Letter to the editor

Response to ‘Does the purchase of voluntary renewable energy certificates lead to emission reductions? A review of studies quantifying the impact’

Hydrogen produced via electrolysis is expected to play a key role in decarbonising hard-to-electrify sectors, providing feedstocks and reducing agents for industrial processes, as well as fuel for aviation and shipping (Zeyen et al., 2023, IRENA, 2020, Anon, 2019, Staffell et al., 2019). Given its anticipated role in the energy system transformation, ensuring that green hydrogen produced from electrolysis with renewable electricity does not increase overall emissions has become a critical focus of research and policy. Recent studies (Brauer et al., 2022, Ruhnau and Schiele, 2022, Zeyen et al., 2024b, Ricks et al., 2023, Giovanniello et al., 2024) have extensively discussed certification standards for green hydrogen production, focusing on the varying degrees of spatial and temporal alignment required between renewable energy generation and electrolysis demand. Notably, these regulations also have important implications on domestic electricity prices, decarbonisation and hydrogen production costs in potential exporting regions (Schumm et al., 2024).

The European Union (EU) adopted a Delegated Act (DA) in 2023 that imposes, from 2030 onward, hourly matching between renewable generation and electrolysis demand from additional renewable capacity installed within the same bidding zone (EU Commission, 2022). These requirements aim to ensure that green hydrogen production contributes to emission reductions and drives investments into new renewable generation capacities. However, there is an ongoing debate about how different modelling approaches affect the evaluation of these criteria and their impact on emissions, hydrogen production costs and energy infrastructure.

In a recent publication ‘Does the purchase of voluntary renewable energy certificates lead to emission reductions? A review of studies quantifying the impact’ (Langer et al., 2024) the authors of the article (henceforth ‘the authors’) analysed different studies focusing on renewable energy certificates (REC) for corporate electricity procurement (Xu et al., 2022, Xu et al., 2024, Riepin and Brown, 2024) and renewable hydrogen regulation (Zeyen et al., 2024b, Ricks et al., 2023). Their work discusses how different modelling choices, such as how additionality requirements or national renewable generation targets are modelled, impact the emissions and evaluation of different policy regulations. In comparing emission impacts, the authors adopt an approach that contrasts a regulated case (e.g. additional renewable generation with annual or hourly matching) with an unregulated scenario (e.g., electrolysis connected directly to the grid). This differs from previous studies such as Ricks et al. (2023) and Zeyen et al. (2024b), which compare the emission impact based on a comparison between regulated cases with scenarios without hydrogen production.

The authors caution against drawing policy recommendations from Zeyen et al. (2024b), claiming that modelling assumptions regarding enforced additionality and national renewable generation targets produce results that favour annual matching. In this response article, we argue that the authors’ conclusions on the impact of additionality and renewable targets are misguided. These arguments were already addressed and refuted in the original publication by Zeyen et al. (2024b), where it was shown that the emission impact of annual matching is driven primarily by operational flexibility and the background grid mix rather than the assumptions about additionality or national renewable targets. The reason is that with fully flexible operation, the electrolysis only runs at times of low prices, thus avoids the hours with high emissions that are characterised by high prices. If the electrolysis is not flexible, it is forced to continue running during these hours with high-price and high-emissions, without the guarantee of hourly matching that forces a corresponding increase in clean production during these times. We further criticise the selective scenario representation of the authors’.

We structure our response as follows: We first address the authors’ claims regarding the role of modelling assumptions for additionality (Section 2) and national renewable generation targets (Section 3) and highlight alternative factors that drive emissions outcomes. We address further issues of the authors’ scenario selection in Section 4 and finally draw a conclusion (Section 5).