Developing net-zero carbon supply chains in the European manufacturing industry – a multilevel perspective

Abstract

Purpose

The European Union (EU) and European companies are striving for net-zero carbon targets by 2050 and are therefore focused on urgent decarbonization efforts. Manufacturing contributes to 20% of European carbon emissions, although the primary challenge lies in supply chain (SC) emissions, which highlights the field's need to transform. Amid the dissonance between public and private net-zero commitments and persistent carbon emissions, uncertainties surround the development of net-zero carbon supply chains (NZCSCs). This paper aims to address this lack of knowledge by presenting an exploration of the development of NZCSCs within the EU through 2050.

Design/methodology/approach

Using a real-time Delphi methodology and tool from durvey.org, this study involves a multiphase panel discussion process with 67 SC and sustainability experts. Twelve prospective theses for NZCSC development in the EU were formulated through desk research, interviews and an expert workshop. The panel assessed these theses in terms of impact, desirability and anticipated occurrence year and provided justification for their evaluations.

Findings

The study identifies three clusters that influence NZCSC development, comprising 68 implications that scholars, managers and policymakers should consider during this transition.

Originality/value

This study contributes to the available information regarding NZCSCs by offering insights from a multilevel perspective into the influences on NZCSC development in the EU's manufacturing sector.

Citation

Steiner, B., Münch, C., Beckmann, M. and von der Gracht, H. (2024), "Developing net-zero carbon supply chains in the European manufacturing industry – a multilevel perspective", Supply Chain Management, Vol. 29 No. 7, pp. 164-181. https://doi.org/10.1108/SCM-06-2024-0372

1. Introduction

In response to escalating concerns regarding global warming's environmental impacts − such as diminished crop yields, severe weather events and potential political destabilization (IPCC, 2022) − 197 countries signed the Paris Agreement in 2015. Translating this global ambition to the regional level, the EU Green Deal is aimed toward a 55% reduction of carbon emissions by 2030 and net-zero emissions in the EU by 2050 (European Commission, 2019). Despite ambitious commitments both globally and within the EU, decarbonization targets remain unmet, with global greenhouse gas (GHG) emissions rising from 11 billion tons in 1960 to 36.8 billion tons in 2023 (Budget, 2023). To adhere to the 1.5°C warming limit, drastic measures are essential, as current trends fall short: global carbon intensity decreased by only 2.5% in 2023, whereas a 6.5% annual reduction is needed to meet the 2°C target, and a 17.2% reduction is required for the 1.5°C goal (PwC, 2023). This significant divide underscores the urgent need for transformative approaches to decarbonization.

Companies contribute to approximately 70% of global GHG emissions (CDP, 2023a); consequently, reducing their corporate carbon footprint is vital for overcoming the disparity between ambition and implementation. However, reducing a company's carbon footprint extends beyond its direct operations (Scope 1) to include emissions from purchased energy (Scope 2) and its entire value chain (Scope 3), which covers both upstream (e.g. materials, inputs) and downstream (e.g. transport, usage, end-of-life) emissions (Ranganathan et al., 2004). Scope 3 emissions constitute approximately 75% of total emissions across industries (CDP, 2023a, p. 6). A comprehensive supply chain (SC) perspective is, therefore, essential for effective GHG reduction. Researchers have explored various practices for decarbonizing value chains, from supplier relationship development and carbon pricing to the role of technology (Steiner et al., 2023; Xia et al., 2018; Xu et al., 2023; Zhang and Wang, 2022). Yet, mere incremental improvements are insufficient to eliminate the discrepancy between ambitious reduction targets and actual decarbonization. A systemic, disruptive change is required, shifting manufacturing from a fossil-fuel base to a low- or zero-carbon model. The multilevel perspective (MLP) is a valuable framework for analyzing such transformations (Geels, 2006).

From the MLP framework, companies' decarbonization efforts interact with evolving socio-technological landscapes and are influenced by technological innovations, stakeholder expectations, legislation and industry behaviors (Geels, 2019). To date, however, little is known about how this field will develop and how corporate decarbonization efforts can shape it. This research, which focuses on EU manufacturing, explores this uncertainty. The research question is this: How can net-zero carbon supply chains (NZCSCs) develop within the socio-technological landscape of European manufacturing until 2050?

The contributions of this paper are threefold. First, this study investigates the influences on the development of NZCSCs in the EU as a multilevel phenomenon. The clusters identified through the analysis are valuable tools for informed decision-making when formulating effective strategies to achieve carbon neutrality goals. The study also presents 68 implications that are relevant to researchers, practitioners and policymakers who are developing NZCSCs.

2. A multilevel perspective of supply chain decarbonization

A prominent theory in the literature that effectively captures the complexity of radical decarbonization is the MLP, which is often used in transition studies (Geels, 2006). The MLP provides a robust framework for comprehending socio-technological transitions, such as the movement toward sustainable practices across various industries. At its core, the MLP examines the interplay across three distinct levels: the niche, the regime and the landscape (Geels, 2006).

The niche level is where radical innovations become established. These new developments are often insulated initially from market pressures, thus allowing new ideas and technologies to emerge and evolve (Geels, 2006). This level is crucial for the novel technologies that are central to the analysis and are examined in greater depth.

The regime represents the confluence of dominant industry practices, cultural norms, regulatory frameworks and established technologies that define the current system. Due to their entrenched values, norms and infrastructures, regimes often exhibit resistance to change (Geels, 2006). In this study on SC decarbonization, the existing regime comprises the actors, institutions and practices that have historically supported fossil fuel-based manufacturing SCs. The analysis focuses on changes within the regulatory regime (policies) and the industry regime (market stakeholders), aiming to establish decarbonized value chains as the new standard.

Finally, the landscape level encompasses the broader socioeconomic and political context, which includes long-term trends, macroeconomic conditions, societal values and environmental challenges. Shifts at this level, such as those in public opinion, global crises or environmental impacts, exert pressure on existing regimes that open pathways for niche innovations to emerge and disrupt established systems (Geels, 2006). In the analysis, relevant landscape pressures consist of political decarbonization commitments, such as the EU Green Deal’s targets discussed in the introduction (European Commission, 2019); society’s heightened awareness of the climate crisis, which includes social movements such as Fridays for Future, Extinction Rebellion and Last Generation; and the direct consequences of climate change on SCs, such as the physical risks posed by extreme weather events for sourcing and logistics (Er Kara et al., 2021; Ghadge and Seuring, 2020).

In addition, it is crucial to integrate the United Nations’ Sustainable Development Goals (SDGs) (United Nations, 2015) into the discussion of the subject matter, particularly when addressing NZCSCs by 2050. A key aspect to consider is the alignment with SDG 12, which focuses on responsible consumption and production (United Nations, 2015). The relevance of this goal becomes particularly significant when examining upstream processes, as they play a central role in ensuring responsible sourcing and the sustainable production of raw materials, components and parts. These processes, embedded within the manufacturing scope, are critical drivers in achieving sustainable development. By minimizing environmental impact and promoting resource efficiency, they directly influence the sustainability of future SCs. Therefore, discussing the alignment between SDG 12 and the transition to NZCSCs highlights the interconnectedness of responsible consumption, production practices and the broader global sustainability objectives. In following the MLP, 12 prospective theses were categorized into three key areas: policy regime changes, market stakeholder regime changes and niche innovations regarding novel zero-carbon technologies. Moreover, these prospective theses were linked to the SDGs whenever possible.

2.1 Changes in the policy regime (P1–P4)

In the policy regime context, four prospective theses (P1–P4) regarding changes for measuring and reducing emissions throughout manufacturing value chains were examined.

Measuring and disclosing carbon emissions is foundational to their management and reduction (Azadi et al., 2020; Dahlmann et al., 2023). The European Commission's recent proposal on battery transparency exemplifies this fact, marking a move toward regulating the mandatory disclosure of product carbon footprints (PCFs) across the EU (European Commission, 2023). The transparency on PCFs enables low-carbon procurement, which is connected to SDG 12 (United Nations, 2015), namely, sustainable consumption. Based on this precedent, the following is proposed:

The product carbon footprint must be disclosed for every product, including finished products, intermediaries, and raw materials. (P1)

A significant policy that encourages decarbonization is carbon pricing (Teixidó et al., 2019), as exemplified by the EU's emissions trading system's (ETS) cap-and-trade regulation (European Union, 2003). Despite its implementation, the exclusion of certain sectors and a carbon price insufficient for achieving the Paris Agreement targets highlight the need for more inclusive and aggressive pricing mechanisms. The following is suggested:

Carbon pricing mechanisms are implemented in the manufacturing industries with carbon prices aligned with the 1.5°C reduction path. (P2)

Governmental subsidies complement carbon pricing by directing industries toward decarbonization (Meckling et al., 2017). Germany's financial incentives exemplify national efforts (Federal Government of Germany, 2021), although the broader application across the EU remains uncertain. The following is proposed:

Governments have established positive incentives for carbon reduction measures within the EU. (P3)

The circular economy's (CE) 9R strategies − refuse, rethink, reduce, reuse, repair, refurbish, remanufacture, repurpose, recycle and recover − (Potting et al., 2017) offer significant potential for decarbonization (Meini et al., 2021). To reduce GHG emissions, the European Commission thus embraces an agenda that promotes a CE (European Commission, 2020). An EU policy initiative includes a draft regulation for specific products, such as large batteries, that require the inclusion of secondary materials − 16% cobalt, 6% lithium and 6% nickel − by 2030 (European Union, 2023). As the development of the CE promises economic opportunities (European Commission, 2020), this is connected to SDG 8 and 12 (United Nations, 2015), advancing sustainable production while laying the foundation for economic growth. Anticipating future developments, the following is proposed:

Comprehensive circularity regulations in the manufacturing industry are implemented within the EU. (P4)

2.2 Changes in the market stakeholder regime (P5–P8)

In the market stakeholder regime, four prospective theses (P5–P8) that are identified as significant through the literature review and expert interviews were explored.

The European industrial sector is a major consumer of electricity within the EU (Eurostat, 2023). To reduce their Scope 2 emissions to zero, manufacturers and suppliers are heavily dependent on decarbonizing the electricity mix (Victoria et al., 2020). To align with SDG 7, which aims to ensure access to affordable, reliable, sustainable and modern energy for all, the EU has committed to significantly increasing the share of renewable energy in its energy mix (United Nations, 2015). This transition is critical for the industrial sector’s shift toward 100% nonfossil energy technologies. Furthermore, achieving SDG 7 targets necessitates substantial investments in clean energy infrastructure and the implementation of supportive policies to accelerate the adoption of renewables (United Nations, 2020). Currently, however, the EU’s electricity mix continues to depend significantly on carbon. Anticipating necessary changes, the following is posited:

The electricity market shifted towards using 100% nonfossil energy technologies. (P5)

Transitioning to NZCSCs necessitates significant investments (Monasterolo et al., 2022); accordingly, investors and financial markets are increasingly channeling funds into sustainable models (van Nieuwenhuijzen, 2023) because they view decarbonization as a crucial performance indicator (Neri et al., 2021). Initiatives such as the Glasgow Financial Alliance for Net Zero highlight a commitment to the 1.5°C goal but have yet to represent mainstream investment practices. Anticipatorily, the following is proposed:

Investment institutions exclusively invest in companies aligned with the 1.5°C Paris reduction path toward carbon neutrality. (P6)

From an investment standpoint, GHG emissions are significant (Griffin et al., 2011), as poor carbon performance introduces financially relevant transition risks. Consequently, financial institutions are screening decarbonization strategies and capabilities before investing (Boffo and Patalano, 2020). An emerging strategy to mitigate carbon risk involves adjusting interest rates, thereby influencing the cost of debt based on a company's net-zero alignment (Caragnano et al., 2020). The following is proposed:

Financial institutions charge higher interest rates the more a company is misaligned with the 1.5°C Paris reduction path toward carbon neutrality. (P7)

Suppliers are more likely to invest in decarbonization if they can transfer some or all of the associated costs to their customers. In sectors such as the automotive industry, original equipment manufacturer, are beginning to support the decarbonization of essential materials including steel in an effort to reduce their SC emissions (Muslemani et al., 2021). Although low-carbon alternatives often incur a price premium (Mccollum et al., 2018), the acceptance of higher costs for reduced carbon footprints continues to develop. Looking ahead, the following is proposed:

B2B customers are willing to pay a significantly higher price for reduced product carbon footprints of products. (P8)

2.3 Niche innovations regarding novel zero-carbon technologies (P9–P12)

In the current landscape, various innovations and key technologies are surfacing as pivotal elements for radical decarbonization; this represents the technological innovation facet of the MLP framework (Geels, 2006). These niche innovations are crucial for system transformation, although they typically feature lower initial performance (Geels, 2006). Four prospective theses (P9–P12) at the niche level that were identified as crucial through the comprehensive literature review and expert interviews were explored.

In alignment with SDG 9, which focuses on building resilient infrastructure, promoting inclusive and sustainable industrialization and fostering innovation, emission transparency across SCs is essential for effective management of life-cycle emissions (Stenzel and Waichman, 2023). Digital ecosystems, such as Catena-X in the automotive industry, have been developed to facilitate carbon emission data transfer (Catena-X, 2024), although widespread adoption by the EU’s manufacturing industry remains limited. By leveraging these digital infrastructures, companies can monitor and control suppliers’ carbon emissions Europe-wide, ensuring transparency and accountability across the entire value chain. The advancement of these digital networks is pivotal for driving innovation and supporting sustainable industrial practices, thereby contributing to the overall goal of sustainable development (United Nations, 2020). Anticipating future development, the following is proposed:

Companies monitor and control suppliers' carbon emissions Europe-wide for the entire value chain by utilizing transparent data-sharing networks. (P9)

Innovative zero-carbon technologies, including nuclear fusion, are emerging as highly desirable solutions for dramatically decarbonizing the electricity mix and ensuring energy stability (Khan et al., 2019). Despite significant investments in nuclear fusion research, those in the development process are confronting fundamental challenges and progress is advancing slowly (U.S. Government Accountability Office, 2023). However, novel disruptive energy solutions connect with the SDGs 7 and 9, targeting affordable low-carbon energy through innovative technologies (United Nations, 2015). In anticipation of technological breakthroughs, the following is proposed:

Novel disruptive energy sources are exploited and gain mainstream adoption. (P10)

Carbon capture and storage (CCS) technologies are a key niche innovation in the decarbonization space; they enable carbon emissions to be separated directly from the air or during the manufacturing process and stored permanently (Anwar et al., 2018; Hanssen et al., 2020). While existing CCS technologies are operational, their high energy demand renders them inefficient and difficult to scale (Reiner, 2016). However, CCS technologies target SDG 9, aiming to enable an innovative net-zero carbon infrastructure (United Nations, 2015) helping to develop carbon-neutral supply chains. Looking ahead to future advancements, the following is proposed:

Carbon capture and storage technologies have matured and scaled, allowing companies to neutralize their residual emissions. (P11)

Logistics service providers seek carbon-neutral technologies, including battery electric and hydrogen vehicles, to reduce their carbon footprint (McKinnon, 2022; Transport and Environment, 2021). The successful adoption of these technologies hinges on substantial investments that support infrastructure, which is a critical factor that remains uncertain in facilitating this transformation (McKinnon, 2022). This is connected to SDG 9 targeting sustainable infrastructure development within the industry (United Nations, 2015). Thus, the following is proposed:

Logistics service providers use fossil-free transport modes and warehousing technologies. (P12)

3. Methodology

Implementing changes at the systemic level entails the engagement of diverse stakeholders (Geels, 2002), which requires arduous prognostication and anticipation of future developments. This study addresses these uncertainties by considering the perspectives of various stakeholders and using long-term foresight to forecast NZCSC developments within the European manufacturing industry.

To explore the research question, this research conducted a Delphi-based scenario analysis. This proactive and interdisciplinary approach involves systematically crafting an interpretative consensus within a specialized group (Prasad and Prasad, 2002). Delphi studies exhibit distinctive features, including anonymity, multiple iterations, statistical group response and controlled feedback (Dalkey and Helmer, 1963; Rowe and Wright, 2011). These studies gather expert opinions within a specific area by soliciting assessments regarding prospective theses (Linstone and Turoff, 2011). The iterative nature of the Delphi method, which involves multiple survey rounds, allows experts to adjust their evaluations through controlled feedback (Rowe and Wright, 1999): individual assessments from previous rounds are presented in aggregate in subsequent rounds, thus fostering effective communication, discussion and mutual learning. Consequently, this recurring process contributes to a higher level of rationality and confidence among participants (Rowe and Wright, 1999). The Delphi-based scenario development process adheres to a four-step approach: Delphi prospective thesis development and formulation, Delphi panel selection, Delphi survey execution and scenario development (Figure 1).

3.1 Delphi prospective thesis development and formulation

The initial phase in envisioning potential NZCSC futures involves the development and formulation of prospective theses. Crafting impactful and inspiring prospective theses in the Delphi study (Markmann et al., 2021) is essential; therefore, a methodical, multistep approach to guarantee data triangulation, validity and reliability was used (Markmann et al., 2021; Nowack et al., 2011). In the first stage, this study conducted desk research to acquire insights into NZCSCs by reviewing academic articles, trend reports and consumer behavior surveys from academic databases, news portals and press releases. Based on the initial research findings, 27 semistructured interviews were conducted (13 interviews with practitioners, 7 interviews each with academics and politics). The participants were provided with a questionnaire in advance to ensure consistent responses, and were recorded, transcribed and analyzed the interviews alongside the literature review. The synthesis resulted in 20 preliminary prospective theses for the future of NZCSCs, as findings from both the literature review and the interviews were combined. During two consecutive internal evaluation workshops, the prospective theses were thoroughly discussed and revised to mitigate ambiguity. Subsequently, similar prospective theses amalgamated and further were streamlined (Roßmann et al., 2018; Rowe and Wright, 2011).

In the subsequent phase, this study implemented an anonymous online expert workshop with five practitioners, four academics and four politics). During the 90-min session, a collective introduction of the research topic was provided and the experts were asked to anonymously rate the prospective theses regarding meaningfulness and probability of occurrence on a five-point Likert scale, where 1 = the least sensible or probable option for each prospective thesis and 5 = the most sensible or probable option for each prospective thesis; additionally, they were asked to offer justification for their rating of each thesis. Following the rating process, experts accessed their groups’ average results via, for example, boxplots and then engaged in a moderated 5-min discussion regarding the outcomes of the assessments. This process leads to more accurate results, especially since negative psychological group biases such as halo or bandwagon effects are minimized (Beiderbeck et al., 2021). Following an analysis of the workshop results, a third internal workshop to further refine and specify the prospective theses were convened. To synthesize the outcomes of the entire development process, a fourth internal formulation workshop was conducted, which entailed discussions, revisions and condensation of the prospective theses, ultimately resulting in 12 diverse, thought-provoking and comprehensible prospective theses. To mitigate fatigue and high dropout rates, the number of prospective theses were held as low as possible while continuing to offer a wide array of prospective developments (von der Gracht and Darkow, 2010). This research adhered to recommendations from relevant literature while formulating the prospective theses (Markmann et al., 2021; Rowe and Wright, 2011), and content validity and comprehensibility of the prospective theses as well as plausibility and consistency of the questionnaire were ensured by independently conducting a pretest with two researchers and two practitioners.

3.2 Delphi panel selection

Ensuring the validity and reliability of a Delphi study involves a careful selection of experts (Landeta, 2006; Spickermann et al., 2014). NZCSCs are interdisciplinary by nature, and disparate expert panels impact accuracy, scientific robustness and the mitigation of cognitive biases; therefore, significant effort to invite a diverse group of experts was expended (Ecken et al., 2011). Additionally, this study ensured that the appropriate knowledge and a representative panel size were considered (Rowe and Wright, 2011). Moreover, to account for the theoretical proposition that socio-technological transitions involve various actors, a multistage selection procedure was used (Roubelat, 2000) and diverse methods for expert identification (Mauksch et al., 2020).

The expert identification process involved three key stakeholders: practitioners, academics and policymakers. In this study specific criteria for each group were established, identifying practitioners by brainstorming to determine the industries to target, which included automotive, logistics, electrical equipment and aluminum; then a desk review to identify relevant companies was conducted while considering individuals with management responsibilities in strategic departments. For academics, recent articles and publications on SC sustainability, decarbonization and corporate sustainability were analyzed. In the policy category, searches on policy, advocacy, associations and city sites were conducted; this study focused on SC decarbonization, sustainable development and energy transformation. In the assessment of identified experts, both surface-level diversity factors (field of work) and deeper-level diversity characteristics (experience) were incorporated (Spickermann et al., 2014). Therefore, a self-assessment of expertise that used a seven-point Likert scale ranging from 1 = very low expertise to 7 = very high expertise was included (Landeta, 2006).

Through the identification process, 550 potential experts with diverse academic, professional and regional backgrounds were identified in an effort to minimize bias and enhance information processing. Invitations were sent to participate via individualized emails, which resulted in 110 experts who were willing to participate in the Delphi survey. Of these, 67 experts completed the entire questionnaire, thus yielding a response rate of 12.12%.

3.3 Delphi study execution

To capture the experts' long-term judgments, this study chose a real-time Delphi (RTD) format using the RTD tool durvey.org, which streamlines the overall process while achieving comparable results to a classic Delphi format (Gnatzy et al., 2011). In an RTD setup, the survey is executed online and the group opinion is immediately calculated and returned. The questionnaire comprised three parts: The initial segment introduced the topic and provided the data protection declaration as well as instructions for the response process. The second part of the questionnaire sought surface-level information (Spickermann et al., 2014) and prompted experts to disclose personality traits (Loye, 1980), as such deeper-level information is increasingly used in business research to yield new insights and correlations between participants and their response behavior (Spickermann et al., 2014). The participants were given a seven-point Likert scale to perform a self-assessment of their expertise in decarbonization as well as supply chain management (SCM). To gauge individual attitudes, two matrix questions were incorporated to specifically assess their level of openness regarding decarbonization.

The third segment of the survey featured questions on the 12 prospective theses regarding the future of NZCSCs; each question was accompanied by brief statements regarding its present level of adoption. To evaluate each prospective thesis, experts rated the expected year of occurrence (T), impact (I) and desirability (D). This study also encouraged participants to provide voluntary written justifications to support their personal assessments; the goal was to acquire both quantitative and qualitative data (Tapio et al., 2011).

Similar to other studies (e.g. Bathke et al., 2022; Förster, 2015), 2023−2050 as potential years of occurrence was used; experts selected the exact year (e.g. 2025, 2026, 2027, etc.) or 2050+, which was considered as nonoccurrence. Impact and desirability were rated on a seven-point Likert scale ranging from 1 = very low to 7 = very high.

3.4 Scenario development

The scenario development process with a quantitative analysis of prospective theses was performed, measuring three criteria (T, I, D) in two rounds with each expert, which resulted in 4,824 numeric values. In the initial quantitative analysis, means, standard deviations and interquartile ranges were calculated. Subsequently, clustering using the rating scores were performed, following the approach of Tapio et al. (2011). For qualitative data interpretation, the fuzzy c-means (FCM) algorithm was applied (Ketchen and Shook, 1996) to group the theses based on the mean values of expected year of occurrence (T), impact (I) and desirability (D) (Roßmann et al., 2018). FCM clustering is consistent with the methods used in recent Delphi studies by Roßmann et al. (2018) and Bathke et al. (2022). Unlike k-means clustering, FCM assigns each thesis a membership value between 0 and 1 for all clusters, avoiding fixed single cluster centers (Bezdek, 1981; Bezdek et al., 1999). The iterative nature of the FCM clustering allows for a clear assignment of membership grades for prospective theses situated between two clusters (Bezdek et al., 1999). This study determined that three clusters were appropriate based on the consideration that the number should be lower than the square root of the analyzed prospective theses but higher than one (Bezdek et al., 1999).

4. Analysis

4.1 Descriptive results

This research first analyzed the quantitative results of the Delphi survey; Appendix 1 enumerates the 12 prospective theses regarding the future of NZCSCs. Furthermore, the experts' aggregated quantitative evaluations were incorporated, encompassing the estimated year of occurrence (T), the impact of occurrence (I) and the desirability (D). The findings reveal that the anticipated impacts associated with each of the 12 prospective theses were nontrivial, as evidenced by the mean impact for all prospective theses, which surpassed the 3.5 scale mean.

Consequently, all prospective theses were considered in the subsequent analysis of the Delphi survey results. P5 (Nonfossil electricity) exerted, as anticipated, the most substantial impact (should it materialize), boasting an average impact rating of 6.06. Next were P2 (Carbon pricing) with a rating of 5.76 and P3 (Carbon neutrality incentives) with a rating of 5.18. Conversely, P12 (Zero-carbon logistics) was expected to yield the least impact, garnering an average rating of 4.42. The estimated commencement year for the prospective theses ranged as follows: from 2029.42 for P3 (Carbon-neutrality incentives) to 2040.75 for P10 (New zero-carbon energy). Notably, experts expressed skepticism regarding P10 (New zero-carbon energy), voting for its nonoccurrence on 14 occasions.

Perceptions of socio-technological transitions are often contingent upon stakeholder group membership and associated interests (Geels, 2002); hence, three group comparisons were conducted after the overall panel analysis. Initially, expectations articulated by experts from science (n = 22), industry (n = 39) and politics or associations (n = 6) were juxtaposed. This research used a Mann−Whitney U-test to discern significant differences among the three stakeholder groups, which revealed eight assessments that varied significantly (p < 0.05). Subsequently, this research evaluated these for within-group consensus. A significant disparity in anticipated impact for P11 between business and science experts (p = 0.044) as well as between business and politics experts (p = 0.049) was noted. Practitioners consider this projection to be less influential compared to academics and politicians, justifying their assessment, among other factors, by the high energy requirements relative to carbon reduction and the currently still sobering findings from pilot projects. Moreover, this study ascertained dissimilar assessments between science and politics experts for P8 (p= 0.034). The political perspective here particularly reflects a lack of willingness to pay, whereas academics see no alternative means to finance carbon neutrality. Regarding projected entry years, four prospective theses exhibited discrepancies among stakeholder groups. Specifically, this research discovered significant differences between assessments from business and politics experts as well as between science and politics experts for P10 (p = 0.029). Academics and practitioners, in particular, view this projection as unfeasible in the near future. They justify this stance by citing the required systemic changes, the lack of financing and the technological maturation needed. A significant disparity between the opinions of business and science experts for P4 (p = 0.037) and P5 (p = 0.001) was also noted. For desirability, this study discerned a notable difference between business and science experts for P12 (p = 0.026) only.

A second group comparison for sentiment analysis was conducted, examining evaluations from experts versed in SCM and decarbonization. Thus, this study grouped participants based on their self-rated expertise and performed a mean value comparison, which revealed significant differences for two prospective theses: P7 displayed variations in the assessment of expected entry years among experts with differing levels of SCM knowledge, while P3 exhibited significant differences in desirability assessments among experts with differing levels of decarbonization knowledge.

To explore the potential influences of participants’ personality traits, experts were categorized as green-minded or green-emergent based on their sustainability orientation. This study again used a Mann−Whitney U-test to identify possible differences between the two groups. For P8, two statistically significant differences (p < 0.05) were identified for desirability and impact dimensions between green-minded and green-emergent experts. Additionally, two statistically significant differences (p < 0.05) for P7 concerning desirability and estimated year between green-minded and green-emergent experts were noted. This study detected that other significant differences were absent between the two groups and in the diversity of expectations and responses, which indicates that the test results affirmed the credibility and validity of the findings, as the anticipated future aligns across various dimensions and topics from those with disparate perspectives.

4.2 Cluster development

This research identified three clusters based on the evaluated prospective theses. Therefore, this study determined that the assessed impact and desirability were the key determinants. The clusters are illustrated in Figure 2, displaying all prospective theses within a 3D plot that includes their impact, desirability and year of occurrence.

Cluster 1, termed Complimentary drivers, contains six prospective theses: P7 (Carbon-adjusted interest rates), P1 (PCF transparency), P11 (CCS), P8 (B2B customers pay for carbon neutrality), P12 (Zero-carbon logistics) and P10 (New zero-carbon energy), which were assessed with the lowest impact and desirability compared to other prospective theses. This scenario contains a conglomeration of prospective theses, all of which positively impact the development of NZCSCs, which, on their merits, have a market-driven impact.

Cluster 2, termed Catalyzing forces, is formed by four prospective theses: P3 (Carbon-neutral incentives), P4 (CE regulation), P9 (CO₂ data sharing) and P6 (Carbon-neutral investments). This scenario includes prospective theses with a substantial impact and desirability regarding the development of NZCSCs. However, these propositions have moderate impact and desirability when compared to other clusters.

Cluster 3, termed Game changers, includes two prospective theses: P2 (Carbon pricing) and P5 (Nonfossil electricity). These prospective theses received the highest impact and desirability compared to all other prospective theses. For developing NZCSCs, this cluster offers the greatest impact, which would change the entire sector over the long term. The high desirability of the prospective theses suggests that this cluster also represents an urgent need to act.

5. Discussion

Reflecting on the study’s results, which entailed clustering the 12 prospective theses and using the MLP as the theoretical framework, the discussion was structured in three subsections. Therefore, this research delves into each cluster to examine the implications for practitioners, policymakers and academics (Sections 5.1–5.3), which are summarized in Table 1. In the analysis, this study leverages the rich qualitative feedback from experts in the Delphi study and examine these findings in the context of existing literature.

5.1 Implications for Cluster 1: complimentary drivers

Practitioners should consider the benefits of PCF transparency regulations (P1) to enhance companies' decision-making and foster competition in the decarbonized products market offers (He et al., 2019). Collaboration within the Tier-N SC and participation in initiatives to standardize carbon calculation methodologies are essential for effective PCF development. Scholars have recently addressed carbon methodology harmonization issues for corporate carbon footprints (Jensen, 2012; Klaaßen and Stoll, 2021), and the experts have recommended that industries, can strengthen their market position by accepting paying a premium (P8), which could enable increased investments in decarbonization. The experts have also advocated for investments in impactful CCS (P11), especially in hard-to-decarbonize sectors such as steel (e.g. Acquaye et al., 2014; Lei et al., 2023) or textile industry (Shen et al., 2017). Companies should adopt science-based standards that allow only 10% of residual emissions to be offset or removed (SBTi, 2021) while awaiting emerging low-carbon technologies. Decarbonizing the logistics sector (P12), particularly heavy-duty fleets and shipping, is crucial for reducing the EU's carbon footprint (Bonilla et al., 2015; Lu et al., 2023). Strategic investments in infrastructure and demand for decarbonized fuels are recommended based on recent studies on biofuels and renewable fuels (Panoutsou et al., 2021; Stolz et al., 2022). Connecting interest rates to a carbon budget is suggested as an effective financial market strategy (P7), but its impact varies based on a company's credit needs. Introducing new sustainable energy sources (P10) in manufacturing, the third-largest energy consumer in the EU (Eurostat, 2023), can significantly impact decarbonization, despite technological maturity challenges (Lerede et al., 2023). The experts recommend upscaling existing decarbonization measures, including the use of renewable energy, during this transition.

From a policy perspective, experts argue that the PCF transparency regulation (P1) must have accompanying incentives or penalties, such as carbon pricing, to have a meaningful impact (Sun et al., 2022). For CCS (P11), the experts recommend government investment in research and development for cost competitiveness and efficiency. Therefore, policymakers should carefully consider the development pathways for carbon capture technologies in policy implementation (M. Shen et al., 2022). Policymakers should also prioritize carbon-neutral logistics (P12) by providing incentives for fleet decarbonization, using public funding for charging infrastructure and exploring alternative fuels (e.g. renewable fuels; Stolz et al., 2022). In steering the financial market (P7), policymakers should focus on an internationally aligned, carbon-adjusted interest rate policy and carbon border adjustment mechanisms (Böhringer et al., 2022). Public funding should support new zero-carbon energy sources (P10) despite their current lack of cost-competitiveness, as they are a crucial element of the long-term energy transition (Lerede et al., 2023).

Researchers should examine the impact of PCF declarations (P1) on market decisions and their connection to corporate carbon accounting and disclosures. Additionally, further investigation into CCS (P11) is warranted, focusing on energy efficiency advancements and collaborative models across industries that build on recent academic progress in the EU's strategies for CCS (e.g. Turgut et al., 2021). Existing research addresses energy storage systems for sector-coupled energy systems decarbonization (e.g. Victoria et al., 2019); however, the experts recommend exploring decentralized energy storage for heavy-duty vehicle charging infrastructure and cross-sector energy coupling to advance carbon-neutral logistics (P12). Academic exploration into carbon emissions associated with debt financing (e.g. Palea and Drogo, 2020; Van Den Bremer et al., 2018) should also extend to include analysis of the impact of carbon-adjusted interest rates (P7) on companies of various sizes, business models and financial situations. Finally, future research should delve into the implementation of new sustainable energy sources (P10), including carbon fusion, while considering potential environmental and societal side effects.

5.2 Implications for Cluster 2: catalyzing forces

For practitioners, the experts suggest enhanced transparency in sharing carbon emissions data across the entire value chain to identify emission hotspots and implement collaborative decarbonization measures (P9). The experts also encourage active participation in standardizing carbon accounting technologies and forming industry alliances, such as Catena-X (Catena-X, 2024), for SC decarbonization. The experts highlighted that financial institutions (P6) should align internationally in the capital market and establish collective portfolio decarbonization targets. Therefore, practitioners should consider tailoring rules for emerging markets that lack financial resources, which has been addressed by researchers within limited areas (e.g. Ng et al., 2023). Moreover, the experts stated that internationally harmonized methodologies for carbon emission accounting support financial institutions in decision-making. Concerning carbon-neutral government incentives (P3), companies should transparently engage with national governments, showcasing the impact of public funding on industrial decarbonization. The experts further urge active involvement in CE development (P4) to conserve resources (Ghisellini et al., 2016) and reduce dependencies (European Commission, 2020); they also emphasize CE strategies beyond recycling and investing in more efficient recycling technologies. This may be fostered by close interfirm research and development cooperation (Liu and Song, 2017).

From a policy perspective, incentivizing carbon emissions data sharing (P9) throughout the SC is crucial for effective decarbonization. Therefore, establishing data-sharing networks, minimizing bureaucratic burdens for small and medium-sized enterprises (SMEs) and preventing data manipulation are key recommendations for policymakers. When implementing carbon-neutral incentives (P3), such as subsidies, policymakers should use a mix of financial methods, including carbon pricing, to ensure lasting effects. Additionally, as indicated by authors of recent studies (e.g. Peñasco et al., 2020), avoiding negative market effects in financial support allocation is essential. Concerning the development of comprehensive CE regulations (P4), policymakers should recognize that recycling quotas may not be suitable for all materials or products; additionally, they should investigate other CE strategies (e.g. 9-R CE strategies [Potting et al., 2017]). CE regulations alone are insufficient; a combination of incentives, market education, best practices and networks is vital for both CE and NZCSC development.

For academics, the experts stated that researchers should focus on maintaining emissions data quality in SC sharing, transitioning from generic carbon emission databases, such as ecoinvent (Perdersen et al., 2013), to primary carbon emission data sources (P9) and integrating digital ecosystems (e.g. Catena-X) with minimal bureaucracy. Thus, researchers should investigate issues regarding the alignment of international financial markets (P6) toward common decarbonization targets and strategies, especially concerning climate pledges (e.g. Invesco, Vanguard and BlackRock [GFANZ, 2024]). Considering government incentives for carbon neutrality (P3), future researchers should design equitable allocation methods for maximum SC decarbonization effectiveness. Regarding CE regulations (P4), researchers have investigated diverse aspects, including the EU's institutional framework (Halkos and Panagiotis-Stavros, 2024) and barriers (Kirchherr et al., 2018). However, academic discussions on different CE strategy regulations for materials and use cases as well as potential technological advances in material recycling would support the development of NZCSCs in the European manufacturing industry.

5.3 Implications for Cluster 3: game changers

Practitioners benefit from aligning with the European climate budget through carbon prices (P2) that are connected to a carbon budget, as seen in the existing European ETS system, even at low carbon prices (Bayer and El Aklin, 2020). However, the experts stated that smaller companies face profitability challenges that warrant advocacy to policymakers. The experts stressed the prioritization of business investments in mass development and deployment of renewable energy (P5), including on-site generation. To achieve carbon neutrality, practitioners must also electrify processes such as steelmaking (Muslemani et al., 2021) that are crucial for decarbonizing energy in industrial processes. Additionally, the experts recommend focusing on renewable energy development and production research (e.g. efficiency improvements [Drechsler et al., 2017]).

Politicians must be careful that a carbon budget connected to carbon pricing does not become a measure of protectionism, as researchers emphasize that protectionism hinders the transition to carbon neutrality (Wang et al., 2023). Thus, to avoid penalizing domestic companies, the pricing should be internally aligned with other economies. Social compensation schemes in the EU, which are backed by research on positive public acceptance (Belfiori, 2017; Klenert et al., 2018), should also be investigated by future researchers. Policymakers should prioritize the broad introduction of nonfossil energy markets (P5), including renewable energy, within the EU (Zappa et al., 2019). The experts recommend incentivizing the deployment of renewables while supporting the development of energy storage solutions. Furthermore, potential drawbacks, such as waste from nuclear power plants including carbon emissions (Liu et al., 2023), should be carefully considered when promoting the development of nonfossil energy markets in the EU.

From an academic perspective, the present research primarily focuses on differentiated carbon pricing and economic costs (e.g. Landis et al., 2018). Future researchers should shift to exploring the implementation mechanisms of carbon budget pricing (P2) and the international alignment of pricing policies (Böhringer et al., 2022), which includes an assessment of the impact on emerging economies and industries that are difficult to decarbonize. EU-specific social compensation systems may also be explored in this context. While scholars have addressed the achievement of a fully renewable energy grid in the EU (Zappa et al., 2019), academicians should explore the emissions and environmental impacts of alternative nonfossil energy solutions (P5), such as nuclear power, as well as reducing energy consumption in industrial processes through synergies between industries. Finally, the study of the balance between the EU industrial sector’s energy needs and local renewable energy production, including energy storage and grid integration issues, is vital in future explorations.

6. Conclusion

Rising global temperatures require urgent commitments to carbon neutrality from EU regulators and companies, particularly in the manufacturing sector, which is responsible for 20% of the EU's carbon emissions. As the largest contributor to any company's carbon footprint, SC emissions require careful management for a successful transition to carbon neutrality. Using the Delphi methodology with the MLP framework, the research was guided by the following question: How can NZCSCs develop within the socio-technological landscape of European manufacturing until 2050?

By conducting a series of interviews and workshops, 12 prospective theses were developed, which were carefully formulated to navigate the European manufacturing industry's path toward NZCSCs. Within an RTD survey using the tool from durvey.org, 67 experts in SCM and sustainability assessed the prospective theses, exploring impact, desirability and estimated year of occurrence.

The assessment of the 12 prospective theses, which were categorized into three clusters −complementary drivers, catalyzing forces and game changers − demonstrates a lack of consistency in the impact, desirability and year of occurrence among the prospective theses. Navigating through this era of decarbonization dissonance and dynamism, 68 implications were developed considering each prospective thesis that was further tailored to practitioners, policymakers and academics. Recognizing the need for active engagement and management during this period, these implications are crucial considerations for those striving to meet the pressing decarbonization targets of climate neutrality in European SCs by 2050.

6.1 Implications for supply chain theory and practitioners

This research has implications for SC theory, particularly in applying the MLP framework to carbon-neutral supply chains. By incorporating enthalpy theory, these frameworks can be adapted to decarbonize SCM from a market stakeholder perspective. This transition involves moving an entire industry toward carbon neutrality, where market stakeholders often play a leading role in driving change. Additionally, the integration of frameworks such as the SDGs can be explored further, connecting transition theory with societal landscape drivers.

Practitioners may find value in this study by recognizing the main drivers of the upcoming industrial transformation toward net zero carbon emissions. In addition to the findings presented in Table 1, practitioners can develop strategies related to all 12 prospective theses to gain a competitive advantage in the market. This enables them to be at the forefront of developing carbon-neutral supply chains within the European Union.

6.2 Limitations and future research

This article, like others, is not without limitations. First, the participants in the study were predominantly practitioners. Although many prospective theses are driven by industry, regulatory prospective theses P1–P4 could be better assessed by a large number of policymakers. Hence, greater involvement of policymakers in the future would increase the accuracy of the assessments. Additionally, this research was focused on 12 prospective theses that emerged from the first phase of the study. However, there may be more influences that could play a role in the development of NZCSCs, which would further enrich the discussion in this area. Third, future researchers could focus solely on the landscape or technological innovations and delve deeper into one of the two influential levels within the MLP framework.

Figures

Figure 1

Applied Delphi process

Figure 2

Clustering of all prospective theses

Table 1

Framework implications

Implications for practitioners	Implications for policymakers	Implications for academics
P1: PCF transparency
Collaborate with Tier-N suppliers for data Engage international carbon emission calculation standardization initiatives	Combine PCF transparency regulations with other market steering measures Internationally align carbon emission calculation methodologies	Investigate the effects of PCF declarations on market choices Understand market conditions during the transition to carbon neutrality
P2: Carbon pricing
Engage in internationally harmonized carbon accounting methodologies Advocate for policies that support profitability for small enterprises with limited capital	Ensure the carbon budget attached to carbon pricing is not a measure of protectionism Align pricing measures internationally and consider effects on emerging economies Introduce carbon pricing policies with considerations for carbon-reduced alternatives	Research carbon budget pricing implementation mechanisms for the manufacturing industry Analyze international markets' alignment with pricing measures, including effects on emerging economies Investigate social compensatory systems within the EU
P3: Carbon-neutrality incentives
Evaluate the impact of public funding on research and development for industrial decarbonization Address short-term uncertainties with incentives to foster decarbonization. technology commercialization Focus on long-term decarbonization opportunities through developed technologies	Combine carbon-neutral incentives (e.g. subsidies) with financial steering methods Ensure fair and accurate allocation of financial support Prevent negative market influences when implementing carbon neutrality incentives	Design allocation methods for fair distribution to ensure maximum SC decarbonization Investigate criteria for short-term financial incentives that ensure long-term effectiveness
P4: CE regulation
Support the development of a CE for resource conservation Invest in efficient recycling technologies to reduce PCFs Explore new economic business opportunities within CE practices (9-R strategies)	Recognize that recycling quotas may not be universally applicable Combine CE regulations with incentives, market education, and best practices Emphasize the importance of networks for effective CE impact	Discuss regulations for different materials and use cases in CE strategies Explore potential technological advancements in material recycling
P5: Nonfossil electricity
Expand the purchase of electricity based on renewable energies Increase electrification of diverse processes in industries to reduce Scope 1 emissions Promote on-site energy production for sustainability Research and advance renewable energy production to enhance efficiencies	Focus on widespread adoption of nonfossil energy markets and emphasize renewable energies Incentivize adoption while supporting energy storage solutions Consider potential drawbacks, including waste from nuclear power plants	Investigate emissions and environmental impact of nonfossil-based energy solutions, such as nuclear power Explore ways to further reduce energy demand in industrial processes through industrial-energy synergies Research the alignment of EU industries' energy demand with local, renewable energy production, including energy storage and grid implementation issues
P6: Carbon-neutral investments
Strive for international alignment in capital markets for collective decarbonization goals Consider differentiated rules for emerging markets that lack financial resources Engage in internationally harmonized carbon accounting methodologies	Engage on common decarbonization goals with regulatory bodies within the international financial markets Address financial inequality in emerging markets by adopting climate rules and strategies	Investigate international financial markets' alignment on common decarbonization goals and strategies Consider the financial disparities in emerging markets and their adoption of climate rules
P7: Carbon-adjusted interest rates
Engage in initiatives that support international carbon emission calculation standardization Develop systems that account for a firm’s size and financial situation	Strive for internationally aligned carbon-adjusted interest rate policies Consider effects on border adjustment mechanisms Consider effects on hard-to-abate sectors	Analyze the impact of carbon-adjusted interest rates on companies based on size, business model, and financial situation
P8: B2B customers pay for carbon neutrality
Maximize customer adoption with willingness to pay a premium for decarbonized products that transform the business model	Establish a combination of market steering methods that benefit low-carbon products	Research market steering methods that support companies that offer low-carbon products Investigate how the international market can adjust through carbon border adjustment mechanisms
P9: CO2 data sharing
Share carbon emission data along the entire value chain for collective insights Engage with Tier-N SC actors for increased carbon emission data Provide training on decarbonization and digitalization for efficient carbon emission data-exchange platforms	Incentivize emission data sharing through policies Promote international engagement on common carbon accounting methodologies Minimize bureaucratic efforts for small and medium-sized companies in emission data accounting while offering public education	Investigate emission data quality standards concerning primary carbon emission data Explore the integration of digital data ecosystems with high data quality and minimal bureaucratic effort Examine data privacy considerations in digital emission-sharing networks
P10: New zero-carbon energy
Invest in the development of new sustainable energy sources Increase the adoption of renewable energy	Support strategic long-term research projects for new sustainable energy sources	Investigate the applicability and scalability of diverse new sustainable energy sources, including nuclear fusion Assess potential side effects of nuclear fusion on the environment and society
P11: CCS
Develop and commercialize CCS technology, especially for hard-to-abate sectors Focus on simultaneous implementation of actual reduction measures	Publicly support CCS efficiency development through investments and subsidies	Research energy efficiency advancements, considering industry-wide collaboration models
P12: Zero-carbon logistics
Strategically invest in decarbonizing heavy-duty fleets and shipping infrastructure Demand and invest in decarbonized fuels to generate market demand	Incentivize fleet decarbonization and decarbonized fuel development Fund decentralized charging infrastructure, including local decarbonized energy supplies	Investigate decentralized energy storage applications for decarbonizing the charging infrastructure Explore intersectoral energy coupling to advance the charging infrastructure for heavy-duty vehicles

Source: Authors’ own work

Table A1

Quantitative Delphi results

		I-statistics				D-statistics				T-statistics
Nr.	Prospective thesis	Mean	IQR	SD	CV	Mean	IQR	SD	CV	Mean	SD	CV*	Never
1	PCF transparency	4.7	2	1.24	−8.05	4.67	2	1.47	−6.30	2033.12	4.41	−16.37	1
2	Carbon pricing	5.76	2	1.24	−5.59	5.60	1	1.22	−3.00	2031.31	5.24	−23.14	3
3	Carbon-neutrality incentives	5.18	2	1.13	−10.31	5.33	1	1.26	−7.20	2029.42	4.32	−16.52	1
4	CE regulation	4.9	2	1.14	−8.92	5.22	1	1.15	−8.25	2032.88	4.50	−23.73	1
5	Nonfossil electricity	6.06	1	0.99	−12.28	6.18	1	1.00	−4.27	2038.71	4.59	−26.67	4
6	Carbon-neutral investments	4.88	2	1.26	−13.36	5.10	1	1.43	0.16	2033.94	5.65	−12.36	4
7	Carbon-adjusted interest rates	4.87	1.5	1.04	−19.61	4.79	1.5	1.24	−13.31	2031.50	5.46	−13.00	3
8	B2B pays for carbon-neutrality	4.57	2	1.51	−11.08	4.91	2	1.39	−6.12	2032.51	5.50	−21.96	2
9	CO2 data sharing	4.9	2	1.22	1.00	5.15	1	1.13	−6.26	2031.37	4.26	−13.14	2
10	New zero-carbon energy	4.6	2	1.64	0.98	4.94	2	1.54	−12.83	2040.75	4.90	−14.87	14
11	CCS	4.45	2	1.56	−3.59	4.82	2	1.49	−1.02	2036.47	4.96	−17.30	5
12	Zero-carbon logistics	4.42	1	1.16	−17.81	4.90	2	1.37	−2.77	2037.89	5.49	−17.79	1

Notes:

Prospective theses with consensus among panelists are marked in italic; I = impact in case of occurrence (seven-point Likert-scale; 1 = very low; 7 = very high); D = desirability of occurrence (seven-point Likert-scale; 1 = very low; 7 = very high); T = estimated time of occurrence (2023–2050); IQR = interquartile range; *experts rating T as “never” in one of the two rounds were excluded.

Source: Authors’ own work