Frames, interests, and incentives – a typology of institutionalizing RRI in the business sector derived from ten pioneering projects

ABSTRACT

While in scientific and policy discourse on responsible research and innovation (RRI) the desirability and sustainability of innovations are central, the business sector follows different logics revolving around competitiveness, profits, and growth. They manifest themselves in different discursive frames, indicate potentially conflicting interests, and allow specific incentive mechanisms to be utilized. This paper builds on discursive institutionalism, presents a qualitative comparative case study of ten EU-funded pioneering projects, and derives a generalized typology. It contributes to the theoretical debate on institutionalizing RRI by applying discursive institutionalism and through a nuanced understanding of three leverage points for organizational change in the business sector. We conclude that frames, interests, and incentives are interlinked phenomena and thus should be considered comprehensively. As our typology addresses specific contexts at regional, corporate and process level, it can be further utilized to develop tailored policies and strategies capable of institutionalizing RRI in the business sector.

KEYWORDS:

• Responsible innovation
• business sector
• institutionalization
• competitive environment
• EU projects
• discursive institutionalism

Introduction

The business sector is a key player in research and development (R&D): in 2021, about two-thirds of the R&D activities in Europe were performed by companies and 58% of the total R&D expenditure in Europe was funded by companies.¹ Designing R&D processes and outcomes in a responsible manner has to address the business sector in general and individual companies in particular.

Scholars have discussed different aspects of the responsibility of research and innovation since the 1970s (Owen and Pansera 2019). During the last decade, European policy took up these ideas and supported their further development and dissemination under the term Responsible Research and Innovation (RRI) von Schomberg (2013). In this process, the EU Framework Program ‘Horizon 2020’ played an important role and supported several hundreds of RRI-related projects with more than 300 million Euros (see Chapter 3 of this paper). These projects were frontrunners in further developing RRI-related concepts and tools, mainly for public research-funding organizations, e.g. governmental science funds and regional development programs, and semi-public research-performing organizations, e.g. universities and higher education institutions (Owen, von Schomberg, and Macnaghten 2021).

The concepts, methods, and tools developed in these EU projects have not yet been taken up by companies on a large scale (Schönherr, Martinuzzi, and Jarmai 2020), raising the assumption that institutionalizing RRI in the business sector may follow different pathways than in the semi-public sector (Owen et al. 2021). RRI in the semi-public sector prioritizes desirability, acceptability, and sustainability. The business sector, however, is shaped by market forces and competitiveness and therefore has to focus on profits and shareholder interests (Garst et al. 2017). Challenges in implementing RRI in the business sector, therefore, stem from potentially conflicting interests of society and corporations (Martinuzzi et al. 2018). These different priorities manifest themselves in different discursive frames (Randles, Tancoigne, and Joly 2022), which influence and legitimize the decisions and actions of individuals and corporations. To overcome such conflicting interests, policy can make use of a broad range of incentive mechanisms (Steurer 2011) and specific incentives are needed to support RRI in the business sector (Gurzawska, Mäkinen, and Brey 2017). Discursive frames, potentially conflicting interests, and incentive mechanisms are therefore important aspects of institutionalizing RRI in the business sector and have been occasionally discussed by scholars. However, as of now, there is scientific work at hand that would integrate them into a nuanced typology. Our paper addresses the scientific debate on institutionalizing RRI, by (a) drawing attention to the importance of discursive framings to institutionalize RRI in the business sector, and (b) developing a typology that is based on discursive institutional theory, and thus (c) providing the foundations of effective institutionalization strategies for future research and innovation projects.

Our study follows a key concept of new institutional theory – discursive institutionalism – which focuses on the interplay of ideas, discourse and institutions, on competing logics and on how content and interactive processes of discourse lead to change (Campbell and Pedersen 2001; Schmidt 2002; 2008; 2010). While discursive institutionalism originates from political science, it also influenced organization scientists, who developed a discourse-centered understanding of organizational change (Grant and Marshak 2011; Heracleous and Barrett 2001; Marshak 2002). Following discursive institutionalism, change happens through re-framing, adaptation and/or adoption of frames (Werner and Cornelissen 2014). In the RRI-related scholarly debate, discursive framing has been rarely addressed (Randles et al. 2016), and our study contributes to the theoretical debate on institutionalizing RRI (Owen et al. 2021), by placing the emphasis on discursive frames – a key concept of discursive institutionalism.

In discursive institutionalism, interests and incentive structures are perceived as specific types of ideas and discourses, as socially constructed, and – in contrast to rational choice institutionalism – not as neutral or as material (Schmidt 2008). In the scholarly debate on institutionalizing RRI, the conceptual link between discursive frames, interests and incentive mechanisms has not yet been drawn. Potentially conflicting interests and resulting tensions – such as between openness and intellectual property rights – have been highlighted by Brand and Blok (2019), but have not been applied on a broader scale to explain the challenges of institutionalizing RRI. Incentive mechanisms have been analyzed in the general policy field of CSR² (Steurer 2011), but not in the more specific area of RRI. By linking discursive frames, potentially conflicting interests and incentive mechanisms, our study aims to provide a nuanced view on institutionalizing RRI rooted in discursive institutionalism.

To do this, we asked the research question: ‘How did EU funded pioneer projects apply discursive frames, address potentially conflicting interests and utilize incentive mechanisms for institutionalizing RRI in the business sector?’ The EU Framework Program ‘Horizon 2020’ is a particularly valuable source of information and evidence, as during the last years the European Commission set out to institutionalize RRI in the business sector (Delaney and Iagher 2020). The projects funded in these areas should be well acquainted with the concept of RRI, provide real-world experiences, and be able to substantially reflect on them. They can provide insights and experiences from a broad range of companies and act as frontrunners for adapting RRI to the contexts and needs of the business sector. Their results and experiences can be utilized for developing a better understanding of how RRI can be institutionalized in the business sector, which can be the basis for new theoretical considerations.

To address the research question, we analyzed activities, outcomes, and learnings of ten selected pioneer projects on RRI in the business sector and aims to (a) explore how these projects addressed discursive frames, potentially conflicting interests, and incentive mechanisms and (b) derive a typology that helps develop strategies for institutionalizing RRI in the business sector. To gain rich insights, we selected projects that cover a broad range of approaches, sectors, and regions, processed all project deliverables and publications, carried out a series of expert interviews with project coordinators, and carried out a qualitative comparative case analysis (Halkias et al. 2022; Stake 2013) mainly following the approach of Gioia, Corley, and Hamilton (2012) of finding ‘qualitative rigor’.

Our typology provides a nuanced understanding of three relevant aspects for institutionalizing RRI in the business sector. For policy making it can support the design of effective incentive mechanisms and strategies, something that is not captured in the recent RRI related discourse. On a more general level, our paper contributes to the scholarly debate on discursive institutionalism, by providing insights into a field of application with potentially conflicting interests between businesses and society. Application here can also be understood in terms of Czarniawska and Joerges (1996) concept of ‘travels of ideas’. The authors discuss the concept of how ideas are translated into actions and why certain ideas are implemented as organizational measures at specific times. They argue that conventional explanations for organizational change are insufficient and suggest a closer examination of the process of idea materialization, which we do using the example of RRI. This aligns with the idea of discursive institutionalism highlighting the importance of understanding the narratives, discourses and framing strategies that influence policymaking, governance and social practices.

The following background section presents a short summary of the origins of RRI and the current challenges of applying it in the business sector. In the literature review section, we summarize the different strands of research on discursive framing, potentially conflicting interests and incentive mechanisms, what other research has contributed to the debate on institutionalizing RRI, and highlight that a theory-based, integrated and nuanced typology is currently missing. The method section explains how we selected the case studies, the information we processed, and how we carried out the within-case and the cross-case analyses. The findings section provides rich insights into the ten projects, which are analyzed, interpreted and integrated into a typology in the discussion section. The final chapter provides recommendations for research and innovation policy, institutionalizing RRI in the businesses sector and for future research in this field.

Background

In science and policy, the discourse on the societal responsibility of research and innovation is diverse and framed by similar but still substantially different concepts, such as Responsible Innovation and Responsible Research and Innovation.³ Responsible Innovation (RI) is rooted in scholarly debates on Science and Technology Studies, anticipatory governance, technology assessment, and social innovation (Barben et al. 2008; Lubberink et al. 2017). It emphasizes the future-oriented notion of responsibility, care, and responsiveness, acknowledges the power of innovation to create futures, and aims at engaging with society (Owen and Pansera 2019). Stilgoe, Owen, and Macnaghten (2013) define Responsible Innovation as ‘taking care of the future through collective stewardship of science and innovation in the present’. Tensions between society’s interests and corporate interests (e.g. regarding the protection of intellectual property rights) are discussed (Blok and Lemmens 2015) and challenges of the predominant paradigm of the market economy are addressed. However, concrete guidance for the business sector or individual corporations acting within this paradigm is still missing. Responsible Research and Innovation (RRI) originates from a public policy discourse within the European Commission. While in the beginning RRI also followed a process-related approach⁴, it became an umbrella for repositioning action lines of Horizon 2020 (Owen and Pansera 2019) and a cross-cutting issue for the whole Framework Program. It was substantially put forward by the Horizon 2020 sub-program ‘Science with and for Society (SwafS)’ (Delaney et al. 2020), which consisted of a series of calls for proposals in the areas of gender equality, public engagement, science education, open access and data, governance and ethics, and science education. The interim evaluation of SwafS criticized the relatively low participation of companies (mostly suppliers of services and consultancies) in the funded projects and suggested a greater involvement of for-profits to increase the integration and institutionalization of RRI in the business sector (European Commission 2017).

Literature review

While EU policy tried to support the transfer, application, and institutionalization of RRI in the business sector, only a few scientific studies have linked RI/RRI to business practices: Iatridis and Schroeder (2016) demonstrated the usability of CSR⁵ tools for RRI, arguing that to be practically relevant to private companies, RRI needs to take their interests and incentives into account, although they did not provide any insights into potential resulting tensions or underlying conflicts of interest. Dreyer et al. (2017) presented a consensus view on Responsible Innovation by a group of industry practitioners who shared the view that RRI fails to have an impact on the business sector because RRI-related concepts, tools, and methodologies are not aligned with current industrial practices. They suggested linking RRI better to discourses on CSR, Shared Value, the moral responsibility of businesses, responsible digital innovation, and sustainable investment policies, but did not offer any framework for how this could be done. Lubberink et al. (2017) assessed the conceptual differences between responsible innovation and social innovation and derived conclusions as to how they could be implemented in a business context, but did not provide further insights on different contexts and how they lead to heterogeneous practices. van de Poel et al. (2017) developed a conceptual model of RRI for the business sector that links a company’s RRI strategy to its context and translates it into activities that result in RRI-related outcomes that can be assessed by company-specific RRI key performance indicators. However, it does not provide any further details on potentially conflicting interests or tensions. Scherer and Voegtlin (2020) developed a framework for responsible innovation based on reflexive and participative corporate governance, derived corporate governance policies, and highlighted that current practices fail to set the right incentive structures for businesses to meet societal challenges, but did not provide more in-depth insights into such incentive mechanisms. Owen et al. (2021) identified factors influencing the dynamics of the institutionalization of RRI in a major national public research-funding organization, while Steen, Nauta, and Gaasterland (2018) provided insights into institutionalizing RRI in a major research-performing organization. Both suggested similar studies on institutionalizing RRI in the business sector and assumed that their findings might not be transferable because of different logics and interests, but did provide a nuanced view of the differences between the different contexts.

While most of the aforementioned studies focused on the analysis of specific theoretical and/or practical challenges of RRI in the business context, no typology or framework has shed light on the deeper roots of the difficulties in transferring RRI to and institutionalizing RRI in the business sector. In particular, recent scientific debates on frames, interests, and incentives have not been analyzed in an interlinked way and no empirical evidence on these aspects has been analyzed.

Our paper addresses this research gap by developing a comprehensive typology based on empirical evidence that considers the particularities of RRI in competitive environments. It highlights the suggestions of previous studies that framings of RRI might need to be adapted to the business sector (Dreyer et al. 2017), that RRI has to consider the profit- and competitiveness-driven interests of the business sector and potentially resulting tensions (Brand and Blok 2019), and that specific incentives are needed to support RRI in the business sector (Gurzawska, Mäkinen, and Brey 2017).

Discursive frames

RRI is a multifaceted concept that aims to align research and innovation with the values, needs, and expectations of society (Rome Declaration 2014). The framing, interpretation, and operationalization of the general idea of responsibility are significant to the implementation, dissemination, and institutionalization of RRI.

Frame analysis, a broadly applied and flexible method for studying social constructions of reality, originates in the work of Erving Goffman (1974). He highlighted that culturally determined definitions of reality allow people to make sense of objects and events and that an issue or situation can be viewed from a variety of perspectives. In this context, frames are resources for meaning construction (Cornelissen and Werner 2014), sense-giving processes (Fiss and Zajac 2006), and understanding and interpreting different issues (Entman 1993). Cornelissen and Werner (2014) differentiate between micro-level frames (i.e. cognitive frames that influence the decisions and perceptions of individuals), meso-level frames (for collective meaning construction between in-groups), and macro-level frames (structuring the construction of meaning, expectations, and behavior of institutional fields). Girschik (2016) differentiates between institutional frames (widely-shared and taken-for-granted schemes), interactive frames (processes in which actors co-construct meaning), and discursive frames (rhetorical strategies to influence others). Through discursive framing, social actors can shape and direct audiences’ perceptions and decisions (Waller and Conaway 2011). In addition, their understanding and interpretation of an issue can be guided to influence their sense-making processes (Fiss and Zajac 2006). Over time, discursive frames can become institutionalized in the form of cultural frames and internalized by individuals as cognitive frames (Lakoff 2010).

Our study focuses on discursive frames at the meso-level (Nadkarni and Narayanan 2007), where frames are used to influence perceptions and decisions in and between organized groups. Discursive frames can be applied for strategic purposes to reduce the resistance to change and support the process of reshaping and reimagining current frames (Fiss and Zajac 2006; Gioia and Chittipeddi 1991; Maitlis and Lawrence 2007; Mantere, Schildt, and Sillince 2012). Introducing new frames has to consider context, often builds on already established frames, and makes use of creative extensions and combinations in language (Cornelissen and Werner 2014). Kaplan (2008) offered a framing contests model, which suggests that actors’ framing in social interactions reflects their interests and supports a strategic direction or change. In the business sector, different discursive frames of responsibility are well established, such as corporate social responsibility (Rodriguez-Gomez et al. 2020), corporate citizenship (Matten and Crane 2005), shared value (Porter and Kramer 2011), social impact measurement (Maas and Liket 2011), and corporate sustainability (Montiel and Delgado-Ceballos 2014).

In innovation policy, the quadruple helix model (Carayannis and Campbell 2010) is a well-established frame of collaborative innovation involving research, business, policy, and citizens. In innovation management, a variety of frames focus on different aspects of participation, and deliberation can be found in parallel, such as open innovation (Chesbrough 2003), co-creation (Sanders and Jan Stappers 2008), user-centered design (Wilkinson and Angeli 2014), and design thinking (Brown 2008).

In the context of RRI, discursive frames have played an important role, as three frameworks are most commonly used⁶: von Schomberg (2013) differentiated between the product dimension of RRI (if a product is ethically acceptable, socially desirable, and sustainable) and the process dimension of RRI (referring to responsive, adaptive and integrated management of an innovation process). Stilgoe, Owen, and Macnaghten (2013) distinguished four dimensions of RRI constituting their so-called ‘ARRI framework’: Anticipation (comprising activities like foresight, technology assessment, and scenarios), Reflexivity (with approaches like embedded social scientists and ethicists, codes of conduct, and multidisciplinary collaborations), Responsiveness (referring to value-sensitive design, open access, and moratoriums), and Inclusion (concerning consensus conferences, deliberative polling, focus groups, and citizens’ panels). The EC followed a less conceptually coherent understanding with their so-called RRI keys (engagement, gender, ethics, science education, and open access), which reflected action lines in SwafS and its predecessor program, SIS (Science in Society), as well the organizational structure of the EC units associated with them (Owen, von Schomberg, and Macnaghten 2021).

There is currently no comprehensive analysis of RRI frames and related frames in the business sector.⁷ Our study applies a discursive institutionalism approach and addresses focuses on the discursive frames of RRI and their interrelations with interests and incentive mechanisms.

Interests

The debate about conflicting interests of businesses and society and the responsibility of companies beyond making profits can be traced back to the 1920s (Berle 1931; Dodd 1932; Donham 1927), was taken up after the second world war (Abrams 1951; Bowen 1953), and has been a contested field ever since. While several highly influential scholars highlighted that the business sector should only aim at achieving profits, growth, or market shares (Friedman 1962; 1970), there is still an emerging debate as to if and how societal goals and interests can be integrated into corporate strategies, decisions, and actions (Carroll 1979; 1999; Dahlsrud 2008; Matten and Crane 2005; Rodriguez-Gomez et al. 2020).

A substantial number of scholars highlighted the converging interests of society and businesses and aimed at establishing a business case for responsible and sustainable business conduct that would motivate companies to consider societal interests out of self-interest. Among the theoretical contributions, the shared value concept (Porter and Kramer 2006; 2011) was a highly influential (Bosch-Badia, Montllor-Serrats, and Tarrazon 2013; Epstein-Reeves 2012; Moon et al. 2011), but also contested approach (Aakhus and Bzdak 2012; Baraka 2010; Crane et al. 2014). The authors claimed shared value as a new conception of capitalism, seeking to improve the relationships between business and society by aligning social benefits with corporate self-interest, reconceiving products and markets, redefining productivity in the value chain, and enabling local cluster development. Conceptually not linked to shared value, several thousand empirical studies aimed at showing positive relationships between corporate social, environmental, and financial performance (Friede, Busch, and Bassen 2015; Hirsch et al. 2022). Most of them found positive but small links, complex interrelations, and a wide variety of mediating and moderating variables (Dixon-Fowler et al. 2012; Endrikat, Guenther, and Hoppe 2014; Schnippering 2019).

In contrast to the assumption of converging interests, a substantial number of critical voices highlighted the conflicting interests of society and businesses and called for state intervention and strong policy instruments to internalize social costs and correct price signals (van den Bergh 2010), a role that states can often not fulfill due to globalized markets, regulatory gaps, and loss of national sovereignty (Scherer and Palazzo 2008). As a consequence, businesses tend to fill the gaps of governments, become public actors, and thus become subject to the principles of democratic governance (Scherer and Palazzo 2011). On the corporate level, the shared value concept was criticized as ignoring the tensions between social and economic goals (Crane et al. 2014), and conflicting interests were perceived as the main drivers of greenwashing (Reich 2008) and CSR decoupling (the difference between what is being portrayed in CSR reports and what firms are doing) (García-Sánchez et al. 2021; García-Sánchez et al. 2021).

While the debate about conflicting interests has a long history in economics, political science, and CSR management, the debate about potentially conflicting interests in implementing RRI in general and in businesses in particular is just beginning. Dreyer et al. (2017) argued that RRI-related concepts, tools, and methodologies are not aligned with current industrial practices, but did not further expand on conflicting interests. Iatridis and Schroeder (2016) argued that to be practically relevant to private companies, RRI needs to take their interests and incentives into account, but they did not provide any insights into potential resulting tensions. Schönherr, Martinuzzi, and Jarmai (2020) outlined a business case for RRI, showing that responsibility and innovation can mutually strengthen each other, and suggested additional research to provide insights into how responsible innovation may support companies in generating competitive advantage. Brand and Blok (2019) identified three forms of tension in the ways that competitive markets operate on the one hand, and the social and ethical requirements of innovation processes on the other: (a) tensions resulting from competing for resources and time, as stakeholder involvement may lead to delays in the innovation process, potentially reducing both competitiveness and the innovative capacity of companies; (b) tensions between openness and business interests in protecting intellectual property, as transparency might risk a company’s potential to achieve competitive advantage from their innovations; (c) tensions between deliberative engagement and corporate governance structures, as most corporate governance bodies have to prioritize financial interests while deliberative engagement and responsible innovation move innovation governance beyond mere monetary considerations.

An in-depth analysis of potentially conflicting interests of businesses and society related to RRI and how to deal with them is currently missing. Our study addresses this by focusing on potentially conflicting interests concerning RRI and their interrelations with discursive frames and incentive mechanisms. As some of these phenomena might be implicit, we applied methods of data collection and analysis that allow shedding light on underlying assumptions.

Incentive mechanisms

In order to disseminate and institutionalize RRI in the business sector, a wide variety of policy instruments can be used by governments to incentivize desired corporate behavior (Bemelmans-Videc, Rist, and Vedung 1998; Howlett and Ramesh 1993; Jordan et al. 2003; Steurer 2011; Steurer, Martinuzzi, and Margula 2012): (a) legal instruments derived from legislative, executive, and judicial powers, impose certain behaviors, and based on underlying rationales of hierarchy and authority (e.g. laws, directives, and regulations); (b) Financial instruments based on the taxing authority and the treasury of states that influence behavior through financial incentives and market forces (e.g. taxes and subsidies); (c) Informational instruments based on knowledge delivery combined with moral or factual persuasion, highlighting options and possible consequences (e.g. government-sponsored campaigns, trainings, or websites); (d) Partnering instruments involving government agencies and businesses to exchange complementary resources aimed at self-governance of the business sector (e.g. public-private partnerships, negotiated agreements, stakeholder forums, and dialogues); and (e) hybrid instruments combining several of the abovementioned instruments (e.g. strategies or platforms).

In the area of corporate responsibility, weakened nation-states have increasing difficulties in regulating global business (Steurer 2011) and therefore make use of a variety of soft-law policy instruments, such as (a) awareness campaigns and capacity building, (b) provisions for information disclosure and transparency, (c) activities to facilitate responsible investment, and (d) sustainable public procurement. A similar typology has not been applied to the specific field of RRI and comprehensive and structured analyses of RRI policy instruments addressing the business sector are currently lacking. This early stage of research is mirrored by Scherer and Voegtlin (2020), who highlighted that it is unclear whether existing incentive mechanisms support or impede sustainable and responsible innovations, called for ‘right incentives’ and ‘favorable conditions’ (without specifying what they would look like), but also noted that such conditions are difficult to achieve in a globalized environment.

The most specific categorization of incentives for RRI in the industry was developed by Gurzawska, Mäkinen, and Brey (2017), who developed a matrix of incentives for stimulating the adoption of RRI based on (a) three dichotomies (external and internal, instrumental and non-instrumental, and direct and indirect incentives) and (b) a causal loop diagram, which helped them identify important driving forces for RRI in the industry: corporate reputation and critical consumerism, employee engagement, external (i.e. policy driven) incentives, and internal incentives (i.e. through corporate governance systems). The authors highlight the fact that industries would prefer incentives for the uptake of voluntary RRI over legally binding obligations and perceive soft-law policy instruments as more innovative, flexible, motivating, and better fitting to a globalized economy. At the same time, the authors argue that the effectiveness of such soft-law policy instruments depends on institutionalization and successful orchestration of stakeholder involvement.

A conceptual framework or typology of RRI-related policy instruments and derived incentive mechanisms is currently not available. Our study addresses this by focusing on incentive mechanisms for RRI in the business sector, their interrelations with discursive frames, and potentially conflicting interests.

Summarising the scientific discourse on RRI-related discursive frames, potentially conflicting interests, and incentive mechanisms, it becomes evident that there is neither evidence of their characteristics at hand nor any study that could provide a nuanced view of their interrelations. Our study therefore aims to integrate all three aspects into an overarching typology based on empirical findings from leading EU projects. Our research question is therefore: How did EU funded pioneer projects apply discursive frames, address potentially conflicting interests and utilize incentive mechanisms for institutionalizing RRI in the business sector?

Methods

This paper combines a qualitative case study approach (Yin 2009) with the Gioia methodology to provide insights into and derive theoretical contributions from phenomena in an emerging field of research. Our main goal was to generate a typology on a conceptual level, contribute to theory and, if necessary, challenge or expand it, which is why we chose the Gioia methodology (Gioia, Corley, and Hamilton 2012) focusing on what the authors call ‘qualitative rigor’. EU-funded RRI projects addressing the business sector are such a rich source, as they are (i) multiple-company cases, (ii) of high importance regarding scientific progress, and (iii) act as frontrunners of adapting RRI to the contexts and needs of the business sector. We thus combine rich empirical material with an approach that aims to move away from the purely descriptive towards a theoretical explanation that can contribute to a theoretical debate on institutionalizing RRI in the business sector: The Gioia Methodology. It is a qualitative methodological approach that refers to aspects of Grounded Theory (Glaser and Strauss 1967), provides greater rigor and employs a more systematic approach than unstructured case studies by (a) deliberately differentiating between first order codes (informant centered), second order themes and aggregate dimensions (theory centered), (b) constantly comparing data across cases and informants; and (c) presenting findings with data-based narratives using the second order concepts and aggregated dimensions, with reference to first order quotations.

Although interviews, data, and text analyses allow comparability of results and ensure substantial theoretical contributions and an in-depth understanding of terms, contexts, and logic of individual cases (Stake 2013), we agree with Gioia, Corley, and Hamilton (2012) that case studies do allow for transferability beyond our individual cases ‘if the case generates concepts or principles with obvious relevance to some other domain’ (24) and implemented our sampling strategy accordingly.

Selection of case studies: our sampling strategy

Our cases were selected using theoretical sampling (Glaser and Strauss 1967) to generate relevant concepts that go beyond existing literature. We started by downloading data on all Horizon 2020 projects (https://cordis.europa.eu) on July 25th, 2022, and selecting RRI-related topics and projects in the SwafS sub-program. To include projects that relate to RRI but were funded beyond the SwafS sub-program, we included projects from other sub-programs that contained the terms RRI, responsible research, or responsible innovation in their titles, abstracts, or objectives. We excluded projects that targeted individual researchers (such as ERC or Marie Curie Actions) or only implemented single events (such as the ERA Career and Mobility Day), as these projects would not have any substantial impact on the business sector. We then started with a first classification of the project abstracts and objectives, excluding projects that only mentioned RRI but did not address it in their core activities. As a result, we identified 180 RRI projects funded by Horizon 2020 with a total EC contribution of 364 million Euros. In the next step, we analyzed the project abstracts and identified 20 projects that focused on the business sector.

From this data set, we excluded seven projects that had not ended yet, and could therefore not provide sufficient outcomes for our analyses.⁸ Of the remaining projects we identified a cluster of seven focusing on regional aspects of RRI (funded in SwafS-14-2018-2019-2020), from which we excluded four projects to counterbalance their over-representation in our sample.⁹ During the first round of preparatory talks, several project coordinators drew our attention to a project that had already been funded in the 7th EU Framework Program (RESPONSIBLE INDUSTRY), and that was perceived to be the first to address RRI in the business sector. Therefore, we included this project in our sample, which finally comprised ten projects (see Table 1):

Table 1. Our sample of 10 EU-funded RRI projects addressing the business sector (data source https://cordis.europa.eu/).

Project	Title	Level	Coord.	Start Date	total Cost (mil Euro)	no. of partners	Topics
			funding scheme	End Date	EC contrib. (mil Euro)	business sector partners
SeeRRI	Building Self-Sustaining Research and Innovation Ecosystems in Europe through Responsible Research and Innovation	Regional	REC	01.01.2019	2,0	12	SwafS-14-2018-2019-2020
			CSA	31.10.2021	2,0	2
DigiTeRRI	RRI Approach for Transitioning the Traditional Industry Regional into Digitalized Industry Territories	Regional	REC	01.01.2020	2,0	12	SwafS-14-2018-2019-2020
			CSA	31.12.2022	2,0	2
RRI2SCALE	Responsible Research and Innovation Ecosystems at Regional Scale for Intelligent Cities, Transport and Energy	Regional	REC	01.01.2020	2,0	11	SwafS-14-2018-2019-2020
			CSA	31.12.2022	2,0	2
RESPONSIBLE INDUSTRY	RRI in Business and Industry in the Domain of ICT for, Health, Demographic Change and Wellbeing	Company	HES	01.02.2014	1,9	10	FP7-SIS.2013-1
			CSA	31.07.2017	1,5	-
PRISMA	Piloting RRI in Industry: a Roadmap for tranSforMAtive Technologies	Company	HES	01.08.2016	1,7	5	GARRI-2-2015
			CSA	31.07.2019	1,7	-
COMPASS	Evidence and Opportunities for Responsible Innovation in SMEs	Company	HES	01.05.2016	1,5	7	GARRI-2-2015
			CSA	30.04.2019	1,5	1
SMART-map	RoadMAPs to Societal Mobilisation for the Advancement of Responsible Industrial Technologies	Company	HES	01.05.2016	1,5	9	GARRI-2-2015
			CSA	31.10.2018	1,5	2
LIV.IN	Implementing RRI through Co-Ctreation of Smart Futures with Industry and Citizens	Company	HES	01.05.2018	3,5	17	SwafS-06-2017
			CSA	31.10.2021	3,5	10
SCALINGS	Scaling up Co-creation: Avenues and Limits for Integrating Society in Science and Innovation	Process	HES	01.05.2018	4,0	14	SwafS-13-2017
			RIA	31.07.2021	4,0	-
SISCODE	Society in Innovation and Science through CODEsign	Process	HES	01.05.2018	4,0	18	SwafS-13-2017
			RIA	30.04.2021	4,0	3
		Total			24,2	99
					23,8	20

Our sampling strategy ensured that the ten case studies represent EU-funded RRI projects addressing the business sector, show a wide variety of aspects, and provide sufficient access to information to carry out a comparative case analysis. At the same time, with this careful sampling strategy, we ensured that our results were plausible beyond the selected cases and a ‘thick’ (Geertz 1973) basis for establishing new concepts. In total, 99 organizations from 23 countries collaborated on these projects, of which 85 participated in only one project, 13 in two projects, and one in three projects. Of these 99 partner organizations, 20 were companies, including multinational industrial companies and SMEs, manufacturing companies, service companies, and consulting and communication firms.

Among the ten cases, we identified three different levels, which also served as a heuristic for further analysis:

Regional Level: Three projects (DigiTeRRI, SeeRRI, and RRI2Scale) covering multiple regions in Europe were aimed at integrating RRI into regional development.

Corporate Level: Five projects focused on individual companies: RESPONSIBLE INDUSTRY was a piloting project in RRI for business in the previous EU Framework Programme, PRISMA, SMART-map, COMPASS, and LIV.IN focused on different sectors and sizes of companies, from small and medium enterprises and startups to large multinational enterprises.

Process Level: Two projects (SISCODE and SCALINGS) placed their focus on individual innovation processes and followed the idea of co-creation.

While these levels were set by the requirements of the calls in Horizon 2020, the choice of business sectors was not predetermined by the EC, and the consortia had much more freedom in this regard and set a thematic focus on high technology sectors with high relevance for society, such as nanotechnology, biology, Internet of Things, artificial intelligence, autonomous vehicles, robotics, urban energy, and health.

Data collection and analysis

To deepen the understanding of the observed phenomena, and to increase the generalizability of results, we follow the suggestions of Gioia, Corley, and Hamilton (2012) for analysis and started by assigning ‘terms, codes, and categories’ to open our data. First, we examined the ten projects separately and processed all publicly available information, such as project websites, reports, publications, and other deliverables. Second, we invited the project coordinators to qualitative expert interviews to gain additional information beyond the official documents and provide sufficient insights into our research question. We conducted six of these interviews online in August 2022 (average duration 45 min) and received two written inputs (average length one page). The interviews were conducted by teams of two, with one conducting the interview and the other taking notes (Eisenhardt 1989). We continued by coding this data and then proceeded to a 2nd-order analysis to detect those concepts that ‘don’t seem to have adequate theoretical references in the existing literature’. To enhance internal validity, three investigators analyzed the data, first individually and then together, following the principle of investigator triangulation (Flick 2004). This process brought us to increasingly abstract concepts, which we aggregated at the level of 2nd-order results. This then led us – again in line with Gioia, Corley, and Hamilton (2012) – to a ‘data structure’ and to results that go beyond existing literature. In deriving our typology, we first developed relevant dimensions and grouped cases and the results of analyses, identified meaningful relationships and characterized the constructed types. This approach follows concepts from Grounded Theory (Strauss and Corbin 1998) in the advancements by Gioia, Corley, and Hamilton (2012) and provides a ‘resulting grounded theory model’ (see also Gioia and Pitre (1990)).

Finally, we would like to mention two possible limitations: First, there might be a potential selection bias regarding the projects we excluded in our stepwise approach; second, there could be window-dressing of project outcomes, which we tried to counteract with the in-depth interviews.

Findings

In this section, we first describe our findings alongside the typology of projects (regional, corporate, and process), while in the discussion section, we integrate them into a coherent typology that shows interrelations and starting points for institutionalization strategies. Figure 1 presents an overview of the findings we derived from the project documentation and the interviews with project coordinators, which we detailed in the following sub-sections. The terms used in this figure are grounded in data from the project outcomes and interviews with project coordinators and represent first order codes (informant-centered), following the Gioia Methodology.

Figure 1. Findings from project documentation and interviews.

Regional level projects

Three projects in our sample took a regional approach to institutionalizing RRI in the business sector: SeeRRI focused on developing a framework for integrating RRI principles into regional Smart Specialisation Strategies, developed a conceptual framework, a process model, and policy recommendations for the target regions and shared the project outcomes through a series of webinars, a final conference, and a network of other regions. DigiTeRRI developed RRI-based roadmaps on responsible digitalization in three exemplary industry regions, delivered regional visions, objectives, and implementation measures, and disseminated them through a so-called Digital Cooperation Day, an International Incubator Exchange, and a final conference. RRI2SCALE examined how RRI can be integrated into regional research and innovation ecosystems based on four pilot regions, delivered roadmaps for each of them, and designed implementation processes and exchanges beyond the pilot regions through a series of stakeholder dialogues.

Two of the projects (SeeRRI and RRI2SCALE) did not focus on any specific industry, while DigiTeRRI specifically focused on digitalization and ICT. Most of the target regions were located in ‘most developed regions’ or ‘transition regions’, while only a very few cases were located in ‘less developed regions’¹⁰ and not a single pilot region was located in Eastern Europe. All three projects were funded as Coordination and Support Actions (CSA)¹¹, were coordinated by research organizations (REC),¹² and included public authorities as well as businesses as consortium partners. All three projects were funded by the same H2020 Topic (SwafS-14-2018-2019-2020) which explains a certain coherence in their approaches, i.e. mapping the current regional research and innovation ecosystem, integrating RRI into regional development policies, involving a broad spectrum of stakeholders and citizens, and institutional changes in participating organizations.

Discursive frames

All three projects referred to the ‘Quadruple Helix of Innovation‘¹³ as a fundamental principle of the consortium composition:¹⁴ The SeeRRI website states, ‘All of the ‘quadruple helix’ is represented in the consortium: government (several regional government bodies), businesses (SMEs, an economic cluster), academia (research institutes, universities), and civil society (a confederation of enterprises) and DigiTeRRI describes itself as ‘a 12-partner collaborative project […] conducted in conjunction with European quadruple helix stakeholders in the consortium’ and the RRI2SCALE website mentions the Quadruple Helix as a ‘win-win-approach for reaching out to a diversity of stakeholders’. The expert interviews showed that involving this diversity of consortium partners provided additional benefits for the projects as well, as one interview partner stated: ‘As we already had the Quadruple Helix represented in the consortium, we could understand their different perspectives and objectives’. In implementing the projects the Quadruple Helix served as an inclusive strategy for involving a broad range of actors in workshops, and was referred to in multiple project outcomes, as well as in the interviews, e.g. ‘There must be a framework [i.e. the Quadruple Helix] for the territory to make organizations in their ecosystem work together’. This evidence indicated that the Quadruple Helix served not only as a headline but as a discursive frame of the regional projects.

The three projects interpreted RRI following the RRI keys of the European Commission, but with certain variations: DigiTeRRI stuck to the five original keys (i.e. gender equality, science education, open access, public engagement, and ethics), which were central to the project’s roadmap. RRI2Scale and SeeRRI added sustainability as a sixth key. The dominance of the RRI keys as a discursive frame of the regional projects is caused by the fact that they wanted to meet the expectations of the European Commission as a funder represented by the project officers, who also monitor the progress and quality of the projects’ implementation. One of the interview partners explicitly stated: ‘We received the money from the commission so definitely we need to use something that they define right […] it is their money […] so of course the five keys should be somewhere […] this is what we have to do for the report.’ However, the practical relevance of the RRI keys to the business sector was questioned by some of the interview partners, e.g. ‘We decided to do RRI by design and RRI in practice, but to not talk about RRI [keys with businesses] […] they don't care, they don't even know what it means and they are not their interest.’ These findings indicate, that the RRI keys are an established discursive frame within European Commission and thus taken up by European projects, but not in the business sector. This might be caused by a lack of knowledge, incompatible terminology, or conflicting interests (see next chapter).

Interests

All three regional projects referred to developmental and society-oriented policy goals, such as wealth, wellbeing, and regional economic development, with a focus on the respective region, but very rarely highlight tensions or conflicts of interests between the different actors of the quadruple helix, between different stakeholder groups or between societal goals and the interests of the business sector. In such cases, conflicts were perceived as flaws of individual participation processes - and not as a structural conflict of interest - that could be overcome by better process design and communication. The fact that different social groups have different and potentially conflicting interests is neither mentioned in the project documents nor in the expert interviews. On the contrary, one of the interview partners stated, ‘in the end, is about achieving an impact which is not only economic but is societal and environmental’. Societal aims dominate, as the Quadruple Helix provided a discursive frame that highlights shared interests and outshines potential conflicts, while policy, business, and research are expected to collaboratively achieve them.

Regarding potential conflicts between the principles of RRI and business interests, the project documents did not provide any evidence. However, we were able to conceptualize relevant results from the expert interviews. One interview partner mentioned conflicting interests in the area of Open Access: ‘the EC really wants to see open access […] companies want it because they can learn, but on the other hand they don’t want to share information […] so they only share what everybody already knows’. Another interview partner mentioned that businesses were overrepresented in the development of regional strategies and ‘our challenge was to involve citizens to balance these interests […] some policymakers even complained that the business sector interests were too powerful […] so the policy makers felt less powerful because they were pushed by the business sector and the citizens are not existing.’ One interview partner referred to the vagueness of RRI that allows hiding potential conflicts by stating, ‘The concept is beautiful and everybody loves it, but when you go into details, everybody loses interest […] so if we do not go into details, people cannot see the conflicts’. Conflicts were not even evident and never became dominant in regional projects, as RRI remained vague and was embedded in a well-established and policy-driven frame of collaborative action.

Incentive mechanisms

The most important incentive mechanism for integrating RRI into regional development strategies and - as a consequence - into businesses, stems from the European Union: while the European Structural Funds provide (co-)funding of regional development programs and projects, they also require regions to develop ‘Smart Specialization Strategies‘ aimed at identifying strategic areas for innovation and development with wide stakeholder involvement. All three regional projects took these well-established incentive mechanisms as their main anchor point for integrating RRI into regional development as the principles of RRI were perceived as compatible with the already existing policies, processes, and strategies. As one interview partner stated: ‘it's the best opportunity to integrate RRI because it's already there and you just do an incremental change to create added value.’ As a result, the projects included principles of RRI in the co-development or revision of the regional specialization strategies and aimed to include RRI in the strategy documents as well. Summarizing this evidence, the regional projects utilized policy-driven monetary benefits as incentive mechanisms to disseminate RRI in their target regions and beyond.

Corporate level projects

Five projects in our sample focused on individual companies in their attempts to institutionalize RRI in the business sector: RESPONSIBLE INDUSTRY developed and tested a benchmarking method for RRI maturity and published a guideline for the ICT industry. The project was implemented substantially earlier (2014-2017) and laid the conceptual foundations for the other projects. The PRISMA project developed an RRI-related management standard and key performance indicators (KPIs) as well as guidelines for how to develop RRI-based long-term strategies. COMPASS developed an interactive online platform for small and medium-sized Enterprises, a self-check tool, and a variety of online resources. SMART-map delivered implementation plans on how to integrate RRI in high-tech sectors. LIV.IN focused on large companies from the ICT sector, established an online expert community, organized a series of RRI-related co-creation workshops, published a toolkit for effective co-creation and inclusive innovation, and disseminated showcases of RRI as a business opportunity for companies.

All projects have set a thematic focus on high technology sectors with high relevance for society, such as nanotechnology and biotechnology, the Internet of Things, artificial intelligence, autonomous vehicles, robotics, urban energy, and health. The choice of the business sectors was not predetermined by the EC – the consortia had much more freedom in this regard. All five projects were funded as ‘Coordination and Support Actions (CSA)‘ and were coordinated by Universities (HES).¹⁵ While the business sector was the main target group and involved in a broad variety of activities, only LIV.IN substantially involved companies as consortium partners; three projects also included non-profit organizations as consortium partners (PRISMA, COMPASS, and LIV.IN); and not a single project included a public authority as a consortium partner. Three projects were funded by the same H2020 Topic (GARRI-2-2015), which explains certain similarities in their approaches, e.g. actors from industry, research, and civil society jointly developed roadmaps and implementation plans for the responsible development of particular technologies, products, or services; a comparative assessment delivering evidence of the benefits for industry; activities to facilitate the development of new business models based on ‘an RRI-inspired CSR 2.0’.

Discursive frames

All five projects intensively referred to Corporate Social Responsibility (CSR) in their conceptual work, project implementation, and dissemination: The RESPONSIBLE INDUSTRY project took existing CSR practices as a starting point for developing pathways for how RRI could be implemented in companies. The PRISMA project highlighted that many companies already have a CSR policy and suggested presenting RRI as CSR for the innovation process. The COMPASS project understood RRI as an extension of CSR to the realm of a company’s research and innovation processes and highlighted the importance of a strategic approach in the early stages of research and innovation. The smart-MAP project suggested to build-up alliances between RRI and CSR to generate new ideas on how to foster the adoption of RRI principles. The LIV.IN project analyzed CSR-related multi-stakeholder initiatives and derived a policy brief for establishing such an initiative for RRI in the business sector. The interviews provided additional insights into the projects’ use of CSR as a discursive frame into which RRI could be embedded. One interview partner stated, ‘In CSR you traditionally do not have a strong emphasis on research and innovation [..] and we say, look, if you want to be responsible, you not only have to care for the environment in your production, but you also have to think about what are the ethical and social issues that arise during the process and as a consequence of doing research and innovation’. The project documents and interviews showed coherent evidence that the corporate-level projects utilized CSR as a discursive frame and made use of certain similarities to RRI, as both frames refer to societal needs and concerns and hold companies responsible for their impacts. While this aspect is quite general in CSR (by the most prominent definition of CSR as a company’s responsibility for impacts), it becomes specific in RRI and Technology Assessment. Also, aspects of participation show similarities between the frames of CSR and the frames of RRI, although CSR addresses it in a rather limited form (e.g. in stakeholder dialogues or during materiality assessments), while RRI follows a broader approach (e.g. through deliberative engagement). In addition, CSR and RRI are both perceived as societal responsibilities on the one hand, but also as sources for strategic opportunities for the companies (e.g. by safeguarding their license to operate, reducing sunken costs, and improving their corporate image).

Open Innovation is another well-established frame that was addressed by all projects -sometimes referring to related concepts (such as Design Thinking, Social Innovation, UX Design, Lead User Innovation, etc.), sometimes with slightly different meanings, and often in different contexts. Smart-MAP understood it as the openness of large companies towards collaboration and networking with SMEs and startups. COMPASS highlighted that Open Innovation allows for involving customers and citizens to address sustainability challenges. Two projects (RESPONSIBLE INDUSTRY and COMPASS) incorporated openness of innovation processes in their self-check tools. LIV.IN extended the concept of Design Thinking to ‘Responsible Design Thinking’ and developed tools for inclusive innovation and quality criteria for co-creation processes. As openness is a core element of RRI (following the RRI keys and the ARRI framework) the projects considered it to be a perfect link to Open Innovation as a well-established discursive frame in companies, for which a broad variety of tools are already available.

All five corporate-level projects interpreted RRI following the process-oriented ARRI framework, with a special emphasis on participation and anticipation. Reflexivity and responsiveness were not so often highlighted and two projects (PRISMA and RESPONSIBLE INDUSTRY) added sustainability as a fifth dimension to the ARRI framework. Several interview partners mentioned that the ARRI framework is easier to explain and translate to the needs of the business sector than the RRI keys. However, most projects adapted the core ideas of the ARRI framework to the language and frames of businesses. One interview partner stated that they addressed the benefits of RRI ‘without necessarily using the term Responsible Innovation or the terminology of the Responsible Innovation discourse […] so we translated it into the language that companies may use or understand.’ Another interview partner highlighted the same intention ‘We tried not to use the RRI keys or the ARRI framework’. The LIV.IN project developed a ‘New Narrative for RRI in the business sector’ with the intention of changing the business sector’s perception of RRI from a societal demand and potential threat to a business opportunity. This new framework relates core principles of RRI (such as participation, anticipation, and sustainability) to well-established concepts in the business sector (such as CSR, Open Innovation and sustainability management). All of these activities can be interpreted as attempts to link one of the dominant and mostly cited discursive frames of RRI (the ARRI framework) with the well-established frames of the business sector. However, the projects did not address a single frame (or try to expand or change it), but addressed a wide variety of frames and highlighted a multitude of potential benefits of RRI for businesses. Therefore - and due to the limited budgets for dissemination and limited project duration - their impact on the discursive frames of the business sector was quite limited.

Interests

All corporate-level projects highlighted potential benefits of RRI for companies: LIV.IN developed a new narrative for RRI that addresses business opportunities and highlighted reducing risks, increasing value, developing more and better ideas, and becoming an industry game changer as key benefits of RRI; COMPASS emphasized the strategic value of RRI for high-tech startups; SmartMAP identified anticipating trends, reputation-building, responsible standards, improving product quality, and influencing innovation governance as key benefits; RESPONSIBLE INDUSTRY collected case studies on corporate benefits of RRI; and PRISMA linked RRI to the concept of business value.

Most of the projects observed tensions and conflicting interests as RRI could be perceived as an additional burden by companies that have to compete in global markets. One interview partner highlighted that ‘companies follow a pro-active approach and perceive innovation as a source of business opportunities, while RRI represents societal concerns that potentially limit these opportunities and might represent a negative view on innovation, as risks may be paramount.’ This general concern was specified by our interview partners in two regards: conflicting interests about openness and competitive disadvantages in time-to-market. Two interview partners referred to openness by stating that ‘when it comes to open innovation [… ] companies simply cannot align it with their competitive interest to protect knowledge and intellectual property’ and ‘they need to reveal a lot of information about what they do […] that would reduce their competitiveness’. Two interview partners explained: ‘Of course, RRI could delay [time to market] because of investigations of unwanted effects on the product and process’ and ‘If you do a good technology impact assessment or foresight process it takes a few months or even a year. And if your competitor doesn’t care [about impacts] you lose time to market. That’s a real problem for many companies.’ These examples show that RRI has to consider that companies act in a globalized competitive environment, where protecting intellectual property and time-to-market are important assets that some principles of RRI potentially impede or undermine.

Our findings show the difficulties experienced by company-level projects in their attempts to transfer RRI to the business sector. To overcome potentially conflicting interests, the projects focused on corporate benefits (mainly profits and competitiveness) but did not develop a coherent and sound understanding of obstacles and structural conflicts of interest.

Incentive mechanisms

All five corporate projects utilized two information-oriented incentive mechanisms to disseminate RRI in the business sector: well-documented case studies, which show the benefits of RRI for companies and address their self-interest, and hands-on tools, which allow other companies to follow these cases. In addition, two projects (RESPONSIBLE INDUSTRY and COMPASS) developed benchmarking tools that allow companies to assess their RRI maturity and compare themselves to anonymized data of other companies. One project (PRISMA) utilized international standards as an incentive mechanism and collaborated with the European Committee for Standardization (CEN) and several national standard organizations to integrate RRI into a technical document by CEN (CWA 17796:2021-Responsibility-by-design).

Several interview partners addressed the corporate interest of safeguarding their license to operate as a potentially powerful incentive mechanism for disseminating RRI, especially when societal concerns challenge the license to operate certain technologies, such as genetically modified organisms, nuclear power, or artificial intelligence. They suggested collating cases of major corporate losses due to irresponsibility to motivate others to institutionalize technology assessments and foresight studies. One interview partner emphasized the risk that RRI might be used for greenwashing by companies who would like to protect their license to operate and suggested minimum quality criteria for using the term RRI in corporate public relations.

Most projects tried to establish user groups and online communities beyond the consortium partners and hoped that being a member of such groups could serve as a network-based incentive mechanism for disseminating RRI in the business sector. However, the long-term effects of these networks were limited and one interview partner highlighted the difficulties of establishing self-sustaining communities of practice as follows ‘We only succeeded with synchronous and one-directional communication […] if we do something, they show up, if we don't do anything, nothing happens’. None of the projects succeeded to sustain a self-organized online community beyond the project duration; in some cases, the project websites are no longer online. These findings indicate severe limitations when individual projects try to establish incentive mechanisms on their own because of a limited budget, outreach, and time.

Summarizing this evidence, the five corporate-level projects tried to utilize a broad variety of information and network-based incentive mechanisms. These efforts are limited by the limited resources and outreach of single projects and the fact that these incentive mechanisms only worked when companies understood and embraced RRI as a benefit and not as a burden.

Process level projects

Two projects in our sample focused on individual innovation projects and processes for institutionalizing RRI in the business sector: SCALINGS studied three co-creation instruments across three high-tech domains and developed an interactive EU policy roadmap as well as specific recommendations for co-creation based on RRI. SISCODE experimented in ten co-creation labs with a broad variety of topics and delivered a virtual learning hub with a variety of tools, a platform for interested actors, and an interactive guidebook for practitioners.

Both projects covered a broad variety of topics, sectors, and regions, which were not predetermined by the EC. Both projects were funded as Research and Innovation Action (RIA)¹⁶, were coordinated by Universities (HES), and had substantially higher budgets and larger consortia than the regional- and corporate-level projects. Most of the project partners were Universities (HES) or Research Organizations (REC), while companies or non-profit organizations were rare and not a single public authority was involved as a consortium partner. Both projects were funded by the same H2020 Topic (SwafS-13-2017) which explains certain similarities in their approaches, i.e. aiming at a better understanding of co-creation processes and outcomes under various cultural, societal, and regulatory backgrounds and developing a new way of public engagement with co-creation involving European citizens.

Discursive frames

Both projects used co-creation as their central discursive frame as this was requested by the SwafS-13-2017 topic. Although this clear focus was set, both projects highlighted the broad variety of forms of co-creation: SCALINGS understood co-creation as ‘the practice of bringing together diverse actors in a joint innovation activity to mutual benefit’ and presented its definition of co-creation as ‘When diverse actors such as companies, universities, policymakers, and citizens collaborate in mutually enabled or supported innovation processes, they engage in co-creation practices’. SISCODE’s definition is: ‘Co-creation is a non-linear process that involves multiple actors and stakeholders in the ideation, implementation and assessment of product services, policies, and systems. It aims to improve their efficiency and effectiveness, and the satisfaction of those who take part in the process.’

Both projects carried out literature reviews, comparative analyses of co-creation case studies, and theoretical research on specific aspects (such as normative, political, socio-demographic, regulative, and economic contexts), developed concepts and models of co-creation ecosystems, and published roadmaps and tools. In all of them, co-creation is the dominant term (i.e. 46 out of 48 project documents and publications mention it explicitly, in total more than 10.000 times), indicating that process-level projects aimed at establishing co-creation as a new and dominant frame. Compared to this clear dominant frame, RRI was mentioned comparatively little, and neither the RRI Keys nor the ARRI framework was utilized as a discursive frame. Instead, both projects set co-creation equal to RRI and highlighted its participatory dimension: The SISCODE interactive guidebook emphasizes that RRI is ‘grounded on collaboration and synergy among actors who traditionally operate independently, including civil society, to cooperate in innovation.’ SCALINGS refers to RRI as a particular area of consideration in co-creation, among others, such as ‘Social Robustnes’, ‘Diversity in Innovation’ and ‘Limits of Standardization. Open Innovation is mentioned several times in the project documents, but always as a concept that embraces and fosters co-creation. Aspects of reflexivity were sometimes addressed, but, in general terms, and not in the specific meaning within the ARRI framework, while considering impacts of innovations were not the focus of both projects. CSR and corporate sustainability are not mentioned at all. This evidence is supported by interviews in which we could harden the concept of co-creation through analysis as a concept, while other concepts such as RRI, CSR, sustainability, or open innovation seemed irrelevant for process-oriented projects.

These findings indicate that the two process-related projects aimed to establish co-creation as a new, independent, and unique discursive frame that is only loosely linked to and supported by RRI and does not refer to any other frames of responsibility or sustainability. This attempt to establish a dominant frame is mirrored by one interview partner who aimed to establish a ‘culture of co-creation’.

Interests

Both process-related projects addressed challenges of co-creation in their publications, such as power imbalances among the actors involved, lack of resources – especially from citizens, alignment of goals and values, coordination issues, and limited trust and evaluating impacts when actors have different views on what constitutes success. Whenever such potential tensions were mentioned, they were attributed to power imbalance or social contingencies that could be resolved by ‘a common language and method to reflect on and in the process’ or by ‘a shared understanding developed through innovative thought processes for positive change and societal impact’. To facilitate such processes, one project developed a canvas that focuses on dialogue, transparency, and empathy.

One interview partner was surprised to be asked about conflicting interests and potential tensions resulting from RRI. He perceived co-creation as a new way of collaborative innovation that neither addresses nor creates any conflicting interests but rather follows a vision of a collaborative economy. When directly asked if competitive environments pose certain challenges, the interview partner replied: ‘I don't share your view about the needs of the companies to compete. Companies that are very much protecting their products are very old in their mentality because I think that open innovation is one of the fastest ways to produce new solutions. When solutions or ideas are in the air, I think that completing is much better than competing.’ This quote indicates not just a conflict of interest, but an alternative view on the economy, as for this interview partner, co-creation seems more than a discursive frame, but a different set of values and principles that the business sector should follow.

These examples show that the two process-level projects did not consider, address, or resolve potentially conflicting interests. They did not highlight any tensions before a co-creation process starts (about an innovation agenda or the focus of a co-creation process), or after a co-creation process ends (about the share of profits that an innovative solution, product, or service might generate). This leads to the impression that the strong focus on collaborative and participatory aspects has a certain tendency to hide conflicting interests.

Incentive mechanisms

Both projects relied on two incentive mechanisms: the dissemination of deliverables such as reports, journal publications, and tools (information-based incentive mechanism) and participants of co-creation labs acting as multipliers (network-based incentive mechanism). While SISCODE implemented a detailed and sophisticated exploitation strategy and established a platform to share and learn about co-creation, SCALINGS suggested changes in public procurement law, intellectual property law, and experimental law to incentivize co-creation. As both projects were Research and Innovation Action, they strongly relied on the dissemination and uptake of their research findings, as one interview partner highlighted: ‘Our results are not sold as consulting products or registered trademarks, but I would say they are very much recognized. We are using them in other projects, and we have been involved because of them in other projects […]. Our outcome in terms of institutionalization is our recognition.’ He highlighted that the projects’ approach to co-creation has already been taken up by one of the new missions in the EU Framework Program Horizon Europe.

Discussion

The purpose of this study was to explore how EU-funded pioneer projects applied discursive frames, addressed potentially conflicting interests and utilized incentive mechanisms to derive theory centered second order themes and aggregate dimensions, and to integrate them into a typology that provides a nuanced view on institutionalizing RRI in the business sector. In this section, we first discuss the findings along the three levels the projects addressed, followed by a discussion that refers back to the scientific state of debate on frames, interests, and incentives as aspects of institutionalizing RRI in the business sector.

Our findings illustrate that regional-level projects are linked up to well-established, dominant, and society-oriented discursive frames (i.e. Quadruple Helix). The complementarity of assets, interests, and actors stands in the foreground, while competitive elements are regarded as less important, or only relevant for location competition with other regions. European funding provides incentives for collaboration (e.g. to jointly source funding from the EU) and motivates regions to apply the RRI keys. Strategies for institutionalizing RRI in regional projects are aimed at integrating RRI into regional strategies and roadmaps. All these aspects balance business interests with societal needs and policy goals, establish a narrative of collaboration, and facilitate integrating RRI-related aspects. On a more abstract level, regional projects show that a dominant discursive frame embedded in strong incentive mechanisms and (at least the assumption of) synergetic interests result in favorable conditions for institutionalizing RRI in the business sector.

Corporate-level projects addressed a multitude of discursive frames of the business sector and tried to extend them with elements from the ARRI framework. The diversity of frames and the limited duration of the projects led to more diverse and less coherent actions than the regional- and process-level projects. RRI was often perceived by companies as an additional burden that competitors outside Europe do not have to consider. The projects tried to address this resistance to change by highlighting RRI-related business opportunities, translating societal needs and concerns into profits and losses, or establishing RRI-related management mechanisms, such as key performance indicators, maturity checks, or standards. All of these attempts tried to overcome but did not manage to resolve, the quite explicitly formulated conflicting interests of companies (profits, competitiveness, and shareholder value) and the societal needs and demands that stand behind RRI. The projects relied on the dissemination of information and on establishing networks as informal incentive mechanisms, but could not make use of policies or market forces as incentives for RRI. Potential strategies for institutionalizing RRI in the business sector could be based on monetizing the potential RRI-related benefits and the risks of non-action, as this would allow depicting RRI in profits (relevant for shareholders) and prices (key components of markets). On a more abstract level, corporate projects provide insights into situations where competing frames collide, underlying reasons for tensions and conflicting interests, and market- and profit-based strategies for institutionalizing RRI in the business sector.

Process-level projects emphasized co-creation as a unique, independent, and dominant discursive frame, promote an attitude of togetherness (which perfectly links to the participatory aspects of RRI), and focus on the collaboration of businesses with users, potential customers, stakeholders, and citizens. At first sight, businesses and their counterparts are not competitors, as citizens are interested in businesses addressing their needs, and businesses are interested in learning more about their potential customers. However, it has to be considered that this approach runs the risk of hiding conflicting interests beyond the scope of a co-creation process (e.g. conflicting interests about the ‘right’ innovation agenda before a co-creation process starts, or conflicts about sharing the profits of new products, services, or solutions that were developed in a co-creation process). As all process-related projects were coordinated by universities, publications and recognition of the project and its results were the only incentive mechanisms utilized, while neither policies nor markets were used. Strategies for institutionalizing RRI result from the fact that the EU framework programs publish subsequent calls in similar areas which allow project partners to establish their results and themselves as key players in major projects that follow. On a more general level, the process-related projects show how a dominant frame can be established, potential conflicts can be partly hidden, the origins of project partners predefine the incentive mechanisms used, and that sequences of major projects are important pathways for institutionalizing RRI, including in the business sector.

Summarizing our findings on discursive frames across the three levels, our results show a high terminological diversity, as nearly every project defined its own terms, spoke its own language, and promoted its own deliverables. There is obviously a lack of collaboration among EU-funded projects and of integration of their results into a bigger picture. As a consequence, no overarching discursive frame of RRI in the business sector has emerged or been established to date. The projects followed three different strategies: regional-level projects integrated RRI into a well-established and dominant frame (the quadruple helix), corporate level-projects tried to relate RRI principles to a multitude of existing frames, while process-level projects put all emphasis on establishing co-creation as a new, unique, and dominant discursive frame. These cases confirm what Cornelissen and Werner (2014) highlighted, that introducing new frames has to consider context, build on already established frames, and make use of creative extensions and combinations. Although the attempts of the projects to influence the discursive frames of businesses are evident, our analysis indicates that they were neither strategically planned within the projects nor coordinated between them. As a consequence, the high potential of discursive frames to influence perceptions and decisions (Nadkarni and Narayanan 2007) and to reduce the resistance to change (Maitlis and Lawrence 2007; Mantere, Schildt, and Sillince 2012) of the whole business sector, has not been used to its full extent by now.

Our finding on potentially conflicting interests provide empirical insights on tensions that hamper institutionalizing RRI in the business sector (Brand and Blok 2019). While conflicting interests have been intensively discussed in the scholarly debate about CSR, they were rarely addressed in our case studies and by our interview partners. This might be caused by some of the discursive frames the projects applied (i.e. the Quadruple Helix on the regional level and co-creation on the process level). However, these frames might underestimate, hide, or outshine conflicting interests, while they still exist. For example, interviews with regional-level projects revealed that conflicting interests were perceived as a sort of ‘old-fashioned economy’ that co-creation tries to transcend and overcome. As a consequence, the projects did not provide any further insights into resistance to institutionalizing RRI in the business sector. Only corporate-level projects explicitly addressed conflicting interests, attempted to reframe RRI as a business case rather than a societal demand, highlighted potential benefits, and emphasized risks of non-action.

Our findings on incentive mechanisms provide starting points for designing effective policies and governance systems that support the dissemination and institutionalization of RRI in the business sector. While incentive mechanism are discussed regarding CSR policies (Steurer 2011), our study takes a first step in developing a similar approach for RRI. Our findings show that only regional-level projects managed to utilize financial policy instruments (which are perceived as rather strong policy instruments), while corporate- and process-level projects mainly used informational and network-based policy instruments (which are perceived as comparable weak policy instruments in the CSR policy debate). These findings show that the limited budgets and power of Horizon 2020 projects do not allow the establishment of strong incentive mechanisms themselves. Therefore, the projects tried to link RRI to well-established policies (such as smart specialization strategies) or institutions (such as standard organizations). Another factor that hampers establishing effective incentive mechanisms for RRI is the limited time and lack of collaboration among EU projects, which could only be overcome by better coordination and a well-designed sequence of projects. This would also consider that changing discursive frames and institutionalizing RRI takes more than just a few years.

The typology in Figure 2 synthesizes the results of our analyses and interpretations on an abstract level, goes beyond the individual case studies, and thus provides a data structure that refines the picture of RRI in the business sector and highlights substantial variations depending on the project’s level (i.e. regional, corporate, or process). The terms used in this figure represent second order concepts and aggregated dimensions (theory-centered) stemming from interpretative work following the Gioia Methodology.

Figure 2. Typology for institutionalizing RRI in the business sector.

By providing comprehensive insights and deriving a typology, we sought to generate relevant concepts that go beyond existing literature. While previous studies criticized that RRI-related concepts, tools, and methodologies are not aligned with current industrial practices (see Dreyer et al. (2017)), our study highlights the roles of discursive frames and how they can be used to influence perceptions and decisions of companies, especially when societal and corporate goals conflict. While Lubberink et al. (2017) highlight that RRI is shaped by different contexts but did not provide any insights on what these different contexts are and how they lead to heterogeneous practices, our typology provides additional insights by differentiating between regional-, corporate-, and process-level approaches to RRI. Our typology takes up the call of Scherer and Voegtlin (2020) and (Gurzawska, Mäkinen, and Brey 2017) for effective incentives for RRI and provides insights into what such mechanisms would look like and how they are interrelated with discursive frames and potentially conflicting interests. In addition, our typology integrates these three aspects – frames, interests, and incentive mechanisms – which were considered in previous studies separately (Brand and Blok 2019; Dreyer et al. 2017; Owen et al. 2021) and provides evidence that they are interlinked and thus should be considered comprehensively rather than independently. Our typology thereby contributes to the debate on institutionalizing RRI and adds a complementary view to the work of Owen et al. (2021) and Steen, Nauta, and Gaasterland (2018) by providing insights into the specific contexts and dynamics of the business sector.

For the scholarly debate on institutionalizing RRI, our findings indicate that discursive frames are not homogenous in the business sector, but are influenced by context (regional-, corporate- and process-related), which shapes the interplay of ideas, discourse and institutions. Discursive frames can indicate potentially conflicting interests (such as those we have found at corporate level, where societal needs are often perceived as trade-offs to competitiveness, growth and profits), hide them (as our results show at process level), or provide an overarching narrative (as at regional level where the Quadruple Helix highlights the shared interest of stakeholders involved). Integrating RRI into existing frames can be interpreted as attempts at re-framing, adaptation or adoption of frames (Goffman 1974; Werner and Cornelissen 2014) and create new hybrid frames in order to stimulate organizational change. Such processes have been described in discursive institutionalism (Schmidt 2008; Werner and Cornelissen 2014), but not yet considered in research on institutionalizing RRI.

Future attempts might make use of frame blending (Werner and Cornelissen 2014), a strategy derived from discursive institutionalism whose use is opportune when seemingly competing sets of cultural beliefs exist. This is of particular interest at corporate level, where the most common RRI-related frames (the RRI Keys and the ARRI framework) have not so far been linked to the dominant logic of profits and growth.

While our study links aspects of discursive institutionalism with the practical considerations of incentive mechanisms in RRI for the first time, it only provides a starting point for the development of RRI-related policy instruments that consider specific contexts and framings. Another interesting avenue for future research lies in the interfaces and interconnections between the three different types of projects we have identified: innovation processes are often implemented within companies; companies are located in regions. Therefore, our study could provide a starting point for more in-depth research on the interplay of frames and a multi-level perspective on the discursive institutionalization of RRI.

Conclusions

Our paper provides both a typology of discursive frames, potentially conflicting interests, and incentive mechanisms; as well as regional-, corporate- and process-related contextual factors. The empirically based findings highlight that institutionalizing RRI in the business sector shows substantial variation depending on the level on which the business sector is approached and reveals that frames, interests, and incentive mechanisms are interlinked phenomena that should be considered comprehensively in future research as well as in developing strategies for institutionalizing RRI. Based on these key findings and the rich picture this comparative case study provided, we can draw several conclusions for business, policy and future research.

Institutionalizing RRI in an individual company can lead to tensions if RRI is perceived as an additional burden that competitors do not have to consider, and therefore might have lower costs and shorter time-to-market. Reframing RRI as a business opportunity is therefore an important step, especially in areas in which business success depends on trust and legitimacy, and the social license to operate is currently challenged (e.g. in big data, AI, and surveillance technologies). Understanding RRI as a particular application of a company’s CSR commitments would foster institutional linkages between CSR units and innovation departments. This could lead to compatible discursive frames, derived strategies and actions, and provide benefits for both. While CSR units would become more innovative and gain influence on decisions with a high impact on society and the future of a company, innovation departments could learn how to better deal with the diversity and ambiguity of societal needs and concerns.

For future research policies and programs such as Horizon Europe, our study showed that although individual projects put high efforts into developing tools and strategies, they often lacked outreach, and could not show substantial impacts on institutionalizing RRI in the business sector. Limited financial and time resources and a lack of integration of results into a bigger picture led to diverse and partly competing framings that have not managed to establish RRI as a dominant discursive frame in the business sector to date. Our typology highlights the importance of a more collaborative and coherent approach covering different contexts (i.e. the regional, corporate, and process levels) while considering frames, interests, and incentives as interrelated aspects of institutionalizing RRI. This integrated and nuanced view on institutionalization is of relevance for other areas of society-business-relations, such as Circular Economy, Supply Chain Responsibility and Responsible Digitalization. In all of these areas, future policies and programs should ensure more continuity of activities, better integration of results, consider discursive frames and conflicting interests, and utilize strong incentive mechanisms.

For future policies seeking to disseminate and institutionalize RRI in the business sector, our study provides conceptual foundations on incentive mechanism derived from CSR. By considering aspects of discursive institutionalism, an explanation is provided on why disseminating RRI-related case studies, tools and roadmaps has not led to substantial organizational change in the business sector. From a discursive institutionalism perspective, such change would result from new ideas and the related discursive process, thus establishing new frames which would then influence the construction of meaning and lead to change. In this discursive process, the establishment of new frameworks is more challenging and substantially more time-consuming than adapting existing frames. Most RRI projects in our sample (and probably beyond) have tried to establish the RRI Keys or the ARRI framework as new frames, instead of taking up or modifying the dominant frames of the business sector. Our examples at regional level (quadruple helix) and project level (co-creation) show the high potential of utilizing, extending and adapting existing frames – a strategy called ‘frame blending’ in discursive institutionalism – which could be further developed through future research.

Future research could provide more in-depth analyses of the three aspects of our typology. Regarding discursive frames, future research could provide additional empirical insights into usage, creative extension, and combination, and how they influence and legitimize corporate decisions and actions. While discursive frames that are highly compatible with the principles of RRI already exist on the regional level (Quadruple Helix) and process level (co-creation), there is currently no such frame available on the corporate level. A new narrative that translates RRI into business opportunities, monetizes the costs of non-action and thus better links up to the logic, terms, and tools that shape business practices, is needed. Regarding conflicting interests, a systemic, theory-based, and nuanced view of synergies and trade-offs would provide a detailed picture of the business case of RRI. Future research could go beyond the scope of our study, include corporate data, and carry out interviews with corporate managers to provide insights into their perceptions of the synergies and conflicts of societal needs and corporate interests. Regarding incentive mechanisms, an evidence-based conceptual framework of RRI-related policy instruments could provide a more detailed view of policy options, consider the specificities of innovation in a globalized economy, and position RRI policies in the context of CSR and sustainability-related governance systems. All three avenues of future research would contribute to the scholarly debate about institutionalizing RRI in the business sector and put forward the scientific discourse on frames, interests, and incentives.

Acknowledgments

We would like to thank our interview partners for the detailed insights into their projects, framings and perceptions, the reviewers and the editor of this paper for their productive feedback, and Rodrigo Lozano for his valuable suggestions.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Eurostat (2023), R&D expenditure (accessed on 11 April 2023).

2 Corporate Social Responsibility is a management concept whereby companies integrate social and environmental concerns in their business operations and interactions with their stakeholders. While in 2001 the European Commission defined CSR as ‘a concept whereby companies integrate social and environmental concerns in their business operations and in their interaction with their stakeholders on a voluntary basis’ (European Commission 2001, 4), a new definition was put forward in 2011 defining CSR as ‘the responsibility of enterprises for their impacts on society’ (European Commission 2011, 6).

3 Considering that our object of research is EU-funded pioneer projects, we use the term RRI in this study, although we explore different discursive frames that might originate in the discourse on Responsible Innovation as well.

4 e.g. von Schomberg (2013) defines RRI as ‘a transparent, interactive process by which societal actors and innovators become mutually responsive to each other with a view to the (ethical) acceptability, sustainability and societal desirability of the innovation process and its marketable products.’

5 Corporate Social Responsibility

6 RRI-related frameworks mostly served as science- or policy-based interpretations, operationalizations, and definitions of RRI. If they are commonly used and thus institutionalized, they become discursive frames that shape and direct audiences’ perceptions, decisions, understanding, and interpretation of RRI.

7 Randles et al. (2016) proposes six narratives for RRI based on theoretical considerations and literature analysis but did not carry out any empirical analysis of the frames used in practice.

8 The projects TetRRIS, RIPEET, RRI-LEADERS, WBC-RRI.NET, RRIstart, MOSAIC, and Critical Making

9 The projects TeRRItoria, TeRRIfica, CHERRIES, and TRANSFORM. The inclusion of the three projects SeeRRI, DigiTeRRI, and RRI2SCALE was based on random selection.

10 The European Regional Development Fund and the European Social Fund define less developed regions with a GDP per inhabitant less than 75% of the EU average, transition regions between 75% and 90%, and more developed regions with more than 90% of the EU average (https://ec.europa.eu/regional_policy/policy/how/is-my-region-covered_en).

11 CSAs in Horizon 2020 do not focus on research and innovation but on accompanying measures such as standardization, dissemination, awareness-raising, networking, policy dialogues, and mutual learning.

12 REC in Horizon 2020 stands for Research Organizations, which are not universities nor corporations.

13 The Quadruple Helix model can be interpreted as a frame of collaborative innovation involving research, business, policy, and citizens (Carayannis and Campbell 2010). Cunningham, Menter, and O’Kane (2018) focused on boundary-spanning activities between these four groups, and mentioned different value motives, but did do not shed further light on conflicting interests.

14 This is most likely caused by the H2020 topic which states ‘Local and regional authorities should be active partners of the consortia […], alongside organizations representing the other parts of the quadruple helix’.

15 HES in Horizon 2020 stands for ‘Higher or Secondary Education Establishments’

16 RIAs aim to establish new knowledge and/or to explore the feasibility of a new or improved technology, product, process, service or solution and may include basic and applied research, technology development and integration, testing and validation.