Involvement of multinational corporations in social innovation: Exploring an emerging phenomenon

While studies on social innovation (SI) have expanded significantly, our understanding of multinational corporations (MNCs) in the “ social ” innovation arena remains in its infancy. Through a systematic literature review, we consolidate and examine literature linking the MNC with the ongoing discussion of SI based activities. Based on a qualitative content analysis of 60 articles, this review explores how the MNC is portrayed with regard to its involvement in SI. The analysis identifies a fragmented view of MNCs involvement in SI and reveals a variety of theoretical approaches and conceptualisations in prior research. Our review presents a framework of ‘MNC involvement in SI ’ encompassing: differentiated conceptualisations, a dual value approach, five generic MNC roles, proactive and/or responsive motives, and specific barriers connected to the involvement in SI activities. From this, we suggest a number of implications for theory and practice as well as some directions for future research.


Introduction
International organizations such as the EU and the UN (European Union, 2010;United Nations, 2015), public opinion, and scholars alike have increasingly put multinational companies (MNCs) high on the agenda to become engaged as societally oriented problem solvers.MNCs are expected to leverage their extensive financial resources, management expertise, and global supply chains to find solutions to deep-rooted societal issues (Sinkovics et al., 2014;Varadarajan & Kaul, 2018).In general, the academic literature on social innovation (SI) depicts business organisations as capable of becoming socially engaged actors.Although the pressure to be involved in SI is becoming apparent to MNCs, very little is known about their participation (Mirvis et al., 2016;Tabares, 2020).A compilation of existing research would make an important contribution to this novel field.
Over the past two centuries, the concept of SI has been met with scepticism, and attitudes to SI activities have gone from mockery and violent opposition to, and increasing acceptance (Mulgan, 2007).For some, SI is the study of non-profit organisations; for others, it can encompass almost anything, from new types of democracy to the design of products for poor consumers.At its simplest, SI implies new ideas that work to address pressing unmet needs (Mulgan et al., 2007).SI can take the form of specific ideas, actions, frames, models, systems, processes, services, rules, regulations, and new organizational structures (Nicholls et al., 2015), as long as they are "social in their means and ends" (Young Foundation, 2012).Recently, there has been a call for business scholars to address SI (Foroudi et al., 2020), in particular, MNC scholars (Cajaiba-Santana, 2014).However, much of the research that addresses the way large corporations are approaching SI seems fragmented and disconnected from the more scientific and academic research on SI (Mirvis, et al., 2016).In addition, according to scholars such as Cajaiba-Santana (2014) and Foroudi et al. (2020), the issue of innovation within MNCs has primarily been approached from a perspective such as business innovation, technological innovation, or strategic change.Research on SI has found its intellectual home mostly within the field of social entrepreneurship rather than in business and management research.From the SI literature it is evident that for-profit enterprises have become important actors in the solution of social problems (Pol & Ville, 2009) and there has been a huge increase in research "corporations" trying to incorporate SI into their innovation strategies, especially within the past ten years (Dionisio & de Vargas, 2020;Herrera, 2015;Mirvis et al., 2016;Mirvis & Googins, 2018b;Tabares, 2020).However, the "corporate" actors in this research are for-profit organisations that include small-and medium-sized firms, start-ups, spin-offs, corporate universities, social businesses, and different hybrid organisations, while larger companies, in particular, MNCs are used as empirical illustrations at the most.
To fill this gap in research, this paper presents a systematic review of literature that explicitly addresses the MNC as a participant in SI activities, with the aim of mapping out how the MNCs have been portrayed in this literature and addressing the following research question: "What are the main characteristics of MNC involvement in SI, as presented in the current literature?".This paper makes several contributions.First, it adds to the literature on SI by explicitly linking the MNC to the concept of SI.Second, given the increased attention being paid to MNCs' engagement in SI activities at policy levels as well as amongst academic scholars, this paper can serve as a starting point for research on how the MNC can leverage its unique context to address societal challenges.Third, this study is in line with the suggestion of scholars such as Doh (2019, p. 76) that it is time for scholars within the field of international business (IB) and management to reorient themselves "toward global grand challenges" and towards understanding non-market activities of MNCs.

Theoretical grounding
Social innovation has emerged as a normative concept (Moulaert, 2016;Mulgan, 2006;Mumford, 2002;Phills et al., 2008), but the boundaries of "social innovation" as a concept are still rather broad and fluid (Ayob et al., 2016).Unlike business innovation, SI is driven by the pursuit of social benefits, not profit.As such, SI is characterized by the creation of new ideas and new solutions to societal and environmental problems (Phillips et al., 2015).The term "social innovation" was introduced in the work of Drucker (1984) with a call for using management practices in non-profit organizations to increase the efficiency and effectiveness of producing social good.The definition of SI has revolved around innovation and matters such as the 'societal impact' of innovation (Howaldt & Schwarz, 2010;Pol & Ville, 2009) and has, by some scholars, been discussed in terms of its 'purpose' of meeting social goals and its usefulness to society (Bhatt & Altinay, 2013;Froud et al., 2010).Research has also discussed SI in terms of its 'content' where SI is about 'novel' and 'new' solutions to societal issues to accomplish systemic change (Mulgan, 2006).According to Murray et al. (2010) and the Young Foundation (2012), SI entails the development of new ideas and solutions to societal problems, and "the value accrues primarily to society as a whole, rather than to private individuals" (Phills et al., 2008, p. 39).Social innovation is not only about developing a new kind of product or solution to a societal problem but also refers to the process that leads to it (Ayob et al., 2016).Generally, it is argued that SI processes take place in the space between the different sectors of non-profit firms, government and commercial organisations (Salim Saji & Ellingstad, 2016), where the co-creation of value (creation of e.g.new products and services) by multiple cross-sector actors is an important feature.SI "often involves multi-agent and multilateral networks of actors which develop and sustain new and innovative services and improve the quality of the lives of people" (Babu et al., 2020, p. 16).Gough et al. (2017) point out that systematic literature reviews are grounded in a set of common principles guided by a research question and a systematic approach.Literature reviews are conducted for different purposes and can take different forms (Gaur & Kumar, 2018;Torraco, 2005;2016;Tranfield et al., 2003).In particular, in a research area that covers a range of different fields of literature, "…a systematic review can help to consolidate the topic" (Kraus et al., 2020(Kraus et al., , p. 1028)).Furthermore, considering the growth of literature, "the role literature review papers play in advancing a discipline has become even more critical" (Kraus, et al., 2021, p. 3).Systematic literature reviews require a narrow scope of the study, making them well suited to confined or niche research areas, often incorporating a smaller number of papers (Donthu et al., 2021).This review attempts, as Torraco (2016, p. 410) suggests, to provide "a preliminary conceptualisation of the topic rather than a re-conceptualisation of existing models".This is done by the use of a thematic structure for organizing the analysis.According to Gaur et al. (2018, p. 281) "providing conceptual structuring of the topic provides the literature review with coherence and clarity about what is being reviewed and how the main concepts of the topic come together as a unified idea".

Data collection
The first step was to develop a database of documents for systematic review.We identified three search databases, the ISI Web of Science (WoS), Scopus, and EBSCO; these are widely used business resources that contain reliable indicators within scientific debate (Fink, 2010).The Social Sciences Citation Index (SSCI) was used to search the WoS and retrieve citation data to compile our database.Thomson Reuters, the publisher of SSCI, states that "every journal included in Social Sciences Citation Index has met the high standards of an objective evaluation process" (Van der Have & Rubalcaba, 2016, p. 1925).The SSCI has the advantage that it has filtered out bibliographic data from less reputable journals that have a questionable impact on the scientific discourse, thus preserving data integrity.Although the SSCI has broad coverage of the literature, we chose to perform an additional, identical query in Scopus, which is Elsevier's abstract and citation database of peer-reviewed literature in the fields of science, technology, medicine, social sciences, and arts and humanities.We also chose to use EBSCO, which provides an eBook collection, and the EconLit and Business Source Complete databases of economic and business literature.
To account for the various concepts and/or synonyms used in the literature that could indicate a link to the terms "SI" and "MNC", the following keywords and keywords alternatives (with truncation) were used: ("social innovation*" OR "social innovation* in business*" OR "business* in social innovation*" OR "corporate social innovation*" OR "CSI" OR "doing good doing well innovation*" OR "social value creation*" OR "creating shared value*" OR "CSV*") AND ("corporation*" OR "enterprise*" OR "international corporation*" OR "multinational corporation*" OR "MNC*" OR "multinational enterprise*" OR "MNE*").The search was conducted in September 2020 and the only refinement was the limitation to English-written documents and the exclusion of conference papers and proceedings papers.The initial search results showed 1,025 documents from WoS, 2,398 from SCOPUS, and 2,886 from EBSCO.
The search results from WoS, SCOPUS, and EBSCO were compiled into one database and filtered to remove duplicates.Further, publications containing unrelated subject matters such as spectroscopy isolation (SI) from the Journal of Solution Chemistry and CSI/FBI computer crime and security surveys from the Computer Security Journal, were picked up by the search due to similar abbreviations and acronyms.These publications were manually removed because of common acronyms and abbreviations from the natural sciences and wireless communications field.Since the search included all fields, significant numbers of publications were removed because keywords appeared in the title of the journal, for example, but not in the publication itself.This process reduced the search results to 397 publications.The final step was to read through the main body of text to determine if a publication was explicitly connected to MNCs (or their synonyms).This procedure allowed us to empirically identify articles that covered MNCs as part of the main aim of the study, as a phenomenon or object of study, or approached the topic theoretically.Out of these 397, we identified 58 documents that linked SI activities to MNCs.We added additional control by cross-referencing the reference list of the 58 documents to ensure no relevant study was missed, a process that added two additional articles.We checked the reference lists of those two additional articles, but no new relevant article was found.This interpretive validation thus resulted in 60 documents, a sample that represents the purpose of this review and reflects the criteria for the given research question (Gaur & Kumar, 2018).

Data coding and analysis
An analytical review scheme is necessary for systematically evaluating a given body of literature (Ginsberg & Venkatraman, 1985).This study followed the recommendation of Tranfield et al. (2003) to use data extraction during the analysis stage.Accordingly, a database was created with the following bibliometric categories: authors, year of publication, title, journal, paper type (e.g., empirical, conceptual), theoretical basis, research method (e.g., quantitative, qualitative), SI approach, and "features of MNC involvement".
The analysis of the data in the 60 final articles was conducted in two parts.First, a descriptive analysis of key journals, types of articles, number of publications, research method as well as an analysis of the theoretical foundations of the papers in the review set.Second, we performed a thematic analysis to assess the key features of MNC involvement in SI.This analysis resulted in the main themes of conceptual approach, value orientation, the role and motivations of MNC involvement, and the related barriers (see section five).The thematic analysis was conducted by a qualitative interpretative content analysis, which was chosen because it provides a broad scope of application and flexibility (Gaur et al., 2018, p. 281).It entailed coding the textual data in all 60 articles.In content analysis, textual data can be coded at the level of the word, phrase, sentence, paragraph, or theme.We used the human-scored systems approach in our literature review, a common practice in qualitative content analysis (Drisko & Maschi, 2016;Gaur et al., 2018;Short & Palmer, 2008).It entails manual coding by choosing the coding units for classification, developing the coding scheme, and providing a description of each category.
The full text of the 60 papers was read to identify the main features of MNCs' involvement in SI.To ensure accuracy and reliability, the papers were examined independently by each of the co-authors, each of whom codified the data manually in a thematic scheme.The results were discussed and compared between the authors.Subsequently, the scheme was synthesized and organized into themes which formed the basis for further in-depth content analysis.In this, we were following the coding method of Gioia et al. (2013) who proposed the development of first-and second-order coding.

Descriptive findings
The 60 papers in the review set were published between 1987 and 2020 (Fig. 1).Approximately 60% were published during the last seven years.The papers were from 44 different publications, with 38 journals having one publication each.The three main publications were: Journal of Business Ethics, Critical Perspectives on International Business, and Journal of Business Research (Table 1).Within the 60 documents, 51% were empirical papers based on qualitative case studies, narratives, or anecdotal case evidence.Conceptual papers accounted for 29% and showed a particular emphasis on theory building.Review papers made up 16%.Of the empirical papers, only two (3%) used quantitative methods.For 40 of the 60 publications, the research focus was explicitly on MNCs, with the remaining 20 featuring, for instance, MNCs as a part of their data sample or as illustrative examples.Although not clearly specified in all the papers, with respect to the geographic location of the SI activities, there were 32 with projects/programs that had a global outreach and 13 that dealt with activities undertaken local-for-local, as in the US for the US.Twenty-three papers had a focus on developing markets and 11 presented SI activities in developed countries.

Theoretical perspectives of the papers
Argumentations and references to existing theories in the papers were considered using the thematic variable 'theoretical foundation'.Most studies (85%) used a multi-theory approach as a guiding framework, with only about 13 % adopting a single-theory approach.Notably, 18% of the papers did not refer to any specific theory or framework at all.With respect to the single theories used, three papers referred to institutional theory.Institutional theory (cf.DiMaggio & Powell, 1983) is commonly used in the context of understanding the impact of internal and external influences on organisations; it posits that organisational behaviour is highly influenced by the need for organisational legitimacy.Two papers used stakeholder theory, which posits that a business can be understood as a set of relationships among groups that have a stake in the activities that make up the business (e.g., Freeman, 1984).The behavioural theory of social entrepreneurship was explicitly used in two of the papers.This theory concerns the use of managerial skills to address social problems as well as business challenges.Although social entrepreneurship theory is rooted in research in entrepreneurship C. Holmström Lind et al. (Schumpeter, 1943), it is embedded in the social sector and has often been used to explain the occurrence of, for example, social entrepreneurs (Nicholls, 2008).One article adopted the upper echelon theory to investigate the strategic choices of CEOs with regard to social innovation.This theory posits that the path an organization takes can be seen as a reflection of the top managers' values and choices.In the remaining papers, a multi-framework approach was taken in which two or more theories or approaches were used with a combination of different conceptualisations (see section 5 below).The corporate social responsibility (CSR) approach was used in six of the papers in reference to Carroll's (1979) proposed definition of CSR as consisting of four corporate responsibilities to society (economic, legal, ethical, and discretionary).Other behavioural and/or relational approaches included, for instance, the network approach (5 papers), the cross-sector partnership approach (7 papers), and social alliance theory (1 paper).An innovation-based approach appeared by reference to the theory of (social inclusive) open innovation (1 paper).Finally, a number of different strategic management theories and approaches were identified, including the knowledge-based view (2 papers), and conceptualisations such as the "bottom of the pyramid" (3 papers), and the paradigm of strategic fit (3 papers).From an organisational perspective, the phenomenon of 'hybrids' was mentioned (1 paper) in connection with the actions and activities of socially responsible corporations.

Conceptualisation of SI in a corporate context
Not surprisingly, given the above-mentioned fragmented theoretical landscape, a variety of interrelated but not necessarily interchangeable conceptualisations framed the research covering MNC involvement in SI activities.The adoption of SI at the corporate level was conceptualised as corporate social innovation (CSI), creating shared value (CSV), social value creation (SVC), corporate social responsibility (CSR), and corporate social entrepreneurship (CSE).Hybrid concepts were also identified, such as corporate social investment, social impact investment, doing-well-by-doing-good innovation (DWDG), sustainable-led innovation, and corporate community involvement (CCI) (see Table 2).Fifteen of the articles used the concept of SI grounded in definitions provided by the SI literature (cf.Phills et al., 2008;Pol & Ville, 2009) and/or the social entrepreneurship literature (cf.Sullivan Mort et al., 2003).Its main feature is the view of SI as an innovation activity that is not driven by the pursuit of profit, which stands in contrast to the concept of business innovation.In these articles the participation of MNCs in SI activities is referred to as "social innovation performed by MNCs" (Onsongo, 2019) or "social innovation adopted by MNCs" (Chen & Karwan, 2008), to mention just two.
The concept CSI (17 papers) was referred to as a "strategic tool" (Esen & Maden-Eiyusta, 2019;Googins, 2013;Kanter, 1999), a "strategic investment" (Dionisio & de Vargas, 2020;Tabares, 2020) or "an innovation in the corporate sphere" (Mirvis & Googins, 2018b, p. 21) that corporations use to align economic goals with social value.When SI takes place at the corporate level of an MNC, SI is referred to in terms of the concept of CSI, which encompasses a broad range of activities including innovation, social initiatives for changing corporate identities, and strategies to create value for both the corporation and society (Herrera, 2015;Mirvis et al., 2016;Tabares, 2020).The concept of CSI is used to capture corporations' attempts to integrate "social" innovation activities more directly into their business strategies (Herrera, 2015;2016;Jayakumar, 2017, Mirvis & Googins, 2018a, 2018b).CSI is the most advanced and the most common academic concept concerning social innovation at the corporate level (Canestrino et al., 2015;Dionisio & de Vargas, 2020;Herrera, 2015;Mirvis et al., 2016).
Another frequently used concept is CSR (11 articles), understood as "activities whereby companies integrate social and environmental concerns in their business operations and interactions with their stakeholders on a voluntary basis" (Da Silva, 2017, p.61).The CSR concept is concerned not so much with the social obligations of firms, but rather with the idea that successful CSR strategies align the social and environmental activities of corporations with their business purpose and value (Camilieri, 2016), which is in line with the concept of CSI.The terms "strategic CSR" and/or "political CSR" therefore often occur in these articles.The concept of shared value is used in as many as 21 of the papers and refers to the work Porter and Kramer (2011, p. 67), emphasizing the importance of MNCs engaging in "policies and operating practices that enhance the competitiveness of a company while simultaneously advancing the social and economic conditions in the communities in which it operates".The concept of CSV is used similarly and denotes a focus on the creation of commercial and social value (Dionisio & de Vargas, 2020).A few articles used the concepts of SVC (9 articles) and/or CSE (7 articles).However, they appeared mostly in conjunction with other conceptual approaches.SVC broadly denotes behaviour that occurs "within or across the non-profit, business, or government sectors'' (Sinkovics et al., 2014, p. 693).This concept seems to be about exploring corporations' social innovation activities in, for example, BOP markets, where the emphasis is on the difficulty of creating and sustaining business in many developing markets without having a social value creation mindset (Sinkovics et al., 2014;Sinkovics et al., 2015).SVC emphasizes the creation of positive societal or environmental outcomes (Acs et al., 2013;Doh, 2019;Sinkovics et al. 2015) that are said to be created through MNCs via external partnerships such as with governments, nongovernmental organizations, or social entrepreneurs (Acs et al., 2013;Rygh, 2020).It is suggested that corporate social entrepreneurs are the ones who carry out SVC and/or direct actions toward SVC in MNCs (Feraru, 2018;Ghauri et al., 2014;Sinkovics & Archie-acheampong, 2020).The concept of CSE is used in two different contexts, first as a social and economic value creator within MNCs (Acs et al., 2013;Ghauri et al., 2014;Sinkovics & Archie-acheampong, 2020) and, second, as more than just a tool to achieve social and economic value but rather a new version of CSR as "the mixture of entrepreneurship and social value creation which, together with the focus on improvement of the relationship between the business environment and society" (Feraru, 2018, p. 202).

A dual value approach
A majority of the papers (85%) emphasised the importance of balancing business value and social value (Table 3).From this, we are able to argue that one feature of MNC involvement in SI entails a 'dual value approach', which implies an endeavour to simultaneously create social and economic value.This dual value orientation is referred to as a proactive approach to create, for example, "synergistic value" (Camilleri, 2016), "shared value" (Lashitew et al., 2020;Mirvis et al., 2016;Tabares, 2020;Varadarajan et al., 2018), "value creation and value capturing" (Esen & Maden-Eiyusta, 2019;Ghauri et al., 2014), "blended value" (Mirvis & Googins, 2018b), and activities and innovations that "deliver social benefits and business value" (Varadarajan, 2014).This dual value approach stands in contrast to business innovations in which the value primarily accrues to the benefit of the firm.An interesting notion is that with regard to this dual value orientation is that as many as 32 of the papers (53%) refer to the "shared value" concept of Porter and Kramer (2011) in their work (see Tables 4-6).

The contributor
In some of the articles, the MNCs are referred to as resourceful organizations (Ghauri et al., 2014), and the involvement in SI activities is considered an act of "deliberate resource allocation" of a corporation (Varadarajan et al., 2018).Being engaged in SI implies contributing a Note: A majority of articles use a combination of the above approaches.range of different resources varying from "financial assistance, managerial expertise, use of company infrastructure or the fulfilment of other needs expressed by the community directly or via a not-for-profit intermediary" (Cooke, 2010, p. 73).Some MNCs primarily act as financial funders or investors (Le Ber & Branzei, 2010a;Ortega et al., 2014;Jo et al., 2018), which entails engaging in money-value-resource transactions (Ortega et al., 2014), and as key project investors (Chen et al., 2008) with "episodic engagement" (Mirvis & Googins, 2018b).MNCs can also act as 'knowledge contributors', being brought into a social innovation project because of their specific competencies (Raufflet, 2009;Salim Saji & Ellingstad, 2016), such as specific technical or managerial skills (Altuna et al., 2015;Herrera, 2015Herrera, , 2016;;Gupta et al., 2017).In this regard, MNCs "are chosen on the basis of the specific competencies they can contribute to the project" (Altuna et al., 2015, p. 273), sharing their knowledge and competencies within specific areas of expertise with other actors in a project (Bowe & Van der Horst, 2015).

The mentor
Some MNCs are involved in SI activities mainly for their willingness and ability to act as educators or mentors (Camilleri, 2016;Chen et al., 2020;Gentile, 2001;Jo et al., 2018;Mirvis & Googins, 2018b;Mirvis et al., 2016).Such mentorship entails, for example, educating and training social entrepreneurs to enable them to launch their ventures (Mirvis & Googins, 2018a).By engaging in training and education activities at a community level, corporations are able to enhance necessary skills and competencies among indigenous organisations (Jo et al., 2018).According to scholars such as Selsky and Parker (2010), SI can be viewed as solving meta-problems, which require a complex set of skills and competencies.MNCs that adopt the role of mentors are able not only to increase the level of awareness among corporate employees and managers about creating sustainable social impact (Gentile, 2001, Mirvis & Googins, 2018b) but also to improve the skills and capabilities of social entrepreneurs, rural communities or other stakeholders in areas in which they have the expertise and unique know-how (Mirvis et al., 2016).

The innovator
In the innovator role, MNCs are involved based on their capacity to innovate (Canestrino, et al., 2015;Herrera, 2015;Kanter, 1999;Ljung & Bengtson, 2012;Varadarajan et al., 2018) and their ability to act as an innovator for solving social problems (Dionisio & de Vargas, 2020;Mirvis et al., 2016).The engagement of an MNC in SI is not charity, "it is R&D -a strategic business investment" (Kanter, 1999, p. 124).Being engaged in SI implies an obligation "to engage in processes of innovation where the ideas, products, and processes have to be adapted to significantly generate benefits to the business, the community stakeholders, and the wider society" (Da Silva, 2017, p. 65).The view of the MNC as an 'innovator' seems to be strongly associated with the concept of "corporate social innovation" (Canestrino et al., 2015;Dionisio & de Vargas, 2020;Esen & Maden-Eiyusta, 2019;Herrera, 2015;Kanter, 1999).Some companies are said to have embraced social innovations into their strategy" (Dionisio & de Vargas, 2020) and recognized the important role corporate employees play as 'social innovators' for the co-creation of innovations (Mirvis et al., 2016).This line of research emphasises that MNCs engaging in SI represent a new paradigm for innovation, one that encompasses innovation through co-creation (Esen & Maden-Eiyusta, 2019;Herrera, 2016;Lashitew et al. 2020).As stated by Herrera (2015Herrera ( , p. 1471)), "social innovation is more likely to increase competitive advantage when both operational and organizational mechanisms allow co-creation".

The political agent
Some MNCs adopt a political role in the sense that they become "active political actors and agents" (Barnett, 2020, p. 466).Muthuri et al. (2012, p. 5) state that corporations "sometimes act as "governments" or as "citizens" in the administration of citizenship rights for individuals in the community".By being a socially-oriented business and through advocacy, MNCs are able to shape public policy in a way that strengthens society (Barnett, 2020) and anticipates future market opportunities (Sinkovics et al., 2014).Holmström Lind et al. ( 2018) propose that MNC involvement in SI projects may be an important means of gaining legitimacy among important political actors in the home as well as host country environment.

The integrated partner
In some of the papers, the MNC is illustrated as, for example, an alliance partner, a cross-sector partner, a cross-sector collaborator, or a strategic partner (Kolk & Lenfant, 2016;Mirvis et al., 2016;Salim Saji et al., 2016).Such partnership roles entail more fully-fledged responsibilities including, for instance, actual research or information seeking on the part of the MNC into the social problem (Salim Saji et al., 2016).It can also entail knowledge-sharing and a strong commitment to and a greater engagement in a SI project or activity (Allee & Taug, 2006;Carberry et al., 2019) along with "co-creation" to ensure joint learning (Mirvis & Googins, 2018b).This role also encompasses a bridging role, which implies that an MNC acts as a bridge and a boundary spanner between governments, NGOs, and the communities in which they are embedded (Manning & Roessler, 2014;Muthuri et al., 2012).For corporations, close cross-partnerships with NGOs and social enterprises may be the only way to give the company a broader license to operate in society (Lashitew et al., 2020).Being a close partner can be crucial for success when entering arenas, such as the SI domain, where problems are very complex and "no party has full knowledge of the situation and solutions are emergent" (Mirvis & Googins, 2018b, p. 19).A key feature of this close integrated cross-sector partnership is the provision of mentoring, business guidance, and technical assistance to the partner recipients (Mirvis et al. 2016).

Responsive motives
The responsive motives of the MNCs have been categorized along the lines of institutional and stakeholder adaptation.The responsive motive mentioned most often is the MNCs' institutional adaptation to their surrounding institutional setting.These institutions include not only formal ones such as regulatory bodies but also more often informal institutions such as the prevailing values and norms in society, which the firms need to abide by to be considered legitimate (Barnett, 2020;Canestrino et al., 2015;Cooke, 2010;Esen & Maden-Eiyusta, 2019;Ljung & Bengtson, 2012;Maak et al., 2016;Manning et al., 2014).Society expects MNCs to include the societal context as part of their responsibility because of the power held by these large companies (Gentile, 2001) and the opportunity afforded them both to influence policy and to "ensure organizational legitimacy" by engaging in SI activities (Maak et al. 2016, p. 466).The study by Lashitew et al., (2020, p. 432) emphasises the need for "external legitimation" as a driver for corporations' involvement in SI and views SI activities as a "legitimation activity".As mentioned by Barnett (2020), companies are expected to do good for society, and responding to these demands puts pressure on them to make significant changes.The issue of stakeholder adaptation, in which companies comply with demands from stakeholders, was also central among the motives for involvement in SI-related projects (Altuna et al., 2015;Chen et al., 2020;Da Silva, 2017;Esen & Maden-Eiyusta, 2019;Mirvis et al., 2016).While various stakeholders were mentioned as putting pressure on companies (Mirvis et al., 2016;Panchenko & Kiriakova, 2015), a specific group received special attention, namely activists from civil society (Carberry et al., 2019; Sinkovics & Archie-acheampong, 2020).

Proactive motives
One proactive motive that was identified relates to the MNCs' strategic orientation, which includes issues related to the market, costs risk, branding, HR, stakeholders, and policy.The need for strategic fit (Mirvis & Googins, 2018b), that is for SI activities to be aligned with the firm's strategy at large, is central.One common strategic orientation is the 'market orientation' (Ghauri et al., 2014;Herrera, 2016;Varadarajan et al., 2018) The main logic behind the market orientation concerns the MNCs' expected benefits from their market presence by performing SI activities.An illustrative example is given by Acs et al. (2013, p. 794) with SI endeavour of Microsoft, where the engagement was "organized with the principal goal of enabling commercial growth and market presence".While some scholars define market orientation motives in terms of profit gains, several scholars underscore the importance of being a first mover and market leader in the social arena (Harrisson et al., 2012;Rygh, 2020).A 'reputational orientation' (Cooke 2010;Lashitew et al., 2020;Rath et al., 2018;Selsky et al., 2010) is also raised as a strong motive, with the need to strengthen corporate reputation and branding emphasized in some papers as well as the need to attract and retain employees (Abboubi & Nicolopoulou, 2012;Chen et al., 2008;Kanter, 1999).A 'cost-benefit orientation' also seems to underly participation in social innovation activities (Bowe & Van der Horst, 2015;Dionisio & de Vargas, 2020;Jayakumar, 2017;Muthuri et al., 2012;Selsky et al., 2010).The rationale is the idea of a win-win situation for society and business from the SI-related endeavours, described broadly as cost savings and profit enhancement (Esen & Maden-Eiyusta, 2019)."Such a strategic approach can result in new business models and cross-sector collaborations that will inevitably lead to operational efficiencies, cost savings and significant improvements to the firms' bottom lines" (Camilleri, 2016, p. 10).
The second main proactive motive found in this review is a learning orientation with the intention to learn about this 'new' market (Allee and Taug, 2006;Gupta et al., 2017;Harrisson et al., 2012;Mirvis et al., 2016;Mirvis & Googins, 2018b).Mirvis and Googins (2018b, p. 7) argue that "as employees interact with and mentor social actors, they gain new insights into social issues and innovations, sharpen their coaching and project management skills, and develop new relationships that expand the 'social capital' of their companies".Hence this logic is based on the fact that MNCs as business organizations do not hold knowledge concerning societal issues, local needs, etc., and need to connect and learn from other sectors of society that are experts in these fields.Strongly connected to this is the notion of innovation as a motive to engage in SIrelated activities, with some articles noting that gaining new knowledge and insight is important for coming up with new ideas (Sinkovics et al., 2014;Varadarajan et al., 2018), and "…developing innovative products, services, and operations to deal with local social and environmental challenges" (Nguyen et al., 2019, p. 781).

The multiplicity of logics and value frames
SI as a "collaborative venture" (Voltan & De Fuentes, 2016) is faced with a high level of institutional complexity.Manning et al. (2014) refer to an "institutional distance" between collaborators from different sectors with different value systems that can act as a hurdle in recognizing opportunities for collaboration.Conflicting institutional logic, arising from the highly diverse logics surrounding and applied by organizations from different sectors, is a challenge to collaboration and mutual understanding (Altuna et al., 2015;Chen et al., 2008;Jo et al., 2018).Le Ber et al. (2010b) list differences, such as organizational identities, missions, structures, and patterns of activity, that exist between forprofit and non-profit actors as part of the institutional complexity in cross-sector relations.The authors explain this problem in terms of being embedded in distinct different value logics, which may cause what the authors refer to as misunderstandings, misallocations, and mismatches of power and mistrust.Hence, goal and value differences require greater "flexibility within, and strong relationships with outside actors to coordinate action to achieve systemic change" (Lashitew et al., 2020, p. 423).In line with this reasoning, Manning et al. (2014) underscore that different value systems are commonly the cause of the low level of initial trust in cross-sector collaborations.As formulated by Harrisson et al. (2012) and Muthuri et al. (2012), companies and associations do not have the same agendas or ideological foundations and are therefore driven by a set of opposing values and logic, which could lead to social projects characterized by competition and rivalry.

Lack of knowledge and capabilities
One apparent barrier emphasized by Manning et al. (2014) and Mirvis et al. (2016) is the lack of specific knowledge and understanding about the social subject per se.Knowledge about where the main (social) problem resides and what the local conditions are surrounding the specific issue of concern can create problems for MNC involvement in SI (Mirvis et al., 2016;Lashitew et al., 2020).In the article by Holmström Lind et al. ( 2018) it is proposed that MNCs are more likely to engage in SI if they can draw on existing knowledge sources and skills within the corporation.The authors also suggest that the MNCs' traditional focus on business-related issues implies that their expertise in societal development is not a natural part of their skillset.A further problem concerns the organizations' "knowledge about each other", or what Mirvis et al. (2016) refer to as social ties and "know-who".According to Ortega et al. (2014, p. 74), firms with a socially-focused business model are often used to "tackle problems with multiple dimensions of uncertainty simultaneously".Hence, engaging in non-market activities such as SI where inter-sector collaboration is common implies having to deal with the problem of different capability bases (Le Ber et al., 2010a).The lack of collaborative capabilities (Manning et al., 2014;Mirvis et al., 2016) and cultural capabilities (Ljung et al., 2012;Jo et al., 2018;Smith, 2016) are important sources of problems when entering the SI arena.It is argued that companies, especially MNCs, need to develop an understanding and a capability of how to work in cross-sector relationships (Herrera, 2015;Mirvis et al., 2016).Many corporations simply do not possess the appropriate knowledge about how to act in this social space and it is therefore suggested that corporations need to employ "changemakers" (Mirvis & Googins, 2018b, p. 21).

The trade-off between social and commercial value
A number of authors discuss the dilemma of balancing commercial goals and the creation of value for society.Companies wish to maximize their economic value, in comparison with what drives many non-profit organizations, NGOs, and governments, which are guided by a social value orientation (Acs et al., 2013).Hence, being involved in SI requires a trade-off between the creation of social value and commercial value (Canestrino et al., 2015;Chen, et al., 2020) or what some authors refer to as social value creation and value capture (Esen & Maden-Eiyusta, 2019;Ghauri et al., 2014), in which social value creation is measured at the societal level and value capture at the organisational level.The concept of value capture is referred to in the work of Santos (2012) and viewed as a value derived from an activity that "happens when the focal actor is able to appropriate a portion of the value created by the activity after accounting for the cost of resources that he/she mobilized" (Ghauri et al., 2014, p. 579).The challenge of this trade-off is explained by Smith (2016, p. 35) as executives believe that "social engagement should be separate from the business".To manage this trade-off, Mele et al. (2020) discuss moving beyond traditional management, by developing a new value logic for conducting business and by creating new rules and ways to measure social impact.Some authors discuss this issue as the ability to institutionalize SI by integrating the societal aspect of the innovation efforts into the corporate strategy (Herrera 2015;2016;Mele et al., 2020;Mirvis et al., 2016;Mirvis & Googins, 2018a;2018b;Tabares, 2020).

Power balance
The issue of power comes up in reference to the balance of power when different actors from different sectors are engaged in societal projects (Holmström Lind et al., 2018;Selsky et al., 2010).Chen et al. (2020, p. 249) address the issue of power struggles between actors, claiming that there is always a likelihood of power struggles between local actors and MNEs, which "may require adjustments from both sides to achieve the collective goal of success".Similarly, for Muthuri et al. (2012) the balance of power between communities and businesses is often one-sided and very often in favour of the business.According to Holmström Lind et al. ( 2018), the likelihood of MNC involvement in SI projects is related to the ability of the MNC to balance profit-non-profit goals in social innovation projects in line with its interests.However, in a situation of high corporate power, "local communities often seek distributional or procedural justice that can serve as a catalyst for company-community conflicts that range from minor disagreements to escalated or outright violent conflicts" (Muthuri et al., 2012, p. 10).

Findings and implications for theory and literature
The literature reviewed in this paper provides insight into the way the MNC has been portrayed with respect to involvement in SI activities.
From the thematic analysis, we identified five main characteristics of how MNC involvement in SI is portrayed in the literature.First of all, the review shows that the theoretical, as well as conceptual grounding of SI in a corporate context is rather fragmented.Many of the articles do not refer to any specific theory and in a majority of the articles, the theories are vaguely referenced, more or less as background to the study.These observations establish that the study of MNC involvement in SI appears to lack theoretical grounding.One example is the heavy use of the concept of CSI, which appears to be growing and becoming established as the 'norm' to explain MNCs engagement in "social" innovation activities (Dionisio & de Vargas, 2020;Herrera, 2015;Mirvis et al., 2016;Tabares, 2020).Scholars in this line of research anchor much of their reasoning on the rather fluid concept of "shared value" coined by Porter and Kramer (2011), a concern that resonates with the previously expressed critique of the shared value concept as lacking a deeper theoretical foundation (Dembek et al., 2016) and showing disregard for the broader stakeholder theory as well as emerging research on social innovation and hybrid organisations (Crane et al., 2014).We believe that our analysis adds insights into this debate and enriches the literature on SI in a corporate context by widening the understanding of this topic beyond concepts such as CSI and CSV and by recognizing the critical need to develop strong theoretical underpinnings to guide future research on this topic.We also believe that our analysis contributes to the MNC literature, emphasizing the need to link business to society (Doh, 2019) by integrating the social innovation literature into the research.
Second is the notion of a dual value logic.This strongly relates to recent research on organisations with a social purpose and the emerging issue of how firms are able to balance the "new dual institutional logic of social-economic value creation" (Weerawardena et al., 2021, p. 762) for sustainable competitive advantage.Third, in this review, we also identified some specific dimensions of 'MNC involvement in SI' including the MNC roles, motivations for Si involvement and specific barriers relating to SI involvement.We could identify five generic roles characterising MNC involvement in SI (the mentor, the contributor, the innovator, the political agent, and the integrated partner), with regard to the scope and breadth of MNC involvement.We also find that the involvement in SI is discussed and coincides with much of the previous research on motivations for participating in CSR, ranging from a stakeholder-driven to a strategic-driven intent.However, in addition to this the review reveals that there is also a strong learning-driven imperative behind this involvement.This implies a desire to learn about the inner workings of this "social" market.Lastly, our review identified that the involvement in SI activities seems to be strongly linked to specific barriers to C. Holmström Lind et al.
involvement relating to difficulties in combining multiple logics, values, and rationales from different actors across different contexts.For this reason, we believe that the involvement in SI by for-profit organisations in general, and MNCs in particular, as suggested by Dionisio & de Vargas (2020) is a question of how to align existing organisational identities and structures into the formation of new hybrid organizational.In Fig. 2 these main findings of the literature review are summarized and presented in a conceptual framework.

Managerial implications
This study offers many practical implications.First, managers should be aware of and even anticipate that problems will arise during the process of getting involved in SI activities.Although this review has shown proactive behaviour among MNCs concerning their engagement in SI processes and activities, many MNCs may experience disruption when it comes to organizational and managerial practices.Thus, managers should take a cautious and considered approach when engaging in SI given the many barriers that will probably be encountered along the way.Second, managers should be aware that non-business partners (i.e., governments, NGOs, and citizen groups) may operate with different working processes, objectives, rationalities, and organisational logic and values (cf.Sanzo-Pérez et al., 2021).Recognizing these differences acknowledged in this review may help minimize disruptions to crosssector collaborative SI projects.Third, managers should also be aware that operating in the social sector requires the possession and development of new knowledge and capabilities and may also require new organisational-and managerial practices to systematize this knowledge and institutionalize it into the firm for further use.Fourth, SI projects appear to offer managers shortcuts in forging relationships with external local actors.Thus, managers should consider that the relationships initiated and deepened through the SI projects may become a strategic resource for the subsidiary which can result in not only a strategic position of the subsidiary within its local surroundings but also the corporate MNC network.Finally, as the barriers to institutionalizing SI are complex for MNCs, there is a need to employ 'change-makers' to facilitate this process.Identifying these persons may involve costs to the MNC, particularly as the vetting process of internal or external persons to drive these changes is not well understood.This leads us to our suggestion to consider and perhaps even re-think current recruitment strategies and policies, to one that is more suitable to find potential change-makers that can create actions to solve social problems and to deliver value for the corporation.

Directions for future research
Based on the analysis, our directions for future research directions are divided into three areas: theory, research context, and method.Table 7 provides a summary of these suggested research directions for each research area, along with some sample research questions.

Theory
The fragmented use of theories opens up opportunities for research to address the topic of MNC involvement in SI from a number of suggested theoretical perspectives.For instance, the resource-based view explains a firm's competitive advantage as a mix of resources that are valuable, rare, and costly to imitate (Barney, 1991).Early versions of RBV included physical, organisational, and financial capital, whereas later research recognizes the potential to gain resources through networks and firm alliances (Gulati et al., 2000).Since partner resources are emphasized as crucial in SI processes, an interesting question for future research is what kind of network resources or partner resources are necessary for an SI context.As an extension of the RBV to dynamic markets, the dynamic capability theory refers to the firm's "ability to integrate, build and reconfigure internal and external competencies to address rapidly changing environments" (Teece et al., 1997, p. 516).This theory could help explore not only the conditions under which specific social innovation capabilities can be developed but also how these capabilities may be used to overcome problems related to SI activities.In recent research on sustainable initiatives in firms, it has been argued that in particular collaborative capabilities are crucial (Van Hoof & Thiell, 2014) for firms to obtain resources from partners and to learn from them.Hence, a dynamic capability view allows questions such as the following: What type of organisational capabilities are most conducive for MNCs involvement in SI? Are MNCs equipped with necessary collaborative skills and capabilities that allow them to learn from non-business partners?Are there any particular re-skilling capabilities that are necessary for the SI context?
The organisational learning theory offers insights into how organisations process information to expand their range of potential behaviour and thereby allow them to see new things and do new things (Huber, 1991).In line with organisational learning theorists (cf.Cohen & Levinthal, 1990), one way to leverage SI would be through "learning-byinvolving" to build an organisational knowledge base for future application.Interesting questions for future research would be: What means of learning and what learning approaches are MNCs using in the SI context?Does an MNC need to unlearn its traditional mentality to learn  when it comes to SI? How can MNCs speed up their learning ability for societal problem solving?Our findings imply that the involvement in SI implies a merge of different organisational identities and value logics.Hence, the use of organisational identity theory (cf.Albert and Whetten 1985), which deals with the set of meanings or beliefs that answer the question "who are we?" could provide an interesting foundation for future research.For instance, how do the organisational identities per se of the MNCs affect the commitment to SI activities?What is the link between their organisational identity orientation, in terms of individualistic, relational, or collectivistic orientation (cf.Wickert et al., 2017) and the involvement in SI?
With regard to the findings on the issue of power balance between the actors involved in SI activities, one useful theoretical perspective is that on power.According to Avelino (2021, p. 425), this "enthusiasmfor-social-change" requires increased attention to power and politics.The power debate has revolved around the question of whether power is consensual or conflictual (Haugaard, 2002) in a context to a situation where the actors are in, for instance, a project, seek consensus or conflict.This poses the question of whether a social innovation initiative and its outcome are determined by consensual or conflictual processes between the actors involved?Or in other words, is there a struggle between the 'bright' and the 'dark' side of power in SI projects?A closely related behavioural approach is that of micro-political theory (cf.Dörrenbecher and Geppert, 2011), which deals with how to handle organisational bargaining and conflict solving.Not only could this theory inform us of how intra-organisational micro-political negotiating processes within the MNC may affect the choice and nature of the SI initiatives selected.It could also improve our understanding of how MNCs cope with inter-organisational conflicting goals and value frames and under what conditions a specific institutional logic prevails.Research on the use of influence tactics and tools (e.g., Kipnis et al., 1980) could also be of value for this research.
In this review, it was acknowledged that becoming a "first social mover" is important.Hence, one interesting research question relates to the extent to which the opportunity for an MNC to be a "first social mover" encourages engagement in SI.From a 'first mover advantage' approach a company that leads the way into a new market locks in a competitive advantage that ensures superior profits over the long term (cf.Lieberman & Montgomery, 1988).The question is thus whether or not this theory holds within these new and uncontested social innovation markets.We believe that this issue calls for more research by IB scholars since they have the knowledge and insights to grasp the inner workings of MNC behaviour in SI.In one of the review papers (Dionisio & de Vargas, 2020) the phenomenon of hybridity and the emergence of hybrid organisations is emphasised in relation to SI activities in terms of the very nature of such activities of straddling organisational boundaries, resulting in conflicting organisational logics, values, and rationalities.Although research on hybridity is in an early stage (Doherty et al., 2014), it would be interesting to understand how MNCs are able to handle this hybridity and what implications an engagement in hybrid organisational contexts might have for securing long term competitive advantage.

Research context
In this review, we have seen that the vast majority of MNC engagements in SI were based on strategic considerations and therefore characterized as voluntary and "self-regulatory" activities (Haufler, 2001) not guided by formal regulation.However, since the societal sphere is becoming increasingly formalized (cf.EC, 2021), societal problem solving may move from gaining informal legitimacy to an obligation to comply with increasing formal legitimacy.Considering the formal institutional setting surrounding MNCs, future research should include questions such as how new and evolving formal rules and regulations will affect MNE involvement in SI.Over the years, the research context (e.g., economic, social, technological) has also changed.One significant event has been the publication of the UN's Sustainable Development Goals (United Nations, 2015), pinpointing MNCs a particularly vital role because of their worldwide reach and vast resources (e.g.Holmström Lind et al., 2018, p. 79).Interesting questions to investigate relating to the role of MNCs in the fulfilment of the SDGs could be: What types of SI projects have MNCs engaged in that specifically aim at fulfilling the UN's Agenda 2030?Which UN Social Development Goals have been the most or the least supported by MNCs, and what are their issues of concerns for entering these areas?
From an SI activity context, one limitation of this study is that we were not able to determine the type of SI the MNC was engaged in, whether it implied a small or a radical/transformative change.A suggestion for future research, therefore, is to investigate the type of SI (e.g., incremental/radical or catalytic/sustaining) that MNCs are involved in.Future research would also gain from studies on the role of MNCs for the 'scaling up' of SI activities.SI activities often occur in a local community context (Cajaiba-Santana, 2014).However, in this study, we were also not able to discover the locational context in most of the articles.Hence, an issue for future research would be to evaluate the influence of the geographical-and locational context on the outcome of different types of SI activities and projects.
Since the involvement of specific local MNC subsidiaries may be more complex than the review could discern, a suggestion for future research would be to adopt a subsidiary perspective by analysing, for example, the challenges of these geographically dispersed subsidiaries when engaging in community-based SI and assessing to what extent these subsidiaries are given autonomy to act on societal problems in their local surroundings by the corporate headquarters.In addition, in the IB literature the importance of embeddedness in business-tobusiness relations for knowledge development in subsidiaries is well documented (cf.Andersson et al., 2002).Cross-sector relations have generally not been included in this research.One interesting issue for further research would thus be how this cross-sector embeddedness of SI activities affects the 'stickiness' (Szulanski, 2000) of SI related knowledge and the 'absorptive capacity' (Cohen & Levinthal, 1990) of the subsidiaries.In this review, we were not able to detect any studies where the knowledge developed in these SI processes has actually been applied.

Method
One of the findings from this literature review is that the majority of the papers contained qualitative case studies and conceptual research, with many of the empirical papers based on a few cherry-picked successful cases.We believe that this field of research needs more in-depth case studies, in particular, process studies that follow an SI project over a longer period of time.Future research should bring attention to and use longitudinal studies that collect data at two or more different points of time to capture mechanisms of change and stability.As the literature review shows, only two papers (3%) used quantitative methods, which limits the application of this research for managerial or policy recommendations.The lack of quantitative studies, in particular survey studies, suggests that there have been very few attempts to operationalise concepts and cause-and-effect relationships or develop impact measurements and performance measurements related to the participation of MNCs in SI activities.We also suggest that the conceptual framework presented in Fig. 2 could be a promising starting point for developing constructs for testing the interdependencies between, for instance, the motivations for SI involvement and the role adopted in SI activities.We also believe that it would be interesting to explore if there is an actual causal relationship between an MNC's value orientation and its role in SI, and to what extent do the identified barriers in this review affect the level of engagement and role adopted among MNCs in these activities.There is also room for more in-depth case studies in this field of research due to the rather anecdotal use of MNCs and lack of in-depth empirical analysis.For instance, one interesting research avenue would be to look more into how MNC can overcome the various barriers presented in Fig. 2.

Conclusion
This review adds new insights into MNC involvement in SI.It reveals that this research field is still in an embryonic stage and consists of a number of different conceptual approaches and understandings of the role of MNCs in the social innovation space.The review provides principal features of this involvement and raises issues that warrant further research.It contributes not only to the current state of research on SI but also to the recent line of research on CSR and CSI, among other areas.One important implication of this study for MNC managers is to invest in social innovation education and try to learn about SI.Finally, we see a distinct opportunity for IB scholars to engage in and contribute to this topic by applying the rich sets of knowledge on MNCs available to them.

Table 1
Distribution of documents by publication.

Table 2
Overview of the conceptual foundation.

Table 3
Value orientation.
"Traditionally, social innovations have been produced via the public sector and NGOs or by social entrepreneurs.Today, as companies apply "shared value" principles to global supply chains and business lines, green their operations, and strive for more robust social impact"(Mirvis &  Googins, 2018b)C.Holmström Lind et al.
(continued on next page) C. Holmström Lind et al.

Table 7
Future research agenda.