1The core methodological approach of this paper is the use of qualitative methods, with limited use of quantitative methods. The methodological challenge is (1) to reliably and objectively identify which economic sectors (and which actors within them) are supportive of hydrogen and fuel cell technology in Europe, and (2) to analyse what their motives are for doing so. Europe is chosen as it is one of the regions that is globally most involved in hydrogen and fuel cell research (alongside China and Japan), and because there is substantial support from a supranational (i.e. EU) level. This creates an interlinkage with questions about the role of global governance in promoting technology shifts. The paper uses the UN Statistical Division (2020) M49 standard to define the geographic scope of Europe.1

2 The paper identifies support for the hydrogen transition by manually gathering membership directory data of all non-government and non-public entities2 registered in international, national or sub-national hydrogen and fuel cell associations, and categorising these by primary business activity and country. These associations aim to promote hydrogen projects, cooperation between interested parties, and can function as lobby groups in support of the hydrogen transition.3 They are used as proxy for supporting the hydrogen transition. In total, 1670 registrations are categorised, from 39 hydrogen associations. Business activity is registered based on the Statistical Classification of Economic Activities in the European Community, commonly known as NACE (Eurostat, 2008). These are standardised in the European Single Market, and consequently cover most registrations. Several NACE levels are registered, from the overarching section level, to division, to group or class.

3 Despite this standardisation, national systems for accessing business registers vary considerably, and many require payment per information request. Where NACE-registration information is freely accessible (e.g. Belgium, Czech Republic, Denmark, France, Italy, Slovakia, UK), official government business registrations were used. Where information on NACE-registrations is not freely accessible (e.g. Austria, Germany, Netherlands, Sweden) through the relevant government agency, information services based on this information were used.4 The former is much preferred for reliability, and subsequent research should aim to be based on the official government channels for business information.

4 There are validity problems with this classification method. In most jurisdictions, companies choose their classification. This means that companies conducting relatively similar or overlapping activities can register as carrying out different activities. This has mostly been left untouched, except for the manual change of Nouryon5 and the inclusion of transporters via pipeline (e.g. GRTgaz and Gasunie) under electricity and gas instead of under transport and storage. Furthermore, companies active in multiple activities (e.g. integrated oil and gas companies, utilities, or conglomerates) are classified as conducting only one business activity. Finally, large companies tend to register as a head office6 (NACE M70) with subsidiaries conducting different business activities, sometimes differing by country. For consistency and clarity, two principles in categorisation are used: (1) same-name subsidiaries are registered in exactly the same way as the company is registered in the country where the headquarters are located, (2) large organisations registered under NACE M70 are changed to the code of a registered subsidiary at the discretion of the author.

5 The above points are acknowledged methodological flaws. However, it is difficult to avoid manual corrections using any type of sector classification based on official registrations, while at the same time maintaining the validity of the results. Air Liquide, Deutsche Bahn, Électricité de France, Hyundai and Shell are all registered as carrying out the same activity, which is not an accurate reflection of their behaviour and business interests as economic actors. The support of subsidiaries may vary across countries, a nuance which is lost by registering them as “clones” of the parent company. However, here too, analytical consistency is lost when subsidiaries from the same company carrying out the same activities but registered differently abroad are considered as being part of different industries. To some extent, these issues are all the result of artificially rigid statistical groupings. These flaws may be inevitable when analysing sectors, rather than doing a company-by-company analysis for all 1670 registrations. Subsequent research could and should try to improve on this.

6 The paper uses this business activity dataset to analyse support across sectors for the hydrogen transition, and to identify who the main actors are in these sectors. This lays the foundation for answering the second part of the research question on the motivation of these actors. This is done with literature research, such as by going through annual reports, financial reports, sustainability reports, press releases, news articles, and existing (academic) literature on the sector’s or actor’s connection to hydrogen. A serious methodological improvement would be to engage representatives from all identified sectors and all key actors, as well as independent experts from governments and research institutions. Acquiring representative samples to make objective conclusions about all sectors and actors – taking account of large differences even within sectors – requires considerable resources and falls outside the scope of this paper. A larger study could attempt doing this or first focus on specific actors and/or sectors.