Modeling and Implementation of the U.S. Hydrogen Production Tax Credit

Abstract

Low-carbon hydrogen (H2) could contribute to achieving long-term climate goals by supporting the decarbonization of several hard-to-abate industries. The U.S. Inflation Reduction Act includes a tiered hydrogen production tax credit (PTC) awarded for producing H2 below certain emissions thresholds. One pathway for producing PTC-eligible H2 is water electrolysis supplied with low-carbon electricity. But assessing the systems-level emissions associated with electrolytic H2 is challenging, not only because instantaneous power flows from a particular producer cannot be directly associated with a particular user, but also because of the risk that electrolyzers might divert clean electricity away from the grid. Following the passage of the IRA, there has been a vigorous debate focusing primarily on the time-matching requirements — that is, the period over which electricity use must match production from contracted generators — for grid-connected H2 production to receive the PTC.

Applying a macro-energy systems model to case studies of Texas and Florida, we show that divergent results in the literature, which presented a conundrum for regulators trying to pick between policy options, are explained by different interpretations of the proposed “additionality” requirement. Specifically, the emissions associated with H2 production under different “time-matching” requirements are conditional on how additionality is modeled. We further show that the interaction of these qualifying time-matching requirements with other energy system policies could reduce the merits of more stringent time-matching requirements. For instance, if a region has a relatively high renewable portfolio standards (RPSs) to enable grid decarbonization, we show that less stringent (and therefore less costly) time-matching requirements are sufficient to avoid any increases in system-level emissions.

Building on this analysis, we explore how uncertainty in inter-annual variable renewable energy (VRE) generation complicates the implementation of stringent PTC requirements. We confirm that a system design that accounts for inter-annual VRE uncertainty comes at a cost premium — a reality ignored by the existing literature. In addition, we show that inter-annual VRE uncertainty will necessitate the formation of markets for hourly electricity attribution certificates (EACs) to make up for inevitable shortfalls in supply of contracted VRE electricity supply under an hourly time-matching requirement.

We recommend that the Treasury adopt a phased and regionally differentiated approach to implementing the PTC — regions without RPS policies could transition to an hourly time-matching requirement in the mid-term (e.g., by 2030), whereas regions with sufficient RPS policies could continue with looser requirements. In addition to PTC implementation, these results are relevant to the broader field of Scope 2 emissions accounting for voluntary (e.g. corporate net-zero goals) and regulatory purposes. As more private enterprises, such as data centers owners, pursue voluntary measures to reduce their electricity-related emissions, our work provides a foundation for further research into clean energy procurement standards (voluntary or mandated) that support power sector decarbonization.