The sequence matters: Expert opinions on policy mechanisms for bioenergy with carbon capture and storage

Wähling, Lara-Sophie; Fridahl, Mathias; Heimann, Tobias; Merk, Christine Article — Published Version The sequence matters: Expert opinions on policy mechanisms for bioenergy with carbon capture and storage Energy Research & Social Science Provided in Cooperation with: Kiel Institute for the World Economy – Leibniz Center for Research on Global Economic Challenges Suggested Citation: Wähling, Lara-Sophie; Fridahl, Mathias; Heimann, Tobias; Merk, Christine (2023) : The sequence matters: Expert opinions on policy mechanisms for bioenergy with carbon capture and storage, Energy Research & Social Science, ISSN 2214-6326, Elsevier, Amsterdam, Vol. 103, pp. --, https://doi.org/10.1016/j.erss.2023.103215 , https://www.sciencedirect.com/science/article/pii/S221462962300275X?via%3Dihub This Version is available at: https://hdl.handle.net/10419/275739 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. http://creativecommons.org/licenses/by-nc-nd/4.0/ Energy Research & Social Science 103 (2023) 103215 Available online 27 July 2023 2214-6296/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/). Original research article The sequence matters: Expert opinions on policy mechanisms for bioenergy with carbon capture and storage Lara-Sophie W¨ ahling a , * , Mathias Fridahl b , Tobias Heimann a , Christine Merk a a Kiel Institute for the World Economy, Kiellinie 66, 24105 Kiel, Germany b Department of Thematic Studies, Unit of Environmental Change, Centre for Climate Science and Policy Research, Link¨ oping University, 581 83 Link¨ oping, Sweden ARTICLE INFO Keywords: Bioenergy with carbon capture and storage (BECCS) Negative emissions Net-zero Policy instruments Policy sequencing Expert survey ABSTRACT This paper presents the results of a globally distributed survey on policies for incentivizing bioenergy with carbon capture and storage (BECCS). The current lack of policy incentives to support the implementation of BECCS constitutes a major deployment barrier. Therefore, scientists and policymakers are now debating the optimal BECCS policy framework. Previous studies have presented theoretical analyses of policy options to spur deployment, yet despite the considerable influence of experts on policy processes, very few studies have explored expert opinions. Drawing on an online survey of experts (N =46), we explore their policy preferences and whether those preferences differ or converge between experts from different working sectors. The results show that a tax and refund scheme, a flat-rate subsidy, and reverse auctioning are considered more suitable than other measures. Furthermore, most experts agree that rather than a stand-alone policy, a policy mix would be needed in order to support BECCS deployment. Several experts propose a sequence of policies, moving from publicly funded supply-push policies in the short term to budget-neutral demand-pull policies in the longer term. Regarding various subsidy schemes, respondents favor investment subsidies or results-based subsidies based on stored biogenic carbon dioxide. The relatively minor differences in the response patterns between groups of experts suggest that a consensus on a preferred BECCS policy pathway might be forming across different sectors and interest groups. Therefore, our results could inform policymakers on policy instruments for BECCS that are considered most suitable by experts and thus help to shape the policy pathway for BECCS. 1. Introduction Deploying negative emission technologies (NETs) to remove carbon dioxide (CO 2 ) from the atmosphere will be crucial to achieve the targets of the Paris Agreement [1]. Bioenergy with carbon capture and storage (BECCS) has been highlighted as a key technology to deliver negative emissions and limit global warming to 1.5 ◦C or 2 ◦C [2–4]. BECCS is indispensable in the cost-optimized modelled emissions pathways compiled by the Intergovernmental Panel on Climate Change (IPCC) in its 6th Assessment Report. Emissions pathways that hold global warming at or below 1.5 ◦C by 2100 (>50 %), with limited or no temperature overshoot compared to pre-industrial levels, rely on BECCS to deliver removals in the order of 30–780 GtCO 2 accumulated between 2020 and 2100 [3]. To illustrate the scale of BECCS deployment in these scenarios, the lower bound is about the 2021 annual global CO 2 emission from fossil fuels and industry (37 GtCO 2 ), while the upper bound comes close to the cumulated historical (1750–2021) CO 2 emissions from fossil fuels and industry for the United States (USA), European Union (EU) and United Kingdom (UK) combined (793 GtCO 2 ) [5–7]. In the likely event that global emissions will continue to increase during the 2020s and the carbon budget for 1.5 ◦C is eventually overdrawn resulting in a temperature overshoot, the reliance on BECCS and other methods to remove CO 2 from the atmosphere increases even further [3,4]. BECCS can be applied in processes that combust biomass to generate electricity, heat, pulp and paper, and other goods, or those that use biomass in energy conversion plants to generate biogas or liquid biofuel [8,9]. These facilities are equipped with carbon capture technologies, so that the CO 2 released during biomass conversion is captured, compressed, transported and stored in geological formations to prevent it from entering the atmosphere [8,10,11]. One major advantage of BECCS is that it can offset residual emissions from hard-to-abate sectors (e.g., CO 2 from aviation and shipping, diffuse emissions of methane from agriculture and waste) [1,11–13]. Furthermore, the use of biomass as an energy source can substitute the use of fossil fuels [11,14]. In contrast, * Corresponding author. E-mail address: [email protected] (L.-S. W¨ ahling). Contents lists available at ScienceDirect Energy Research & Social Science journal homepage: www.elsevier.com/locate/erss https://doi.org/10.1016/j.erss.2023.103215 Received 10 March 2023; Received in revised form 14 July 2023; Accepted 15 July 2023 Energy Research & Social Science 103 (2023) 103215 2 relying on biomass as an input for deploying BECCS can result in tradeoffs with land and water, and thus become a threat for achieving the sustainable development goals (SDGs) [15–17]. A common argument is that bioenergy production requires large amounts of woody and cropbased biomass, which means upscaling it could create new competition with land used for food production, leading to higher food prices and affecting food security [10,16,18]. In contrast, other studies argue that depending on specific criterions the growing of perennials for biomass production in agricultural landscapes can be beneficial for biodiversity and ecosystem services [19–21]. Further, to meet the increasing bioenergy demand and to limit the trade-offs with land, there is an ongoing research that residues from forestry and agriculture can be additionally used for bioenergy supply in the future [9,16,22]. Additionally, deploying BECCS might exacerbate water scarcity, because the irrigation of bioenergy plantations leads to water withdrawals putting additional pressure on freshwater systems [16,23–25]. Also, CCS is water intensive and especially the capture process relies on large water consumption [26,27]. So far, BECCS is still for the most part pre-commercial. While a few pilot projects exist, e.g., in North America and Northern Europe, a struggle to identify viable business models for BECCS has limited deployment to a few facilities, often publicly supported [9,28]. Hitherto there has been little progress towards implementation on a large scale. The IPCC projects that pathways limiting global warming to 1.5 ◦C will have to rely on NETs, and thus national policy agendas have begun to recognize the importance of NETs for climate mitigation pathways [2,4]. However, many states do not provide precise information on which NETs they intend to implement or how high their removal potential should be, nor a political roadmap to support them [4]. Therefore, one major barrier to the successful deployment of BECCS is the current lack of policy incentives [29,30]. Incentivizing BECCS would require substantial policy reforms, either by redesigning existing climate policy or defining new policy instruments [31]. Possible policy designs have been discussed in the literature, e.g., various subsidies [32] of which several such policies are already in force, including the US federal government's tax credits for geologically stored CO 2 and the EU's Innovation Fund, extension of the EU emissions trading system (EU ETS) to account for negative emissions [33], a carbon tax with a refund scheme for negative emissions [34,35], quota obligations [36,37] while the Swedish government plans to implement a reverse auctioning system [31,36–38]. Considering suitable policy instruments for BECCS, a distinction may also need to be drawn between net-zero and net-negative emissions. The fact that the Paris Agreement's aspiration to limit global warming to 1.5 ◦C cannot viably be achieved unless the world achieves net-zero CO 2 emissions by the middle of this century, followed by net-negative CO 2 emissions to accommodate for residual methane and nitrous oxide emissions from sources that are hard to abate, has potential implications for BECCS policy [39]. Net-negative CO 2 emissions, on the one hand, require a higher reliance on NETs, and on the other, put a heavy mitigation burden on future generations. From a global effort-sharing perspective, it is also quite likely that emissions in developing countries will peak later and that headroom for their emissions will have to be created by achieving net-zero and net-negative greenhouse gas emissions targets in developed countries [40]. Bednar et al. [41] claim that more stringent policies are necessary when considering intertemporal financial transfer to avoid intergenerational inequity. There has been much discussion among researchers about appropriate policy instruments for incentivizing BECCS, but so far experts directly involved in BECCS policy processes and implementation have rarely been asked about their own views on these instruments. This paper addresses this gap in the literature by asking experts from research, business and industry, and policymaking to assess the policy options discussed in the literature. Therefore, we conducted an online expert survey exploring which policy instruments experts would prefer in general, which ones they would consider suitable in their country of residence, and which ones they feel are most likely to be implemented by policymakers in the future. The survey also included questions on whether the respondents' policy preferences would be the same if the objective were to achieve net-zero emissions compared to net-negative emissions. The experts' extensive know-how of BECCS deployment and the substantial influence of expertise on climate policy processes warrant a closer examination of their policy preferences. Gauging their expertise can be useful both in order to explore appropriate policy incentives for BECCS and to project the direction in which policy might develop. In particular, experts from business and industry, who deal with the implementation of BECCS in practice, will be directly affected by proposed policy instruments, and therefore it would be informative to explore their instrument preferences. In line with this, it is of interest to analyze whether the policy preferences of experts from research, business and industry, and policymaking are similar or diverge. A survey looking at the spread of opinions regarding policy incentives for BECCS is yet lacking. The paper is structured as follows: Section 2 provides an overview of studies that include surveys on BECCS and NETs in general. Section 3 describes the survey presented in this paper, while Section 4 shares the results, with a special focus on differences in the assessments between respondents from research, business and industry, and policymaking. In Section 6, which includes a discussion, we draw conclusions about the level of agreement among experts from different sectors regarding suitable policies for BECCS and explore the implications for the governance of BECCS. 2. Literature review Several surveyand interview-based studies with experts have been conducted in recent years, often with a focus on uncertainties related to modeling BECCS and the feasibility of deploying BECCS on the scale suggested in the technology-cost-optimized scenarios produced by integrated assessment models (IAMs). Rickels et al. [42] conducted an online expert survey with Earth system and integrated assessment modelers to evaluate the importance of NETs in future climate policies, as well as the socio-political, techno-economic and biophysical constraints on NET deployment. They found that, of the NETs assessed, BECCS faced the most constraints, but that nevertheless most experts would include BECCS in a NET portfolio. The need to explore the feasible scale of NET deployment was highlighted by Grant et al. [43], who conducted an online expert survey on the potential of BECCS to remove CO 2 from the atmosphere in the period 2020–2100. The authors conclude that the experts' disparate estimates of BECCS's CO 2 removal potential indicate considerable uncertainty about the future role of BECCS in achieving climate policy objectives. Due to this uncertainty, they conclude that policymakers urgently need to offer additional incentives to reduce emissions and minimize dependency on a future deployment of NETs at scale. Haikola et al. [44] arrive at a similar conclusion. Like Rickels et al., they conducted interviews with Earth system and integrated assessment modelers, but also compared the interview data with survey data from delegates participating in UN climate change conferences. Haikola et al. identify a major discrepancy between expectations about the removal potential of BECCS among the UN delegates and the role of BECCS in the IAM pathways, arguing that the detachment of IAMs from real-world developments in BECCS undermines the models' policy relevance. Additionally, Vaughan and Gough [45] carried out an expert elicitation focusing on assumptions made in IAMs to strengthen the feasibility of BECCS for reaching the 2 ◦C target, with the result that only technical assumptions concerning CCS are deemed to be realistic, while assumptions about bioenergy production, social acceptability and policy framework are considered to be unrealistic. Thus, they conclude, that the representation of BECCS in IAMs needs improvement but also the framework conditions to achieve the targets needs to be clarified in order to avoid an overestimation of BECCS' potential. Based on interviews with integrated assessment modelers as well as critics of IAMs L.-S. W¨ ahling et al. Energy Research & Social Science 103 (2023) 103215 3 from different disciplines, Low and Sch¨ afer [46] determined that the respondents had different interpretations regarding the feasibility of BECCS within modeling, which in turn led to different suggestions on how to reform IAMs. They suggest that future modeling exercises should engage experts from a wider range of disciplines, so as to resolve discrepancies and avoid misunderstandings. Forster et al. [47] held an expert workshop discussing feasibility criteria for BECCS and afforestation, resulting in proposed approaches to account for social and political dimensions in IAMs. Based on this research and additional expert elaborations, Clery et al. [48] gathered a wide range of technical and socio-political criteria for the feasibility of BECCS supply chains that need to be considered in future integrated assessments and by policymakers. In brief, previous studies have mainly focused on the feasibility of implementing BECCS at the scale assumed in the mitigation pathways derived from IAMs. In this regard, previous research emphasizes that the lack of appropriate policy instruments constitutes a major obstacle to BECCS deployment. Political inaction has often been discussed in the literature, but to the best of our knowledge, experts in the field of BECCS have never been asked what policy instruments they feel are desirable and politically feasible. This paper addresses this gap by focusing exclusively on policies for incentivizing BECCS, which are already being discussed in political circles and the scientific community, and by asking experts specifically about their views on these instruments. 3. Method 3.1. Survey design and data collection The data analyzed in this paper was collected through a survey of experts from policymaking, research, and business and industry who were involved in BECCS and/or CCS. The survey was distributed in two versions among two partially overlapping sets of respondents. The first questionnaire included a core set of questions about BECCS policies coupled with an extensive set of questions on expected technical developments and costs. The invitation and two reminders were sent by email in December 2021 and were distributed to 145 international experts, with a response rate of 16 % (N =25). Since the survey consisted of a technical and a more policy-focused part, we invited experts who had mainly technical and/or political expertise regarding the value chain of CCS, BECCS and bioenergy production. In order to find qualified experts in this field, we screened suitable scientific papers and their authors, and searched for contacts on a global level from universities, scientific institutions, consultancies, NGOs, as well as practitioners at CCS and BECCS plants, and power and heat providers dealing with bioenergy. It was considerably easier to find contact data for experts from research than from business and industry, or from policymaking. Therefore, we invited more researchers (N =92) than experts from business and industry (N =48) or policymaking (N =5). As it was not always easy to identify respondents' working sector, we added a question about this to the survey. To increase the number of responses on the policy-related questions, and to further refine the data on policy expectations, a second questionnaire with additional policy-related questions was developed. The second questionnaire and two reminders were sent to 125 experts (including the 25 respondents that answered the first questionnaire) in February and March 2022. Overall, this survey reached a response rate of 25 % (N =30, of which nine also responded to the first survey). The second survey targeted only those experts with a more policy-focused expertise on the value chain of CCS, BECCS and bioenergy production. Here, more experts from business and industry (N =67) were contacted compared to those from research (N =18) or policymaking (N =15). Table 1 shows that in total 46 respondents answered the core set of policy questions, while 30 respondents replied to the expanded set of policy questions. All experts were asked background questions about their occupation, e.g., the sector they primarily worked in, their professional background, and the country of their primary employment. As reported in Table 2, a large share of the respondents was based in Europe, followed by North America. Most of the experts worked in research, followed by business and industry, and policymaking (one expert indicated that none of the sectors applied). Further, multiple responses were possible, so that four experts claimed to work in research and policymaking, while three listed research, and business and industry as their working sectors. In the survey, participants were asked to categorize their expertise, so most participants claimed to have profound knowledge on policy instruments for supporting BECCS (N =21), CO 2 capture (N =21), CO 2 storage (N = 20), costs of CCS (N =16) and technical integration of BECCS (N =16). It can be deduced that most of our respondents were knowledgeable concerning CCS in general and BECCS in particular, and further that a significant share of them were familiar with policy instruments to support BECCS, which is a good basis for evaluating the suitability of policy instruments. Furthermore, half of the participants worked in their expert field for >10 years (N =23) (<2 years: N =3, 2–5 years: N =14, 6–10 years: N =6). 3.2. Questionnaire design Besides questions on the respondents' backgrounds, we asked about the general suitability of various policy instruments to support the deployment of BECCS. The core set of policy-related survey items in the first and second questionnaire were based on the scientific literature on BECCS policy. Experts were asked to assess the suitability of the most frequently discussed policy instruments, including: a general subsidy [32], the integration of BECCS in cap-and-trade systems [33], a carbon tax with a refund scheme [34,35], a quota obligation [36,37], and Table 1 Overview of survey topics and their distribution to the expert groups. Topic Group A Group B N Perceptions of BECCS ● ✓ 30 Suitability of policy instruments ✓ ✓ 46 Necessity of a policy mix ✓ ✓ 46 Suitability of policy instruments in own country of residence ● ✓ 30 Suitability of policy instruments in the context of achieving net-zero/net-negative emissions ● ✓ 30 Suitability of different subsidies ✓ ✓ 43 Likelihood of implementation for different subsidies ✓ ✓ 43 Appropriateness of reverse auctioning ✓ ✓ 43 Note: Group A includes experts who responded only to the core set of policyrelated questions (16 experts), while group B include experts who responded to the core and expanded policy-related questions alike (30 experts). The number of respondents, N, could be less than A plus B, as not all respondents answered all questions. Table 2 Respondents (N =46) divided by primary sector of work and region. Research Business and industry Policy Multiple responses Not stated North America 3 3 ● 2 ● Europe 13 8 4 3 1 Rest of the world 1 ● ● 1 ● Not stated 5 ● 1 1 ● N 22 11 5 7 1 Note: North America includes Canada and the USA; Europe includes Austria, Belgium, Denmark, Germany, Italy, the Netherlands, Norway, Sweden and the UK; Rest of the world includes Brazil and South Korea. L.-S. W¨ ahling et al. Energy Research & Social Science 103 (2023) 103215 4 reverse auctioning [31,36–38]. Table 3 shows the explanations of the various policy instruments, which were also outlined to the experts during the survey. The expanded set of questions in the second questionnaire also included items on whether the respondents' assessment of the policy instruments would change if they had to differentiate between a global context and their country of residence. In addition, participants were asked to assess these policies in the context of net-zero emissions targets compared to net-negative emissions targets. In those cases where assessments changed with the context, participants were asked to explain the differences. In addition to the suitability of policy instruments, the questionnaire included a set of questions related to subsidies, reverse auctions, and the need for a policy mix to support BECCS. Since subsidy schemes can take on a variety of forms, we included a question about the suitability of five specific flat-rate subsidies for BECCS, i.e., for (1) biomass production, (2) bioenergy, (3) BECCS investment, (4) emissions stored with BECCS (i.e., a results-based payment), and (5) emissions stored with CCS (with no distinction between fossil and biomass sources) [32,49]. Furthermore, we asked about the likelihood of the respective subsidies being implemented by policymakers in the future. This allowed us to contrast the assessment of the subsidy schemes' suitability with the assessment of the likelihood of their implementation. Also, the core set of questions included questions about the experts' views on reverse auctioning, whether they felt it was an appropriate measure, and the reasons for their response. In the extended questionnaire, we asked experts to agree or disagree with statements concerning BECCS as a suitable measure to address global warming, its contribution to meet the well below 2 ◦C goal, its contribution to mitigation actions in their country of residence, and last but not least, whether or not BECCS should be incentivized by means of policies. The response scales for questions that asked about the experts' views on specific topics were four-point Likert scales ranging from “disagree strongly/very unsuitable/very unlikely” to “agree strongly/ very suitable/very likely.” 4. Results of the expert survey 4.1. Perceptions of BECCS To obtain a picture of how the experts perceived the potential of BECCS, they were asked to agree/disagree with four different statements, shown in Table 4. The respondents generally agreed that BECCS was both a suitable measure to address global warming and that BECCS should be incentivized by means of policies. The high level of agreement was especially true for experts from business and industry, of whom none disagreed with either of the two statements. The responses of experts from research and policymaking were also overall positive, and only a small share of experts disagreed with at least one statement. For the statements about BECCS's contribution to meeting the well below 2 ◦C target globally and its contribution to mitigation actions in the respondents' country of residence, the responses were mainly positive. However, the level of agreement with these two statements was markedly lower compared to the level of agreement on the suitability of BECCS to address global warming in general, and the support for policy incentives. We find again that experts from business and industry tend to agree more with the statements on BECCS's contribution to the 2 ◦C target and to domestic mitigation actions in comparison to experts from research and policymaking, as shown in Table 4. While for both statements the response pattern diverged slightly between experts from business and industry compared to experts from research and policymaking, the Kruskal-Wallis test 1 revealed that the differences were not statistically significant (p: 0.1147 Table 4 Mean agreement with BECCS statements by working sector (Scale: 1 disagree strongly; 2 disagree slightly, 3 agree slightly, 4 agree strongly). Research Business and industry Policy Total Mean N Mean N Mean N Mean N BECCS is a suitable measure to address global warming. a 3.18 11 3.50 8 3.00 4 3.26 23 Globally, BECCS is likely to contribute substantially to meeting the well below 2 ◦C goal. b 2.64 11 3.50 8 2.50 4 2.91 23 In my country of residence, BECCS is likely to contribute substantially to mitigation actions. c 2.73 11 3.14 7 2.00 4 2.72 22 BECCS should be incentivized by policies. d 3.45 11 3.87 8 3.00 4 3.52 23 Note: “Don't know” answers and seven experts who indicated multiple working sectors were excluded. Applied Kruskal-Wallis test: H 0 =No difference in responses between working sectors. a p: 0.3681, b p: 0.1147, c p: 0.1517, d p: 0.1126. Table 3 Overview of policy instruments addressed in the survey, which were also available to the experts. Policy instrument Explanation Subsidy BECCS operators receive a financial benefit from the government to support the deployment phase. A distinction is made between different forms of subsidies: subsidizing investment, bioenergy production, biomass usage, or with focus on emission reductions, i.e., a subsidy on emissions stored with BECCS and a subsidy on emissions stored with CCS with no distinction between fossil and biomass sources [32,49]. Integration in a cap-andtrade system A cap on CO 2 emissions forces emitters to surrender emission allowances in accordance to their amount of emissions. Those allowances would reenter the trading pool when CO2 is removed from the atmosphere [33]. Carbon tax combined with a refund scheme A carbon tax would set a price on fossil and biogenic carbon. But a standalone carbon tax would not incentives emission reductions beyond zero. A possible solution might be a linking of a tax with a refund scheme, where a BECCS plant would receive a refund for each ton of CO2 removed from the atmosphere [34,35]. Quota obligation Emitters of specific sectors are obliged to purchase BECCS units in proportion to their emission release or directly finance BECCS projects and subtract those negative emissions from their own emissions [36,37]. Reverse auctioning The traditional role of buyers and sellers is inversely so that the reverse auctioning differentiates between one buyer (here the government) and many sellers (here actors that offer negative emissions through BECCS). Normally, the lowest bidder wins the auction and receives a differentiated guarantee price for storing CO2 from biogenic sources. Overall, the purpose of a reverse auctioning is to serve as an investment aid for BECCS operators [31,36–38]. 1 The Kruskal-Wallis H test was used to detect statistically significant differences in the response patterns among experts from research, business and industry, and policymaking, as well as among experts with different locations of their primary work. The Kruskal-Wallis test is a nonparametric procedure used to determine whether the tendencies of a variable differ between two or more independent samples [50]. L.-S. W¨ ahling et al. Energy Research & Social Science 103 (2023) 103215 5 for BECCS's contribution to the 2 ◦C target, and p: 0.1517 for BECCS's contribution to domestic mitigation actions). Particularly experts located in Sweden, the UK and the USA agreed that BECCS would contribute substantially to mitigation actions in their country of residence, while experts in other European countries mostly disagreed (e.g., Germany, Belgium and Austria), which is according to the Kruskal-Wallis test significant (p: 0.0216). 2 When asked what motivated their response, one expert argued that there was no room for BECCS in densely populated regions. Other experts also stated that a few countries (e.g., the UK, Sweden and Finland) had substantial potential for implementing BECCS because industrial plants already in place used biomass at scale and were suitable for a BECCS retrofit. Further, several experts indicated that meeting the targets would be difficult without CO 2 removal and that BECCS was one of the most advanced technologies for carbon removal. On the other hand, they also argued that BECCS was unlikely to contribute to climate change mitigation on a global scale due to challenges regarding upscaling, and that its contribution to reaching the 2 ◦C target would therefore be limited: “BECCS is one among many solutions for climate change mitigation. There are many challenges to scaling up BECCS so it will likely not be a substantial contributor to climate change mitigation on a global scale, although it could contribute on a local and national level.” Challenges for the large-scale deployment of BECCS include the conditions for long-term CO 2 storage and land-use implications (e.g., resource competition for land and water) due to the increased biomass use that would result from upscaled BECCS. Here, several experts emphasized the importance of ensuring the sustainability of biomass production: “BECCS is part of the tools to reach carbon neutrality, but it cannot be used to compensate inaction. First the priority is to tackle fossil emissions, with CCS as last resort. Secondly, to ensure sustainable biomass production. Only then, BECCS should have role to play.” On average, especially researchers agreed that BECCS was a suitable method to address global warming, but were more skeptical as to whether it would substantially contribute to meeting the well below 2 ◦C goal. 6 of 17 researchers agreed less with the second statement compared to the first (see Table 4). In contrast, among respondents from business and industry, only 1 of 10 gave different answers on the two statements. Despite being skeptical about the scope of BECCS's contribution to global climate change mitigation, most experts agreed that it needed to be incentivized by means of suitable policy instruments: “There has been a significant volume of independent academic and thirdparty research over recent years that indicates that BECCS will play a critical role to address residual emissions in hard-to-abate sectors across the global economy. Whilst over the long-term a liquid market for carbon removals should provide developers with the revenue streams and price signals required to develop BECCS projects, in the short-term some form of policy intervention will be required given the capital and operational costs of carbon capture technology.” 4.2. Suitability of policy instruments for supporting BECCS Fig. 1 shows that most experts ranked a carbon tax combined with a refund scheme as the most suitable option for supporting BECCS, followed by a general subsidy. The level of agreement on the suitability of a carbon tax combined with a refund scheme was fairly high and only a few experts (N =3) felt the policy was unsuitable. The assessments of the integration in a cap-and-trade system, a quota obligation, and reverse auctioning were more diverse. The majority of experts indicated that an integration in a cap-and-trade system or a quota obligation would be suitable means of supporting BECCS, although nearly one third of the experts felt these policies were unsuitable. Looking at reverse auctioning, it is striking that one third of experts responded “Don't know” – indicating that they felt that they did not know enough to assess the option's suitability. The remaining experts considered reverse auctioning to be a suitable policy (N =25). If the “Don't know” answers were excluded from the analysis, reverse auctioning would be the second most suitable instrument after a carbon tax together with a refund scheme in the ranking. In the open-text answers, several experts indicated that reverse auctioning was a suitable instrument for supporting BECCS at an early stage because it would enable government to support projects at the lowest possible cost and provide predictable financial flows for the operators. Nevertheless, from an operator's point of view, reverse auctioning bears the risk of putting considerable time and effort into the preparation of a bid, only to have it rejected: “If the purpose is for the government to achieve low-cost BECCS then reverse auctioning probably is an appropriate instrument. For me as an actor it's a matter of reducing risks and with reverse auctioning I have to be very sure [to] have counted in all those risks that might appear to leave an offer. That requires a lot of knowledge and investment in time and effort and the possibility that my offer might be refused.” Reverse auctioning could support the market entry of BECCS, but the experts also emphasize that it could only be effective for a limited time because of the limited opportunity to use public finance to create a sufficiently large market for upscaled BECCS. If a reverse auction were not combined with a quota obligation, a tax or a similar instrument to generate government revenues for financing the auctions, it would be costly for taxpayers in the long run. Therefore, reverse auctioning was often supported as part of a policy mix, an aspect that we will discuss in detail below. Table 5 summarizes the respondents' views on the suitability of different policy instruments by working sector. The response patterns diverge slightly between working sectors: experts from business and industry categorized most policies as suitable, whereas those from research and policy were more reserved, particularly regarding a quota obligation and a cap-and-trade system. However, the Kruskal-Wallis test did not identify significant differences in the responses of experts from the three different working sectors for any of the policies (see Table 5 for p-values). Considering all respondents, the Wilcoxon signed-rank test 3 revealed that a carbon tax and refund scheme was significantly preferred over a cap-and-trade system (p: 0.0021) or quota obligation (p: 0.0184). Comparing the sectors shows that this result was mainly driven by researchers. According to the Wilcoxon singed-rank test, researchers significantly preferred a carbon tax with refund scheme over a subsidy (p: 0.0104), cap-and-trade system (p: 0.0023) or quota obligation (p: 0.0086). Furthermore, they significantly preferred reverse auctioning over a quota obligation (p: 0.0263) or cap-and-trade system (p: 0.0263). In contrast, experts from policy and business and industry did not show a significant preference for any policy measure (p: 0.0918–1 for experts from policy, p: 0.1585–1 for experts from business and industry). As mentioned above, one set of questions concerned the experts' opinions on specific subsidies. Fig. 2 presents the assessment of their suitability and the likelihood of their implementation for the various 2 For the other statements and policy instruments, the experts' assessments did not differ significantly based on their location. Therefore, we did not explore it any further in this study. 3 The Wilcoxon signed-rank test was used to detect differences in the response patterns among experts from the same working sector. This allowed us to determine whether the response patterns for the suitability of two policy instruments were significantly different among experts from the same group. The Wilcoxon signed-rank test is another nonparametric test used to determine whether the main tendencies of a variable differ between two or more independent samples [51]. L.-S. W¨ ahling et al. Energy Research & Social Science 103 (2023) 103215 6 flat-rate subsidies. The experts ranked a results-based subsidy based on emissions stored with BECCS as the most suitable option, followed by an investment subsidy. The Wilcoxon signed-rank test indicated that subsidies on emissions stored with CCS (p: 0.0 for subsidy on BECCS/CCS, p: 0.0003 for subsidy on investment/CCS), and on bioenergy (p: 0.0 for subsidy on BECCS/bioenergy, p: 0.0 for subsidy on investment/bioenergy) or biomass production (p: 0.0 for subsidy on BECCS/biomass production, p: 0.0 for subsidy on investment/biomass production) were considered significantly less suitable. It is interesting to note that roughly half of the respondents felt that subsidizing CCS alone would not be sufficiently effective to support the implementation of BECCS. Only 47 % of the experts considered this subsidy to be suitable, while nearly 90 % considered a subsidy on emissions stored with BECCS to be suitable. In comparison, the experts claimed that an investment subsidy was most likely to be implemented, followed by a subsidy on emissions stored with BECCS, although for the latter the assessment of the likelihood was lower than that of the instrument's suitability. However, a Wilcoxon sign-rank test showed that, in these cases, there was no significant difference between the responses on suitability and likelihood of implementation (p: 0.2406). Furthermore, the assessments of a subsidy on biomass production and a subsidy Table 5 Mean assessment of the suitability of policy instruments for supporting BECCS by working sector (Scale: 1 very unsuitable; 2 rather unsuitable, 3 rather suitable, 4 very suitable). Research Business and industry Policy Total Mean N Mean N Mean N Mean N Carbon tax with a refund scheme a 3.50 22 3.30 10 2.75 4 3.36 36 Subsidy b 2.95 22 3.44 9 3.25 4 3.11 35 Cap-and-trade system c 2.71 21 3.25 8 2.25 4 2.78 33 Reverse auctioning d 3.29 14 3.14 7 2.50 2 3.17 23 Quota obligation e 2.63 19 2.87 8 2.00 4 2.61 31 Note: “Don't know” answers and seven experts who indicated multiple working sectors were excluded. Applied Kruskal-Wallis test: H 0 =No difference in responses between working sectors. a p: 0.2743, b p: 0.3771, c p: 0.1856, d p: 0.8346, e p: 0.3147. Fig. 2. Comparison of the suitability of different subsidies (blue bar) and the likelihood of their being implemented by policymakers (red bar) (“Don't know” responses excluded). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) Fig. 1. Results of the assessment on the suitability of different policy instruments for supporting BECCS. L.-S. W¨ ahling et al. Energy Research & Social Science 103 (2023) 103215 7 on bioenergy diverged slightly. As mentioned above, the subsidies were considered to be less suitable than the other suggested subsidies. Despite this assessment, roughly half of the experts considered both subsidies likely to be implemented by policymakers. Here as well, we employed the Wilcoxon sign-rank test, which revealed that the likelihood of the subsidies' implementation was considered to be significantly higher than their suitability (p: 0.0124 for suitability/likelihood of subsidy on biomass production, p: 0.0064 for suitability/likelihood of subsidy on bioenergy). Table 6 compares the assessment of suitability between experts from research, business and industry, and policy, and we can see that the responses differ slightly. In particular, the assessment of a subsidy on biomass production diverges, with experts from policy considering it to be substantially more suitable compared to experts from research and business and industry. However, the Kruskal-Wallis test indicated no significant variation between experts from the different working sectors (see Table 6 for p-values). The Wilcoxon signed-rank test revealed that experts from research and business and industry significantly preferred an investment subsidy and a subsidy on emissions stored with BECCS over the three other subsidies (p: 0.005–0.0872 for experts from research, p: 0.0066–0.0276 for experts from business and industry). Further, experts from research significantly favored a subsidy on emissions stored with CCS over a subsidy on bioenergy (p: 0.0112) or biomass production (p: 0.0274), while experts from business only preferred such a subsidy over a subsidy on biomass production (p: 0.0276). Comparing the assessment of the likelihood of implementation between experts from the different working sectors, Table 7 shows that the response patterns diverged significantly for a subsidy on bioenergy (p: 0.0216). Additionally, the assessment of the likelihood of a subsidy on bioenergy being implemented differed significantly between experts from business and industry and those from policy (p: 0.0283), as well as between experts from research and those from policy (p: 0.0126). We also asked the experts if they would change their ranking of the suitability of policy instruments to support BECCS depending on whether the target were net-zero emissions or net-negative emissions. Only six experts (N =30) across all three working sectors indicated that they would change their ranking. While 17 experts claimed they would not rank the suitability of the policy instruments differently, seven selected the response “Don't know.” Those experts who would change their ranking mainly argued that achieving net-negative emissions would require more ambitious measures and that policies would therefore have to be more stringent: “BECCS for offsetting residual emissions is structurally different from BECCS for providing net negative emissions. All of the above schemes, if appropriately adjusted, can in principle incentivize CDR [CO 2 removal] at no additional costs for tax-payers if CDR only balances residual emissions. None of the schemes, however, are suitable for net negative emissions without consideration of intertemporal financial transfers/ intertemporal emission trading.” Given how heterogeneous the adjusted rankings were, it is difficult to make conclusive remarks on the rankings of the different policy instruments for achieving net-negative targets compared to net-zero targets. 4.3. Need for a policy mix Further, the experts were asked if they felt that a policy mix was necessary to support the implementation of BECCS. Roughly 83 % of experts from all working sectors agreed to this. Those experts who supported a policy mix were subsequently asked, in an open-ended question, to suggest which policy mix they considered to be most suitable for supporting BECCS. As Fig. 3 shows, most experts proposed a policy mix containing the above-mentioned policies and including policies with direct financial support from the government, e.g., a carbon tax (if combined with a refund scheme), a general subsidy, or reverse auctioning. They included market-based instruments as complementary policies, especially capand-trade and a quota obligation. Interestingly, a quota obligation was mentioned comparatively often even though it had been categorized as the least suitable instrument in the previous questions (see Fig. 1). The most frequently suggested policy mix combines a carbon tax with a quota obligation and a carbon tax with a refund scheme, followed by reverse auctioning with either a quota obligation or a cap-and-trade system. A few experts also suggested a sequencing of the policy mix. One expert proposed a subsidy combined with public procurement as a suitable mix for an early phase of BECCS deployment, followed by a market-based mechanism once BECCS technology and the corresponding market were more mature. Another example: introduction of reverse auctioning at the beginning, followed by a cap-and-trade system or quota obligation. Furthermore, some experts called for other measures such as public procurement, government-funded research, mechanisms for CCS (including infrastructure build-up and regulations), as well as regulations for guaranteeing sustainable biomass production. In general, several experts claimed that the ideal policy selection depended on the government's preferred scale and scope: Table 6 Mean suitability of different subsidies by working sector (Scale: 1 very unsuitable; 2 rather unsuitable, 3 rather suitable, 4 very suitable). Research Business and industry Policy Total Mean N Mean N Mean N Mean N Investment subsidy a 2.90 20 3.80 10 3.33 3 3.21 33 Subsidy on emissions stored with BECCS b 3.30 20 3.70 10 3.00 5 3.37 35 Subsidy on emissions stored with CCS c 2.57 19 2.50 10 2.40 5 2.53 34 Subsidy on bioenergy d 2.00 21 2.55 9 2.50 4 2.21 34 Subsidy on biomass production e 1.95 21 1.77 9 3.20 5 2.08 35 Note: “Don't know” answers and seven experts who indicated multiple working sectors were excluded. Applied Kruskal-Wallis test: H 0 =No difference in responses by working sector. a p: 0.0727, b p: 0.1963, c p: 0.9326, d p: 0.2959, e p: 0.0553. Table 7 Mean likelihood of different subsidies being implemented by working sector (Scale: 1 very unlikely; 2 rather unlikely, 3 rather likely, 4 very likely). Research Business and industry Policy Total Mean N Mean N Mean N Mean N Investment subsidy a 2.85 20 3.22 9 3.25 4 3.00 33 Subsidy on emissions stored with BECCS b 2.89 19 3.37 8 2.75 4 3.00 31 Subsidy on emissions stored with CCS c 2.84 19 2.75 8 2.50 4 2.77 31 Subsidy on bioenergy d 3.05 19 2.71 7 1.75 4 2.80 30 Subsidy on biomass production e 2.85 20 2.00 9 2.25 4 2.56 32 Note: “Don't know” answers and seven experts who indicated multiple working sectors were excluded. Applied Kruskal-Wallis test: H 0 =No difference in responses by working sector. a p: 0.4246, b p: 0.2831, c p: 0.5187, d p: 0.0216, e p: 0.0521. L.-S. W¨ ahling et al. Energy Research & Social Science 103 (2023) 103215 8 “All of the policies listed above could have a role in supporting the implementation of BECCS, but the mix of policies that are most relevant vary by national and regional context.” 5. Discussion In this expert survey, a carbon tax with a refund scheme received the highest agreement among experts to be a suitable option for supporting the commercial deployment of BECCS, followed by a general subsidy. Looking more closely at subsidies, the experts preferred technologyspecific subsidies, since they considered subsidizing BECCS directly to be much more suitable than subsidizing CCS in general. On the one hand, Bellamy et al. [36] argue that deciding on the technology specificity of a subsidy reveals a dilemma of incentives, as in this case a subsidy for BECCS alone would undermine the exploitation of possible synergy effects between BECCS and fossil fuels with CCS. On the other hand, a general CCS subsidy might more effectively promote the deployment of fossil CCS, as BECCS is less mature than CCS and therefore associated with higher investment risks. Another very likely explanation for a preference to subsidize BECCS and not fossil CCS is provided by Fridahl et al. [30], who show how BECCS and CCS differ in two principal ways. First, BECCS provides a public good. This motivates support policy that rewards deployment. In contrast, fossil CCS should be incentivized by policies that penalize fossil emissions, so as to internalize a negative externality into the cost of production. Second, since fossil CCS is already at least partially incentivized by means of various instruments such as carbon taxes and cap-and-trade systems, attention should be directed towards technology-specific BECCS policies. If not, the chances that BECCS will complement fossil CCS as a mitigation option are low. It is likely that our participants share this view of the existing policy landscape and concerns regarding the competitiveness of BECCS compared to fossil CCS. This would also help those governments aspiring to develop technology-specific supply-push policies for BECCS to advance CO 2 removal. With regard to reverse auctioning, we found a lack of knowledge about, and experience with, this instrument. On the one hand, one third of the experts selected the “Don't know” response and did not assess the instrument at all. On the other hand, the remaining experts mostly ranked it as suitable. Some participants commented that reverse auctioning was particularly suitable for the early implementation phase, which they justified by the fact that costs are minimized through competitive bidding. In addition, long-term contracts could reduce operators' uncertainty about financing the BECCS plants. These statements are in line with Lundberg and Fridahl [38], who argue that reverse auctioning is a suitable interim policy for BECCS market entry, as cost efficiency and stable revenues provide robust planning horizons for the operators and investors. In our survey, some experts claimed that reverse auctioning was not a long-term policy due to the high cost borne by governments and, ultimately, taxpayers. Zetterberg et al. [37] stress this argument and highlight a sequential policy approach as a potential solution. They suggest that, after an early stage of reverse auctioning alone, the policy portfolio could be extended by quota obligations for scaling up BECCS, and for generating revenues to finance additional reverse auctions. This sequential policy mix was also proposed by some experts in our survey. While the survey participants considered a quota obligation as a stand-alone measure to be less suitable compared to other policies, roughly half of them suggested policy mixes that include quota obligations, most often together with a carbon tax and reverse auctioning. A plausible explanation for these seemingly contradictory views is that quota obligations are considered problematic when used as a standalone instrument, whereas they can work well if combined with other instruments. Take publicly funded flat-rate subsidies or reverse auctions, for example: quota obligations could mitigate the problem of raising long-term public finance for subsidies, regardless of whether they are flat-rate subsidies or subsidies allocated via reverse auctions. Sectors with hard-to-abate emissions, such as agriculture and international aviation, could be obligated to buy and surrender BECCS certificates generated from the auctions, so as to comply with an annual BECCS quota specified as a share of their total emissions. As such, the quota obligation would reduce demand for emissions-intensive goods and services, such as meat and dairy products or flights, and would help to raise private finance for BECCS. A similar policy option is explored by Jenkins et al. [52], who argue that extractors of fossil carbon should be obliged to demonstrate long-term storage of CO 2 corresponding to a quota of the fossil carbon they extract. The quota would increase over time, and regulators could require that a share of the quota be delivered through BECCS. However, Zetterberg et al. concede that while in general a sequential policy approach would be promising, it would also be highly complex to set up. Generally, the response patterns of experts from research, business and industry, and policy were similar and we observed significant differences only for a few questions. One difference emerged for the assessment of an integration of BECCS in a cap-and-trade system. While respondents from business and industry and from policy did not feel that any policy instrument was more suitable than the others, researchers preferred a carbon tax combined with a refund scheme, and reverse Reverse auctioning Subsidy Carbon tax Cap-andtrade Regulation on biomass sourcing Investment funds Carbon tax 1 0 0 0 0 0 Cap-and-trade 2 1 1 0 0 0 Quota obligation 2 1 3 2 0 0 Market-based mechanism 0 1 1 0 0 1 Refund scheme 0 0 3 0 0 0 Public procurement 0 1 0 0 0 0 Price stabilization mechanism 0 0 0 1 0 0 Strategies for CCS 0 1 1 0 1 0 Research 1 0 1 1 0 0 Fig. 3. Overview of the frequency of proposed combinations of different policies and/or other mechanisms. The rows and columns respectively show proposed policies/mechanisms to be combined by the experts. L.-S. W¨ ahling et al.