs.
Questions from the Treasury Department
As part of this proposal, the Treasury Department is taking public comments on a few different methods that hydrogen producers also can use to demonstrate their fulfillment of the incrementality requirement. The first method is to allow EACs to fulfill the requirement if the electricity comes from generators that otherwise would have retired; for example, nuclear power plants that are at risk of retiring. However, demonstrating that a given plant would have retired in the absence of demand from a hydrogen production facility likely would be challenging, and Treasury requests comments on this question.
Treasury also requests comments on how to use modeling to demonstrate fulfillment of the incrementality requirement. Modeling could show that an EAC corresponds to clean electricity that otherwise would not have been generated (called “curtailment”). Models also could show that state-level policies would not allow new emitting electricity generation to be built, which would reduce the emissions that are due to an increase in electricity demand from hydrogen production facilities. RFF Fellow Kevin Rennert and I have proposed that the Treasury Department incorporate the impact of state-level policies, such as emissions caps, into the 45V guidance.
However, as I discuss below, all three requirements work together in a cohesive manner to reduce emissions (as compared to a policy with unrestricted EACs). Consequently, I do not believe the Treasury Department is right to focus on incrementality in this request for comment. Further, I believe it would make sense for Treasury to allow state-level policies to obviate the need for all three requirements.
Finally, the Treasury Department is taking comments on a simplified approach to the incrementality requirement, which would allow 5 percent of existing clean generation to produce qualifying EACs without any further burden of proof. This approach is intended as a proxy for curtailment and the avoided retirement of electricity generation facilities, precluding the need for complicated calculations and other considerations. This approach is intriguing because it avoids complicated modeling and the specification of alternative scenarios, but the resulting impact on emissions or costs of this approach is hard to predict.
White Paper from the US Department of Energy
The Treasury Department bases this guidance on a white paper published by DOE that appears to lay out the reasoning behind the distinctions among the different requirements. When one thinks about emissions that are due to an increase in electricity demand (such as new demand from hydrogen producers), one must consider two effects: First, what, if any, new generation is built to serve the new demand? Second, how will the electric grid operate in response to that new demand? Kevin Rennert and I call these two effects the “capacity effect” and the “dispatch effect,” respectively, in a prior paper.
The capacity effect is by far the most important determinant of emissions. If new emitting electricity generation is built to meet new demand, then the resulting emissions will be significantly greater than the threshold that hydrogen producers need to stay below to receive the 45V tax credit. Thus, a policy that incentivizes the construction of new clean generation is paramount if emissions are the only consideration.
Nonetheless, the dispatch effect is still important and can lead to emissions on the same order as the emissions thresholds in the 45V tax credit. DOE similarly discusses “structural” and “operational” impacts in the white paper, which are akin to the capacity and dispatch effects, respectively.
What Do the Three Requirements Do?
To me, the analysis in the white paper seems to say that the incrementality requirement is about addressing the capacity effect, and the deliverability and hourly-matching requirements are about addressing the dispatch effect. I will start with the latter point.
In the white paper, DOE says, “This demonstrates that the absence of either geographic or temporal matching between load and generation would not reflect important ways in which new loads can impact [greenhouse gas] emissions.” But while the deliverability and hourly-matching requirements can address the dispatch effect, these requirements are not the only way to address the dispatch effect. For example, the Treasury Department could track local changes in emissions that are due to increases in electricity generation or consumption (called local marginal emissions rates). Furthermore, the deliverability regions that Treasury uses in the proposed guidance are much too large to ensure that no transmission constraints exist between the clean energy generation and the hydrogen production, which Treasury acknowledges. In our work, Kevin Rennert and I have shown substantial residual emissions due to the dispatch effect, even with the requirements of temporal matching and deliverability at the scale of the proposed regions.
Paying for New Clean Electricity Generation
DOE correctly states that purchasing renewable energy certificates (which are similar to EACs) from existing electricity generators would not address the capacity effect, because other generation would be built to meet the additional demand for electricity. However, incrementality alone is not sufficient to force all new electricity generation to be clean. If one wants to build additional clean generation, someone has to pay for it. When EACs are traded, this money is expected to come from the revenue that a clean electricity generator earns by selling EACs to hydrogen producers, which means that the price of the EAC must be greater than zero.
Incrementality, by itself, is not sufficient to ensure a positive price for EACs, given that substantial amounts of renewable energy likely will be built regardless of any 45V policy. In fact, my reading of the literature indicates that all three requirements work together to address the capacity effect by reducing the supply of EACs that qualify for the 45V tax credit: EACs must be issued in a certain region (i.e., fulfill the deliverability requirement), come from incremental generation (i.e., the incrementality requirement), and correspond to electricity generated in a certain hour (i.e., the hourly-matching requirement). This combination will sometimes—but not always!—reduce the supply of EACs that is available to hydrogen producers, cause a positive price for EACs, and motivate the construction of more clean electricity than otherwise would be generated without the proposed 45V requirements.
More clean energy means lower life-cycle emissions for hydrogen production. But this clean electricity generation would not guarantee that the life-cycle emissions would be lower than the threshold necessary to receive the 45V tax credit; the additional generation just increases the odds that emissions would be lower than without the requirements. And increased emissions still may arise from the dispatch effect.
A Role for State Policies
I expect that the three requirements in the proposed guidance for 45V likely would lead to a scenario with lower emissions than, say, a scenario in which Treasury requires only that an EAC correspond to electricity generated in a certain year rather than a certain hour. But the analysis that DOE conducts in the white paper seems to go too far. All three requirements in the proposed guidance work together to reduce emissions, but these requirements mainly address the capacity effect (which DOE calls “structural impacts”) and address to just a limited degree the dispatch effect (which DOE calls “operational impacts”).
Thus, state-level policies that restrict the construction of new emitting generation could substitute for all three requirements—not just the incrementality requirement. For example, if a state has a cap on carbon emissions—assuming that emissions reduced by the cap aren’t offset by increased emissions in other less regulated states (i.e., no leakage), and no price cap on emissions permits—then an increase in hydrogen production would not lead to an increase in emissions, because the emissions are capped. As mentioned above, the Treasury Department should allow state-level policies to demonstrate that hydrogen producers are in compliance with low or zero life-cycle emissions overall—and not just that producers satisfy the incrementality requirement.
Abuse of the Inflation Reduction Act: The Never-Ending Tax Credit
The possibility of “stacking” tax credits in the Inflation Reduction Act could lead to an undesirable outcome, given the potential combination of the 45V tax credit with other credits in the law. One can imagine that a firm could earn a tax credit by generating clean electricity from solar or wind energy, earn another tax credit by using that electricity to generate clean hydrogen, then earn yet another tax credit by using that hydrogen to generate electricity—and so on to infinity, with the firm earning tax credits in each step of the process.
This loophole, needless to say, is neither the intended goal of the Inflation Reduction Act nor a good idea for the federal budget. Even without the possibility of stacking tax credits, if the value of the 45V tax credit exceeds the cost of hydrogen production, a firm could make money by producing clean hydrogen and simply releasing it into the atmosphere. The new 45V guidance seeks to avoid these unintended outcomes by including an anti-abuse rule that says potential recipients of the 45V credit cannot use the credit “in a manner that is wasteful, such as the production of qualified clean hydrogen that the taxpayer knows or has reason to know will be vented, flared, or used to produce hydrogen.”
I am not a tax lawyer, so I can’t speak to the implementation or enforceability of this anti-abuse provision, but any rule that works to avoid wasteful outcomes is a good idea.
Informing the Final Guidance for 45V
The Treasury Department has spent a lot of time on the proposed guidance for the 45V credit and is trying hard to reduce any increases in emissions from hydrogen production. The requirements in this guidance likely will further that goal—but at the cost of an increase in the price of clean hydrogen, alongside lower levels of clean hydrogen production than otherwise could have been achieved through guidance with less restrictive requirements. Finding the right balance for this inherent trade-off between stimulating hydrogen production and limiting life-cycle emissions ultimately will be a judgment call.
My hope is that the policymaking process will continue to be informed with further clarity on the underlying forces through which EACs impact emissions. I appreciate that Treasury has provided the opportunity to comment on the proposed guidance, and, most importantly, I hope that the agency will release the final guidance far from any holiday weekend.