The Social Dimension of Corporate Sustainability: Review of an Evolving Research Field

Abstract

The social dimension of corporate sustainability has received increasing scholarly attention, especially during the last two decades. As a result, the research field is growing and evolving in different directions, yet currently missing a structured overview. The purpose of the present paper is to close this shortcoming by means of a scoping review of the relevant literature. A total of 53 papers have been identified for review through systematic database search. The research contributes to the academic discussion by providing the following key results: (1) an overview on the conceptual approach, definitions and industry/region-wise coverage of corporate social sustainability research; (2) a review and discussion of the key elements of the concept and (3) the identification and discussion of gaps and avenues to systematically develop the research field further. The results indicate a need for joint empirical and theoretical development, especially with stronger theoretical/conceptual discussions also from empirical research.

1. Introduction

Social sustainability (SS) is an integral part of sustainable development (SD), addressing social challenges such as poverty, inequality or illiteracy [1,2]. Within the popular three pillar model of SD, it stands next to the economic and ecological dimensions, usually considered equal in importance on a conceptual level [3]. Yet, this has not always been the case, as much of the early sustainability debate focused on the ecological challenges of economic development [4]. Vallance et al. [5] anchor the integration of the social dimension into the SD construct within the 1987 ‘Brundland Report’ by the UN [6]. Describing it as a development, that “meets the needs of the present without compromising the ability of future generations to meet their own needs” [6], it places a highly anthropocentric, thus social notion at the core of SD.

Whilst SD represents a concept at the societal level, it has also received increasing interest in corporate management practice and research over the last decades [7,8]. Hahn et al. [7] credit this to the key role of firms in the pursuit of sustainability, as they embody the main organizational principle of the economy and represent its productive resources [9]. For transposing the notion of SD to the business sphere, the concept of corporate sustainability (CS) is a widely established framework [10]. Although there is no consensus on a single definition of CS, researchers identify a pragmatic, yet implicit consensus in the literature that refers to a multi-dimensional construct integrating environmental, social, and economic outcomes of corporate activity [11,12]. Hence, the general understanding of CS transfers the three-dimensional model of SD to the corporate level. CS in this interpretation also contains the normative position that corporations shall contribute towards the realization of SD on the societal level [9]. The triangular interpretation of SD was most prominently introduced into the corporate context in the late 1990s with Elkington’s [13] ‘Triple Bottom Line’ (TBL) approach. The TBL advocates that the corporate contribution to SD should not only be measured in economic indicators, but analogously in environmental and social terms. This has spread into the further conceptual development of CS until today, so that nowadays the majority of its definitions explicitly references these three dimensions, hence also a dedicated social perspective [12].

Yet, SS is often described as underdeveloped in comparison to the other dimensions, lacking conceptual clarity and a solid theoretical foundation [14,15,16,17,18]. This holds true for the SD discourse overall as well as for CS in particular [19,20] and reaches back already to the initial reception of the TBL [21,22]. In this context, corporate social sustainability (CSS) has been described as the most difficult dimension to conceptualize [23], entailing major challenges in further operationalization and measurement. Both the definition and selection as well as quantification and measurement of social indicators pose a major challenge due the complex and dynamic social context of corporate activity [19,24]. With regard to assessing CS, sustainability tools have usually been described as focusing on and being more sophisticated for the environmental compared to the social dimension [19,20]. Similar to the development of SD research in general [3], also the social dimension of CS has started to gain more scholarly interest starting from the early 2000s onwards [25,26], with relevant journals increasingly picking up the topic [27,28,29]. A variety of proposals to conceptualize and assess CSS are being presented, often with differing country and industry focus [25,30,31] as well as from a generic perspective [32]. On a theoretical level, Galuppo et al. [33] developed a well-recognized, two-fold conceptualization of CSS featuring a substantive and a procedural perspective. The former addresses the question of which goals are defined and pursued in the context of SS, whilst the latter refers to how these are being achieved. Especially the procedural aspect is thereby specific to SS in comparison to the other dimensions, as it inherently refers to how organizations and stakeholder cooperation are managed as social constructs [10,34,35].

Whilst benefitting from increasing interest and noticeable research activity, a structured overview of these topics and discussions is currently missing. The present article aims at filling this gap by means of a scoping review. Reviews are a common method in CS research [36,37,38] and scoping reviews in particular can for instance target key definitions and concepts, analyze research topics and identify knowledge gaps as opportunities for further research [39]. There is a considerable number of literature reviews targeting social topics in supply chain management (SCM) specifically, as presented in

Table 1. Overview on existing reviews on SS in SCM.

Reference	Reviewed Topics
Arslan [40]	Themes of social SCM research
Chiesa & Przychodzen [41]	Focus topics and development of socially sustainable SCM
D’Eusanio et al. [42]	Methods and tools for SS in SCM
Köksal et al. [43]	Social SCM in the textile and apparel industry
Nakamba et al. [44]	Definitions of SS in SCM, theoretical frameworks, emerging themes
Sodhi & Tang [45]	Themes of social SCM research
Yawar & Seuring [27]	Management of social issues in supply chains
Zorzini et al. [46]	Management of and theory usage for socially sustainable sourcing

.

Whilst social topics in SCM are being reviewed extensively, reviewing CSS research from an overall corporate perspective appears to be a blank spot so far. The aim of this endeavor is therefore to develop a structured overview of how CSS is approached, conceptualized and operationalized by researchers. Whilst this requires a generic organizational-level perspective, it does not imply an a priori exclusion of articles from the SCM context. In the light of complex global value chains and growing accountability of companies’ for their upstream operations, SCM is an integral part of corporate management and can therefore also contribute to CSS research [47,48]. Since CS management is a highly practical topic, the authors chose to review empirical research specifically, yet of course not neglecting the potential theoretical contributions of it. Explicitly, the following research questions guided the review:

• RQ1: How is CSS approached and defined in empirical academic research?
• RQ2: How is CSS conceptualized and operationalized?
• RQ3: Which gaps and avenues can be derived to further develop CSS as a research field?

The remainder of the article is structured as follows: Section 2 describes the research method to include the systematic literature selection and analytical framework for the review. In Section 3, we elaborate on the descriptive and qualitative result. The latter includes the conceptual approach, a comprehensive mapping of the research activity along regional and industry focus as well as the presentation of results on the substantive and procedural dimensions of CSS. Subsequently, we discuss the findings, propose a research agenda and point out limitations of the study. Lastly, the paper is closed with concluding remarks.

2. Materials and Methods

To assure scientific rigor, reviews have to follow a systematic process [49]. The approach chosen by the authors draws mainly on the propositions formulated by Tranfield et al. [50] and Linnenluecke et al. [51], which are already proven in CS-related reviews [41,44,46]. Therefrom, the authors deployed a four-step methodology to structure the research: (1) planning, (2) development of literature database, (3) analysis and synthesis, and (4) reporting. With the planning and justification already elaborated upon in the introduction, the following section presents the literature selection.

2.1. Data Collection

The search for and selection of literature followed a structured protocol based on the PRISMA guidelines [49] and their extension for scoping reviews specifically [52]. Following a preliminary scoping [51], and in line with the current scholarly praxis, the authors used variety of terms to address both CS [37,38,53] and its social dimension [54,55] with the search string, as illustrated in Figure 1. The variety of keywords ensured a more inclusive search, thereby adding to the validity of the review. SCOPUS and EBSCO Business Source Premier were chosen as target databases as they are considered to provide an extensive range of academic, peer-reviewed literature and offer the technical capabilities for a rigorous search process [51]. Secondly, they are commonly used in reviews on CS, indicating general suitability for the research topic [38,56,57,58,59]. Whilst the usage of more than one database yields a certain number of duplicates, it allows for a broader and more comprehensive literate body, which also contributes to validity.

The application of the search algorithm including preliminary filters in March 2022 yielded 5141 results from SCOPUS and 3154 from EBSCO. The successive identification of relevant articles was strictly based on the PRISMA protocol [49]. After eliminating duplicates and remaining unwanted items such as editorials, 7038 records remained. Based on the initial scoping [51] and related reviews, as mentioned in the introduction, specific inclusion and exclusion criteria were defined to subsequently identify the relevant articles. These are presented in

Table 2. Criteria for literature selection.

Inclusion Criteria	Exclusion Criteria
Empirical research articles	Editorials, book reviews, non-empirical articles
Articles with sole or strong focus on CSS	Articles not addressing CSS at all
Articles addressing CSS as part of CS with dedicated recognition and further elaboration	Articles addressing CSS only superficially, e.g., as part of CS without further consideration or with strong focus on environmental sustainability
Articles providing deeper analysis and discussion of CSS	Lack of in-depth analysis of CSS, e.g., articles taking a CSR ranking as a sole proxy
Articles addressing SS on the corporate management level	Articles without direct link to the corporate management level, e.g., focusing on life cycle assessment on product level only

.

During the screening phase, the authors performed an overlapping and partly iterative selection process. Firstly, the title and keywords of all records were screened. Those that could clearly be identified as not addressing the research questions were excluded, those clearly of interest were moved to the next step of screening the abstracts. In the case of doubt or disagreement between the authors, the abstract was screened right away. Most records were excluded during this first stage. For the next step from abstracts to full text analysis, the process was repeated analogously. A total of 137 papers remained for retrieval and full text assessment, of which 48 were considered suitable for the purpose of the review. For those, the authors screened the references for potential further relevant articles. Another 5 papers were included in this way based on full text assessment considering the same criteria. The full body of literature is available as a Supplementary Table S1.

2.2. Analysis and Synthesis

To ensure a structured and systematic analysis of the literature, the authors developed an analytical framework, as presented in Figure 2, in line with the recommendations of Tranfield et al. [50] and Linnenluecke et al. [51]. The basic structure of the framework was designed a priori to address the research questions based on the preliminary scoping. Further refinements and specification of the sub-categories as described in the following result sections were made iteratively during the actual analysis of the literature. The Supplementary Table S1 also categorizes the reviewed articles along the analytical categories.

3. Results

3.1. Descriptive Statistics

The final literature body for review comprised 53 papers, which in a first step were analyzed with regard to their formal characteristics. As a result, this section presents a brief overview on the distribution across publishing journals, over time and applied research designs.

The allocation of publications across different journals is presented in

Table 3. Journals ranked by publishing activity in CSS.

Journal	Published Articles
Journal of Cleaner Production	10
Sustainability	10
International Journal of Production Economics	2
International Journal of Sustainable Engineering	2
Journal of Business Ethics	2
Social Responsibility Journal	2
Sustainability Accounting, Management and Policy Journal	2
Journals with one publication each	23
Sum	53

. In sum, 30 journals have published articles relevant to this review, of which 23 contribute one paper each. A number of journals published two articles on the topic, including focused journals such as the Journal of Business Ethics and the Social Responsibility Journal. With 10 publications each, the Journal of Cleaner Production and Sustainability are heading the publication list. The full list of journals is shown within the Supplementary Table S1. Overall, this indicates that CSS, as a subtopic of CS, is perceived and developed as a specialized topic rather than appearing on the mainstream corporate agenda.

Regarding the allocation of publications over time as presented in Figure 3, three different phases can be distinguished. During the first phase, lasting from 2000 to 2013, research activity is picking up slowly, amounting to seven articles during the first 13 years of the researched period. The second phase from 2014 to 2018 comprises 18 papers and the last phase, from 2019 onwards covers 28 papers. From phase 1 to 3, this perspective shows a four-fold increase in the literature, notwithstanding the differing number of years as well as the fact that for 2022, only the period until March is covered. This rapid increase in publications firstly indicates the rising interest of scholarly attention for the topic of CSS. Secondly, it reconfirms the observation of an expanding and dynamic research field as stated during the introduction, which also served as one the main justifications for the present review.

Figure 4 shows the research methods applied among the reviewed articles using the classification of already existing reviews from the wider CS and sustainable SCM context [27,60]. As this paper targets empirical articles, categories such as ‘Review’ or ‘Conceptual’ do not apply. With over 30%, case studies [61] constitute the largest part of the literature body. Whilst some of the cases also contain qualitative data, most stay on a qualitative level. Eleven papers follow a modeling approach, most apparent in this category are interpretive structural modelling and fuzzy logic. Qualitative papers amount to 10 in sum, with interviews and document analysis being the major means of research. Eight studies implement quantitative approaches, for which data is collected via surveys. For statistics, analysis of variance and structural equation modeling are the most utilized methods. Seven articles develop mixed methods approaches, i.e., combining either of the previous methodological categories. Here, the combination of Delphi studies with other approaches such as case studies or quantitative analysis was most apparent. Overall, this picture indicates that CSS research is currently in a developing and expanding stage with qualitative research dominating the field. In a broad sense, this includes the greatest part of case studies, naturally the qualitative papers as well as the qualitative parts of modeling and mixed methods papers. Yet, the significant share of modeling papers might point towards the transition to stronger quantification and theory testing, whilst strictly quantitative approaches still play a minor role.

3.2. Qualitative Results

The following chapters document the qualitative results from the literature review. Section 3.2.1 and Section 3.2.2 thereby mainly present the findings related to research question 1, whereas Section 3.2.3 and Section 3.2.4 relate research to research question 2.

3.2.1. Conceptual Approach towards CSS: Theoretical Frameworks and Definitions

From the literature body, 20 papers explicitly elaborate a theoretical approach to CSS. Stakeholder theory appears to be most common, e.g., investigating the demands of different stakeholders and their ability to realize them [62,63,64,65]. Shareholders and their interests are recognized as part of the stakeholder community [62,66], yet shareholder value as an explicit theoretical approach is not apparent. The resource-based view focuses further on the resources companies’ need to operate and how these can be obtained or built [67,68]. Tate & Bals [66] most prominently use it to develop a social resource-based view, equivalent to the already established natural resource-based view. Institutional theory is employed to examine the adoption of specific CSS practices and phenomena of organizational isomorphism as a result of the social environment of companies [67,69]. This is closely connected to legitimacy theory, which emphasizes the embeddedness of corporations into social systems, creating a need to obtain social approval to do business. The need for legitimacy is referenced to both individual stakeholders as well as society in general [67,68]. Labuschagne et al. [25] utilize capital stock theory, framing CSS as a company’s contribution to various forms of social capital, e.g., human, productive and community capital. Turker & Ozdemir [70] draw on systems theory to address the complex, reciprocal and non-linear relationships of corporations with their social environment. Further theoretical frameworks mentioned are transaction cost economics [71], agency [68] as well as contingency theory [72]. Four papers provide a multi-theory approach by combining stakeholder theory with one or more other frameworks. Amongst those are capital stock and legitimacy theory as well as the resource-based view and institutional theory [67,68,69,72].

A majority of papers (45) refers to a general SD definition or construct, mainly to the well-known Brundtland notion as well as to the TBL, whereby the latter already addresses the corporate context [73,74,75]. However, only 16 articles attempt to further define CSS apart from the notion that it embodies the social or human side of CS [76]. Three exemplary definitions were selected for further discussion based on: (a) clear reference to theory and/or SD, (b) mentioning of both substantive and procedural aspects to CSS. These are presented in

Table 4. Exemplary definitions of CSS.

Reference	Definition of CSS
Mani et al. [77]	“socially sustainable practices can be defined as the product and process aspects that determine human safety, welfare and wellness. Social sustainability issues relate to stakeholders including suppliers, manufacturers, customers, and society.”
Sundström & Mickelsson [68]	“From a substantive “what it is” point of view, social sustainability contributes to internal as well as external stakeholders’ development and growth by achieving goals like wellbeing, equality, inclusion and social cohesion, as well as opportunities for learning and self-development. Taking the procedural (how) aspects into consideration, social sustainability addresses how organizations’ relate to their stakeholders as far as transparency, communication, decision making and management are concerned.”
Turker & Ozdemir [70]	“Social sustainability can be defined to what extent an organization addresses the equity principles [inter-generational, intra-generational, geographical, prodecural, inter-species] by developing innovative approaches to meet the demands of social stakeholders by considering the dynamic interactions across components [Social innovation, societal demand, social stakeholders].”

below.

Sundström & Mickelsson [68] most prominently highlight the distinction between a substantive and a procedural perspective towards CSS. The former refers to the direct impacts a corporation has on its social stakeholders ranging from direct stakeholders such as customers to society as a whole [77]. Hence, it describes ‘what’ can be understood as CSS and what is considered to be within its boundaries compared to the other dimensions of CS. Sundström & Mickelsson [68] further emphasize stakeholders’ goals, to which a company can contribute through its goods and services as well as the operations to produce them. Turker & Ozdemir [70] put this into the broader context of SD, which is framed as a desired societal target overall. To this, a gap arises from the actual state, resulting in corresponding societal demand. CSS is subsequently defined as the corporate contribution towards satisfying this demand through social innovation. For transposing societal demand to the individual organizational level, corporations continuously have to engage with their stakeholders. Hence, the procedural perspective relates to the way a firm is managed and how organizational goals are being defined and realized in this regard. Turker & Ozdemir [70] factor in a principle of procedural equity, derived from the Brundtland SD definition. Sundström & Mickelsson [68] again highlight transparent communication as well as the degree and organization of stakeholder participation in management and decision-making in this context. Whilst the identification of the two dimensions of CSS is very apparent, there are not always clear boundaries between them. Rather, they might overlap especially in praxis and go hand in hand with each other [68,70].

3.2.2. Contingencies: Context Factors Influencing CSS Research

Different theoretical approaches argue that organizations have to adapt to their environment, such as institutional, legitimacy or contingency theory. In the light of CCS, especially the social structures in which companies operate are of interest [69], for which country and industry affiliation have been identified as two major context factors [45]. To highlight the covered as well as uncovered areas,

Table 5. Classification of reviewed papers for industry and regional focus.

Region/ Industry	Africa	Asia	Australia	Europe	North America	South America	No Regional Focus
Technology				[31]
Telecommunications
Health Care		[63]					[79]
Financials
Real Estate
Consumer Discretionary		[64,65,71,74,77,80,81,82]		[83,84]		[85,86]	[48,70]
Consumer Staples		[75,76,77,87,88]		[62,83,89,90,91]	[91,92]	[93,94]	[67]
Industrials	[25]	[74,77,80,95,96,97,98,99]		[31,83,84,100]
Basic Materials		[77]		[30,101]
Energy
Utilities				[84,102]
No industry focus		[66,69,103]	[104]	[47,68,69,105,106]	[47,107]	[72]	[108,109,110,111]

presents the reviewed literature along these variables. For industry classification, the common Industry Classification Benchmark (ICB) standard was used [78]. As there are multi-country and multi-industry studies within the sample, references may appear more than once. The two categories ‘without regional focus’ and ‘without country focus’ were used to classify papers without any relation to a specific region and/or industry.

Region-wise, an emphasis of studies focusing on Asia as well as Europe can be observed, whilst industry-related ‘Consumer Staples’ heads the publication list followed by ‘Consumer Discretionary’ and ‘Industrials’. Within ‘Consumer Staples’, social sustainability in the agriculture & food industry appears to be a topic of interest in both Europe and Asia. Regarding the former, Gaviglio et al. [89] formulate an approach to quantify the social sustainability impact of farming organizations, Vu et al. [90] assess fast food service operations in the UK and Veldhuizen et al. [62] analyze the importance of social issues in the Icelandic fishing industry. Asia shows an emphasis on palm oil production [87,88] as well as food production overall [75] with a country focus on Malaysia. Further studies target for instance the beef industry in the U.S. [92], farming in California and Spain [91] and food production [67] without regional focus.

‘Consumer Discretionary’ has the strongest research interest in Asia with three papers focusing on India [74,77,80], which mainly address car manufacturing. Three studies target apparel production, two of which in Bangladesh [71,81] and one in Sri Lanka [82]. Apart from that, further studies target Europe and South America.

A certain amount of papers was classified without regional focus and/or without industry focus. An emphasis here can be found in Europe without industrial focus with five published papers. To name two examples: Rumanko et al. [106] research the differences in SS practices among a family versus non-family business in Slovakia without any industry consideration and Mani et al. [105] analyze the adoption of SS practices in Portugal. Four further studies classify without regional and industrial focus. As one case of that, Popovic et al. [109] develop and test quantitative indicators for the SS assessment in supply chains. Lastly, there are a few areas covered only very little or not at all. Region-wise, both Africa and Australia only note one paper each, whilst sector-wise, ‘Telecommunications’, ‘Financials’, ‘Real Estate’ and Energy have not been addressed yet.

For factors apart from country and industry affiliation, Huq et al. [81] provide an illustrative example as their study also relates to another contingency in terms of a special, in this case catastrophic, event. The study was performed pre and post the Rana Plaza incident in 2013, where over 1000 people died in a collapsing building that housed apparel suppliers for international brands. For this period, the authors explored the evolution of social management practices observed for instance a strong rise in social standards, buyer-supplier collaboration and joint audits. Furthermore, Ikram et al. [110] consider the COVID-19 situation as another contingency factor and identify pandemic response measures as an important part of CSS.

3.2.3. Substantive Dimension of CSS: Concepts and Measurement

The substantive dimension of CSS relates to the impact of corporate operations on social conditions and which societal goals are considered to be of significance in this context [68]. In short terms, this addresses the question of what is required from a content perspective to achieve CSS. Due to the magnitude of social stakeholders on the one and the variety different industries and business models on the other hand, this is both highly complex as well as critical to CSS and its research [89]. A clear focus of research activity addressing this challenge can be noticed with 43 papers contributing towards it. Based on the results from present literature, the analysis of this part follows a three-step structure: (1) breakdown of the CSS into a framework of thematic categories, (2) definition of key aspects per theme and (3) operationalization via indicators and metrics, which are partly also applied empirically by the actual measurement of CSS in e.g., a case study. For steps 1 and 2,

Table 6. Classification of CSS’ substantive dimension.

Theme Cluster	Frequency	Example Aspects/Indicators
Health & Safety	36	Provision of safe & health-supporting working environment
		Occurrence and handling of health and safety incidents
		Emergency trainings and protocols, pandemic response plan
		Healthcare services for staff and/or overtaking of healthcare costs
		Dedicated measures for jobs/tasks with high-risk exposure
		Provision of safe & health-supporting working environment
		Occurrence and handling of health and safety incidents
Labor, employment & working conditions	34	Salary levels and fairness, incl. paid overtime, sick leaves, etc.
		Stability of employment and job security
		Adequate work environment, e.g., workplace, sanitary facilities, water supply
		Employee well-being, e.g., stress prevention/reduction, job satisfaction
		Prevention of child, forced and compulsory labor
		Salary levels and fairness, incl. paid overtime, sick leaves, etc.
Local community involvement	30	Contribution to local economic development, e.g., through provision of employment and local sourcing
		Integration of local community into company decisions
		Contribution (e.g., grants or donations) to local projects, e.g., local housing, education & health infrastructure
Training, education & competency development	27	Support of employee professional and personal development, e.g., through dedicated training curriculum
		Regular performance and career reviews
		Support employees in higher/further education
Diversity, inclusion & equity	23	Promotion of diversity and equity (gender, race, religion, etc.) across all levels of the organization and beyond company
		Inclusion of marginalized groups, e.g., disabled workers
		Prevention and handling of discrimination cases
Product & customer responsibility	20	Product safety, e.g., no use of hazardous materials
		Product-specific characteristics, e.g., nutritional value (food) or durability (industrial goods)
		Non-discriminative distribution of products and services, e.g., in pricing policy and distribution channels
Society involvement	18	Contribution to economic welfare and goods supply
		Contribution to charity, e.g., through grants or donations
		Support of employees to contribute to public/voluntary work, e.g., through additional leave days
Democracy & stakeholder participation	15	Integration of the public/dedicated stakeholder groups into decision-making
		Support of freedom of association and collective bargaining
		Prevention of corruption and illegal/unfair lobbying practices
Knowledge sharing & innovation	13	Contributing to research & education, e.g., through own research facilities or university/school partnerships
		Contributing to (social) innovation and its dissemination
Supplier & industry involvement	13	CSS implementation along the value chain
		Development of CSS standards & guidelines
		Fair business operations and competition
Transparency & information disclosure	10	Quantity & quality of shared information, e.g., through dedicated sustainability reporting

shows an overview of the content results provided by the literature sorted by their frequency of mentioning.

The first step in examining the content of CSS is usually the breakdown into thematic categories or category clusters such as equity [95,98], employment practices [85,102] or training and competency development [97,100]. Whilst most authors use a direct category breakdown, others cluster the categories further, for instance for internal and external topics [82,83,108] or depending on the proximity of affected stakeholders [72]. Most authors derive their categorization from literature search, mostly referring to key papers from social and political sciences backgrounds, some also use a theory-based methodology. Stakeholder theory again is a common approach for this [72,83,108], sometimes combined with other frameworks. Hutchins et al. [107] for instance develop a stakeholder needs-based approach by factoring in the Maslow hierarchy of needs. Next to treating all categories equally, some authors also develop a ranking of their individual categories [75,88]. For those, ‘Labor, employment & working conditions’, ‘Health & safety’ as well as ‘Local community involvement’ rank as the most important topics.

Twenty-two articles provide a further operationalization of their content categories insofar as they define indicators to substantiate and measure them. Due to the highly qualitative nature of SS, these indicators are most of the time of a qualitative nature [65,79,93]. Yet, researchers also engage in quantification and propose a balance of qualitative and quantitative indicators [73,87,101], whilst some also present purely quantitative approaches [97,109]. Lastly, 15 articles engage in the implementation of their framework via an actual assessment of CSS. Some of these stay on a theoretical level with a hypothetical example [79,108], but the majority of 11 papers provides an empirical case of one or multiple companies [75,89,90]. Surveys with numerical assessment are a common means here [79,88,106] as well as mathematical and fuzzy logic models for quantification of linguistic variables [80,108].

A high relevance of contingency factors regarding the substantive part of CSS was identified, as the interpretation and absolute as well as relative importance of social issues is highly dependent on the territorial context [89]. Huq et al. [71] for instance notice a mismatch between western concepts and definitions and their implementation in developing countries for global supply chains. The fact that both country and industry affiliation have a significant influence is illustrated by the following example from the industry cluster ‘Consumer Staples’. Lim & Biswas [88], addressing palm oil plantations in Malaysia, rank ‘Human needs’ as the most important category and ‘Workers’ accessibility to health care’ and ‘Provision of sanitation facilities to workers’ as the two most contributing indicators. Ahmad & Wong [75], also covering Malaysia but with a focus on food manufacturing, present ‘Efforts to reduce health, safety and environmental hazards’ for the local community and ‘Decent working conditions’ as the top two indicators. Lastly, Veldhuizen et al. [62] in their study of fishery in the northeast Atlantic conclude with ‘Worker safety’, ‘Product freshness’ and the ‘Company’s salary levels’ as the most important factors of CSS. Whilst a certain level of agreement on category level is noticeable, e.g., with the emphasis on labor-related practices, the actual implementation in terms of indicators varies widely. Additionally, industry-specific indicators apply selectively, such as product freshness for fishing companies. Hence, the choice of indicators plays a major role for a valid operationalization of CSS. This in turn can only be defined in relation to the company’s social context on the one hand and its business model and operations on the other hand. Supporting this, Gaviglio et al. [89] name the selection of indicators the most critical part of CSS research and practice.

Apparent in the analysis is not only the absolute, but also the relative complexity of CSS in comparison to the other dimensions of the TBL. For those papers addressing multiple dimensions, the social usually contains significantly more indicators compared to the ecological and the economic [88,99]. Some articles also propose a ranking or weighing of the dimensions and respective indicators, in which the social is partly ranked the most [75] and partly the second most important dimension [88].

3.2.4. Procedural Dimension of CSS: Governance and Practices

The procedural perspective on CSS relates to how companies operate, most notably how they manage and integrate stakeholders into their decision-making and operations [68]. It thereby complements the substantive dimension with regards to how it is being realized, e.g., how organizational goals are defined and achieved. From the literature body, 12 papers explicitly address these procedural aspects.

The results are structured among the two thematic clusters “Communication & collaboration” and “Assessment & monitoring”, as shown in

Table 7. Classification of CSS’ procedural dimension.

Categories	Communication & Collaboration	Assessment & Monitoring
Primary purpose	Information exchange with stakeholders Joint development/implementation of structures and processes to improve CSS results for all stakeholders	Control and evaluation of CSS practices, standards or management Creation and maintenance of credibility towards stakeholders regarding CSS Management of CSS-related risks
Main targeted stakeholder groups	Employees
	Suppliers and partners
	Customers
	Owners/Shareholders
	Local Community
	Society
Example governance & practices	Principles, standards and guidelines for stakeholder cooperation, e.g., CS-oriented mission statement Code of ethics/conduct Internal social management system Stakeholder interviews and/or visits Incentive structures, both internal and/or external	Self-assessment of focal company and/or partners, e.g., suppliers Stakeholder interviews and/or visits Audit, intern and/or external Certification of relevant management standards (e.g., ISO14001, OHSAS 18001)

. The first cluster targets practices with the main purpose of information exchange and improving CSS coordination and implementation among stakeholders [81,83]. The second cluster addresses specifically the control and evaluation of existing practices, which is closely related to auditing and risk management [94]. Both practice clusters address the same stakeholder groups, yet with different approaches and a different aim as elaborated in the following.

Following Huq et al. [81], ‘Communication & collaboration’ relates to “practices that improve the firm’s coordination with its suppliers, consumers and other stakeholders to jointly improve social outcomes”. To make this possible, a close communication loop with stakeholders is considered important to receive and process incoming messages and feedback [70]. Employees are highlighted as a relevant group, as their engagement and motivation are critical for organizational productivity [65]. Explicit consideration is also given to customers, for which a dedicated feedback system should be established, also corresponding to the content of category ‘Product & customer responsibility’ as reported in the previous section [76]. Further stakeholder groups mentioned are suppliers and customers, but also the community both on a regional as well as a more general level. Sundström & Mickelsson [68] also highlight regional clusters as a level of collaboration and Walker et al. [84] also emphasize the local community. Given that SCM is a frequent subject to CSS research, supplier collaboration is a focal topic too. Key to this is a joint development of not only the buyers’ social capabilities, but also those of the suppliers, especially in developing countries. Therefore, collaboration can be established on the strategic level to develop joint social objectives all the way to the operational business to foster efficient implementation on both sides [81].

To frame operations and collaboration towards CCS, respective corporate governance structures can be implemented. Wang et al. [99] propose a social-oriented mission statement, possibly also with reference to publicly acknowledged sustainability initiatives such as the UN Global Compact or Sustainable Development Goals. Pinto [83] and Walker et al. [84] point out ethical codes and codes of conduct as well as a social management system with regards to company internal policies, to which Morais [94] adds incentive and reward structures for employees. In the light of CSS, these work in a dual way: on the one hand they can incentivize desired behavior such as corruption prevention and on the other, they are also seen as essential for employees’ motivation and engagement [65]. With regards to external CSS governance, Morais & Barbieri [72] argue for the definition of standards as well as a formalized process for the selection of suppliers and cooperation partners. Concerning communication with the wider community and society, marketing is highlighted as a critical corporate function. Therefore, ethical guidelines are also to be established, for instance concerning messaging content in advertisements.

Formalized governance structures play a significant role for the theme ‘Assessment & monitoring’. Morais & Barbieri [72] relate assessment to the control and evaluation of CSS governance and practices by companies or partners themselves, e.g., through dedicated visits at a partner site and/or interviews. Monitoring complements this by seeking external verification, for instance through certifications or third-party audits. This is widely recognized as being more transparent, thorough and credible than a sole self-assessment, thus supporting stakeholder acceptance and contributing to the credibility of a company’s CSS engagement [81]. Thereby, monitoring might partly compensate for mistrust among business partners, which is harmful for the effectiveness of cooperation and collaboration initiatives [86]. From the contingency perspective, an emphasis on supply chain topics with regards to developing countries can be recognized. Walker et al. [84] name the importance of supplier evaluation on a continuous basis, including an emphasis on social aspects in this regard. With increasing legal obligations of global companies for their upstream value chain, especially third party audits and certifications also serve a risk management and mitigation purpose, as they can limit the extent of legal liability [81,94].

However, the sole concentration on assessment and monitoring practices entails the risk of ‘window dressing’, i.e., CSS activities simply being tailored towards a specific audit or to mitigate legal and financial risks. This would result in symbolic compliance rather than really addressing and improving social conditions for the affected stakeholders [81]. This complexity again highlights the importance of collaboration, especially in cross-company and cross-country cooperation of a global economy.

4. Discussion, Limitations, and Conclusions

4.1. Discussion

4.1.1. Discussion of Findings Related to Research Question 1

Firstly, a significant increase in research activity has been noticed, especially in the period from 2019 onwards. Within the review’s focus on empirical articles, case studies are the dominant research methodology. This is common among rather nascent and still expanding research fields as cases provide illustrative examples for these under researched topics. Yet, following the line of Seuring & Müller [60], some papers in this category often rather provide case examples in the sense that they present a description of their unit of analysis, without aiming at actual theory development or testing. Going further, there is a variety of papers using other methods from interview- and survey-based to sophisticated mixed and modelling approaches. Especially the latter contribute strongly to both conceptual development as well as quantification of the research field but represent only a minor part of the literature body. Hence, whilst expanding significantly in quantity, the majority of the research activity remains of a qualitative nature. This indicates a lower level of maturity, yet given the qualitative nature of CSS itself, might also be required to a certain extent. Overall, the results from the descriptive analysis indicate a dynamic, evolving research field having passed the phase of infancy but also not approaching maturity yet.

Stakeholder theory appears to be the dominant theoretical approach yet complemented by a variety of other theories. Hence, both the possibility and need for differentiated theoretical analysis of this topic can be noticed. On the aggregate level, this is currently not being addressed adequately, but especially some more recent articles already develop multi-theory frameworks [64,67,68,69] as a starting point. Few authors provide an actual definition of CSS, adding to the impression of still evolving theoretical development of the topic. Critical appraisal of the utilized CS constructs, mainly the TBL, is missing, although these themselves are not undisputed in the scholarly discussion [22,112,113,114]. Certain articles develop a differentiated and multifaceted perspective with reference to key theoretical papers both within CSS [10,33,115] and social sustainability in general [14,15,17,18,116]. A selection of definitions has been presented and discussed. Although a uniform definition cannot be identified at this point in time, the reference to two complementary dimensions of CSS, a substantive and a procedural, is very apparent and provides a starting point for further conceptual work.

The lacking theoretical development could partly be credited due to the review’s focus on empirical papers, yet corresponds to the findings from related reviews [44,60]. However, the review of Nakamba et al. [44] for instance reported less than 5% of the papers providing definitions, whereas in the present analysis, ~30% address this part. Given that SS has often been described as the neglected and under theorized dimension [28,117], this now appears to gain more consideration. The review also shows that empirical articles can provide significant contributions towards the conceptual development of the field. Future research might build on and allow more focus for this to foster empirical and theoretical-conceptual progress of CSS research in an integrated way. Putting this in more specific terms, Missimer & Mesquita [118] for instance propose approaching and defining CSS from a social systems theory perspective.

Contingency factors are highly relevant for CSS research and practice, which is easily overlooked when examining individual studies in an isolated manner. Reflecting the high number of case studies, CSS is mostly approached from a very specific country and/or industry context. Table 5 provides an overview about both the industry and territorial contexts covered so far. On the one hand, this gives an understanding about focus areas of research with regards to those factors. On the other hand, that also highlights less and uncovered areas, which offer further research opportunities. Region-wise, this would be Africa and Australia, but also North and South America, as Asia and Europe note the most extensive coverage by far. Industry-wise, there is a strong focus on ‘Consumer Discretionary’, ‘Consumer Staples’ and ‘Industrials’. In turn, this leaves research gaps in the following sectors: ‘Technology’, ‘Telecommunications’, ‘Health Care’, ‘Financials’, ‘Real Estate’, ‘Basic Materials’, ‘Energy’ and ‘Utilities’. Apart from individual sectors and/or regions, especially cross-region and cross-industry studies that factor in cultural and socio-economic aspects can add value to integrated CSS research, which is consistent with the findings of related, yet more specialized reviews [43,44].

4.1.2. Discussion of Findings Related to Research Question 2

Research question 2 targeted the issue of how CSS conceptualized and operationalized by researchers. Overall, the development of a valid operationalization of the construct in terms of matching categories and indicators has been derived as a central challenge in CSS research and practice. This is due to the combination of two factors that are already complex themselves but have to be integrated in the case of CSS. On a general notion, the social dimension has usually been considered as the most elusive and underdeveloped perspective of SD [14,15,17,18,119]. This lack of conceptual clarity is reflected on the corporate level, as CS is to a great extent derived from the wider societal discourse [70]. In this regard, Huq et al. [71] point out the local character of social conditions as one of the main factors of mismatches between western and developing country interpretations of CSS. The second factor of complexity results from the variety of different business and organizational models necessary for the production of goods and services. Thus, although a great part of the reviewed papers addresses the substantive dimension of CSS, these naturally apply only for their respective specific setting without the possibility to claim universality. Through the review, a certain agreement on categories could be derived, yet on an indicator level, only examples can be given. Future endeavors could explore that more in-depth and for instance research the possibility of a distinction between common and differentiated indicators. Therefore, existing methodologies such as materiality analysis [24] might be transferred from related fields to the CSS specific context to identify relevant indicators. Additionally, to further analyze the individual categories, related research fields might offer valuable findings. As one example, sustainability reporting, relating to theme “Transparency & information disclosure”, has seen significant research attention over the last years. Topics of interest from this research field could be the analysis of CS reports and reporting indicators [24,120] or the relationship between CS reporting and CS management practices [121].

As the second dimension to CSS, the authors identified and analyzed a procedural perspective. It is widely recognized and finds its application also in the discussed definitions, yet research activity in this context is rather scant. Few papers address the procedural dimension at all and if so, rather superficially and not as the sole topic or focus. The generated results are mainly limited to the description of example practices from case reports. In comparison to the substantial dimension, the approach to procedural topics appears less structured and systematic, as for instance different and sometimes overlapping or conflicting terminologies are applied. This opens up potential for future research to address the topic systematically and in detail, for which both the presented definitions and the theoretical discussion already provide starting points. Sundström & Mickelsson [68] for instance define CSS through a company’s contribution towards its stakeholders’ goals. Taking a stakeholder theory view on this, firms can be seen as a place of social cooperation, by means of which the different players aim to realize their own goals [122,123]. Competing interests and resulting conflicts are an inherent part of that, which Galuppo et al. [33] address by putting conflict management in the center of SS. Hence relevant literature from the wider field of stakeholder theory and management could provide a starting point to further develop the procedural aspects of CSS [35,124]. Moreover, already existing knowledge from corporate decision-making [125] and strategic management [36,126] with regards to sustainability might be useful in this context. Moreover, the role of social innovation as mentioned by Turker & Ozdemir [70] deserves further exploration with regards to its procedural aspects. In that understanding, an actual social innovation not only solves societal challenges, but also in a socially sustainable manner, e.g., regarding selection, prioritization and approach of such issues among the involved stakeholders. To develop this further, findings from innovation strategy and management especially at the intersection to CS [127,128] might provide a starting point.

Lastly, few articles engage in the practical assessment or quantification and even fewer attempt to aggregate a score for CSS. Although the possibility of the latter is discussed controversially in the literature [22,117,129], a push towards more empirical validation and practical advice especially on the implementation level is necessary. On the one hand, relevance of future CSS research for business and CS managers would benefit significantly from that, as they ultimately have to implement it in practice [118]. On the other hand, the findings from empirical testing will flow back into and help fuel the required conceptual development of CSS.

4.1.3. Derivation of Research Agenda Related to Research Question 3

From the discussion of the findings and limitations, the following consolidated research agenda can be derived to tackle current gaps and address research question 3:

• Lacking theoretical development: CSS research requires differentiated theoretical analysis and development. Whilst many of the present studies draw on stakeholder theory exclusively, promising examples for multi-theory approaches have been identified. Research can build upon and expand these to build a stronger theoretical basis. This also applies for case study research, which especially tends towards descriptive presentation rather than theory-backed analysis and discussion [60].
• Lacking conceptual clarity: Compared to previous reviews [44,60], authors increasingly engage in defining CSS. Yet, the majority of papers do not provide a definition of CSS as their subject of research. Stronger conceptual focus in turn could provide the foundation to address the challenges in further operationalization, most notably the choice of relevant and suitable indicators [19,23].
• Unresearched regions and industries: CSS is often approached from a specific region and/or industry perspective. As these factors are highly relevant for the social context of companies, especially the uncovered regions and industries deserve research attention. This could help to broaden and complement the understanding of CSS, both with regards to its substantive and procedural dimension.
• Cross-region and cross-industry studies: Related to the previous point, such studies are especially required to account for the local character of social conditions in the light of diversified and global value chains. Thereby, SS issues arising from, e.g., different cultural contexts [71] can be addressed, pushing towards a better understanding and management of such issues between involved stakeholders.
• Further systematic development of CSS’ substantive dimension: This dimension is addressed by most researchers, yet shows potential for stronger empirical, especially quantitative validation. One the one hand, categories and indicators need to be tested for their validity. On the other hand, the empirical findings need to be integrated back into the theoretical discussion to further develop and improve existing constructs. Sustainability reporting, which has seen a steady increase in importance over the last years [120], provides a pool of empirical data in this context and could be further explored with special attention to its social aspects.
• Systematic development of CSS’ procedural dimension: Research on this dimension appears to be less developed, in both quantity and quality, than for the substantive part. To address this, we propose an integration with relevant findings from fields such as stakeholder management [35,124], overall CS management [36,125,126] and social innovation [70,127]. These can provide a starting point to derive structures and processes for integration and decision-making for CSS specifically.
• Practical assessment of CSS: Many studies do not operationalize their proposed categories and/or indicators fully. This in turn is needed, not only to test the theoretical constructs but also to assist practitioners in dealing with this task in corporate management [35]. The latter is especially important considering practical implications of future CSS research, as the selection and measurement of indicators is described as a major challenge [19].

4.2. Limitations

The presented findings are subject to a number of limitations, beginning the literature search and selection process. The variety of keywords has been chosen on purpose and a large number of results was generated, yet exhaustiveness can of course not be claimed. Moreover, the timeframe of the search was limited to the year 2000 onwards. Whilst CSS has evolved rather lately as a research field, some articles from before might have been missed in that way. The same accounts for the final selection of articles, which always remains subjective to a small remaining part, even when multiple researchers are involved. The analytical framework was developed based on existing conceptual work to ensure structured and focused analysis, yet this of course also limited the content analysis to a certain extent. Alternative ways of structuring and coding the articles would of course also be possible. Lastly, the review focused narrowly on the social dimension of CS. Whilst papers covering other aspects of sustainability were not excluded per se, there was only a little consideration of the links and interrelations from the social to the other dimensions. These also could be taken into account as another avenue for future CSS research.

4.3. Conclusions

As the least researched dimension of CS, CSS constitutes a dynamic and expanding field, which has not been presented yet by a scoping review. The overarching contribution of this paper towards the research field is thus two-fold. Firstly, it provides a structured overview on how CSS has been approached, defined and conceptualized by academic research. Secondly, it points out the white spots in the current research landscape and offers avenues to address by future endeavors.

To this end, the scoping review was guided by three research questions. Addressing RQ1, different approaches to CSS and combinations of those have been identified. Detailed theoretical discussions exist but constitute a minority of the reviewed papers with stakeholder theory being the most common framework. From an empirical perspective, researchers approach the topic mostly from a specific country and/or industry perspective. Going further, CSS can be defined by a two-fold approach featuring a substantive as well as a procedural dimension, which overlap especially in practical implementation. Turning to RQ2, further conceptualization is usually performed by means of thematic clusters, which are then operationalized via indicators. These address a broad range of social challenges, ranging from basic human needs and decent living wages to higher education support and corruption prevention. Lastly, a research agenda has been presented to address RQ3 and provide scholars with suggestions for the further exploration of the field.

Future work might address those gaps so that CSS research can be developed from both a theoretical and empirical perspective. This could help reinforcing the current research dynamic towards conceptual clarification on the one hand and empirical examination on the other hand.